{
  "best_global_step": 31600,
  "best_metric": 0.14016859233379364,
  "best_model_checkpoint": "saves/ia3/llama-3-8b-instruct/train_multirc_1745950263/checkpoint-31600",
  "epoch": 6.525328330206379,
  "eval_steps": 200,
  "global_step": 40000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008157272208173587,
      "grad_norm": 5.225855350494385,
      "learning_rate": 4.999999876629946e-05,
      "loss": 1.5764,
      "num_input_tokens_seen": 7920,
      "step": 5
    },
    {
      "epoch": 0.0016314544416347174,
      "grad_norm": 10.407909393310547,
      "learning_rate": 4.999999375439123e-05,
      "loss": 0.9171,
      "num_input_tokens_seen": 16848,
      "step": 10
    },
    {
      "epoch": 0.002447181662452076,
      "grad_norm": 5.306243419647217,
      "learning_rate": 4.9999984887169785e-05,
      "loss": 0.9666,
      "num_input_tokens_seen": 26592,
      "step": 15
    },
    {
      "epoch": 0.0032629088832694347,
      "grad_norm": 5.998384475708008,
      "learning_rate": 4.9999972164636506e-05,
      "loss": 1.0607,
      "num_input_tokens_seen": 36864,
      "step": 20
    },
    {
      "epoch": 0.004078636104086793,
      "grad_norm": 6.128674507141113,
      "learning_rate": 4.999995558679334e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 45360,
      "step": 25
    },
    {
      "epoch": 0.004894363324904152,
      "grad_norm": 5.801575660705566,
      "learning_rate": 4.999993515364287e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 54992,
      "step": 30
    },
    {
      "epoch": 0.005710090545721511,
      "grad_norm": 5.770157814025879,
      "learning_rate": 4.999991086518822e-05,
      "loss": 1.0241,
      "num_input_tokens_seen": 64752,
      "step": 35
    },
    {
      "epoch": 0.0065258177665388694,
      "grad_norm": 7.130281448364258,
      "learning_rate": 4.999988272143315e-05,
      "loss": 0.9916,
      "num_input_tokens_seen": 74960,
      "step": 40
    },
    {
      "epoch": 0.007341544987356228,
      "grad_norm": 4.7758402824401855,
      "learning_rate": 4.999985072238199e-05,
      "loss": 1.1844,
      "num_input_tokens_seen": 84576,
      "step": 45
    },
    {
      "epoch": 0.008157272208173586,
      "grad_norm": 3.806382894515991,
      "learning_rate": 4.999981486803969e-05,
      "loss": 0.9189,
      "num_input_tokens_seen": 94896,
      "step": 50
    },
    {
      "epoch": 0.008972999428990946,
      "grad_norm": 1.667703628540039,
      "learning_rate": 4.999977515841176e-05,
      "loss": 0.4957,
      "num_input_tokens_seen": 104352,
      "step": 55
    },
    {
      "epoch": 0.009788726649808304,
      "grad_norm": 2.352666139602661,
      "learning_rate": 4.9999731593504344e-05,
      "loss": 0.7593,
      "num_input_tokens_seen": 113296,
      "step": 60
    },
    {
      "epoch": 0.010604453870625663,
      "grad_norm": 0.13051988184452057,
      "learning_rate": 4.999968417332415e-05,
      "loss": 0.2978,
      "num_input_tokens_seen": 122144,
      "step": 65
    },
    {
      "epoch": 0.011420181091443021,
      "grad_norm": 3.989936351776123,
      "learning_rate": 4.999963289787848e-05,
      "loss": 0.2247,
      "num_input_tokens_seen": 131168,
      "step": 70
    },
    {
      "epoch": 0.012235908312260381,
      "grad_norm": 0.7946998476982117,
      "learning_rate": 4.999957776717526e-05,
      "loss": 0.3238,
      "num_input_tokens_seen": 140704,
      "step": 75
    },
    {
      "epoch": 0.013051635533077739,
      "grad_norm": 0.09399475902318954,
      "learning_rate": 4.9999518781222984e-05,
      "loss": 0.3161,
      "num_input_tokens_seen": 149328,
      "step": 80
    },
    {
      "epoch": 0.013867362753895097,
      "grad_norm": 0.19451294839382172,
      "learning_rate": 4.9999455940030746e-05,
      "loss": 0.1879,
      "num_input_tokens_seen": 158576,
      "step": 85
    },
    {
      "epoch": 0.014683089974712456,
      "grad_norm": 4.243290424346924,
      "learning_rate": 4.999938924360824e-05,
      "loss": 0.1447,
      "num_input_tokens_seen": 168576,
      "step": 90
    },
    {
      "epoch": 0.015498817195529814,
      "grad_norm": 5.374866962432861,
      "learning_rate": 4.999931869196575e-05,
      "loss": 0.44,
      "num_input_tokens_seen": 177440,
      "step": 95
    },
    {
      "epoch": 0.016314544416347172,
      "grad_norm": 4.244633674621582,
      "learning_rate": 4.999924428511416e-05,
      "loss": 0.3155,
      "num_input_tokens_seen": 187888,
      "step": 100
    },
    {
      "epoch": 0.017130271637164532,
      "grad_norm": 0.06263105571269989,
      "learning_rate": 4.999916602306494e-05,
      "loss": 0.167,
      "num_input_tokens_seen": 197824,
      "step": 105
    },
    {
      "epoch": 0.01794599885798189,
      "grad_norm": 5.20764684677124,
      "learning_rate": 4.999908390583016e-05,
      "loss": 0.7361,
      "num_input_tokens_seen": 208352,
      "step": 110
    },
    {
      "epoch": 0.01876172607879925,
      "grad_norm": 0.2456064373254776,
      "learning_rate": 4.999899793342247e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 216624,
      "step": 115
    },
    {
      "epoch": 0.019577453299616607,
      "grad_norm": 3.5186610221862793,
      "learning_rate": 4.999890810585516e-05,
      "loss": 0.1994,
      "num_input_tokens_seen": 225616,
      "step": 120
    },
    {
      "epoch": 0.020393180520433967,
      "grad_norm": 0.08733075112104416,
      "learning_rate": 4.999881442314206e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 235312,
      "step": 125
    },
    {
      "epoch": 0.021208907741251327,
      "grad_norm": 3.4193387031555176,
      "learning_rate": 4.9998716885297617e-05,
      "loss": 0.2132,
      "num_input_tokens_seen": 244688,
      "step": 130
    },
    {
      "epoch": 0.022024634962068683,
      "grad_norm": 0.4572628438472748,
      "learning_rate": 4.999861549233688e-05,
      "loss": 0.3448,
      "num_input_tokens_seen": 254720,
      "step": 135
    },
    {
      "epoch": 0.022840362182886043,
      "grad_norm": 0.3799428939819336,
      "learning_rate": 4.999851024427548e-05,
      "loss": 0.1192,
      "num_input_tokens_seen": 263616,
      "step": 140
    },
    {
      "epoch": 0.023656089403703402,
      "grad_norm": 0.03015916235744953,
      "learning_rate": 4.999840114112965e-05,
      "loss": 0.1949,
      "num_input_tokens_seen": 273120,
      "step": 145
    },
    {
      "epoch": 0.024471816624520762,
      "grad_norm": 2.963172197341919,
      "learning_rate": 4.999828818291621e-05,
      "loss": 0.3865,
      "num_input_tokens_seen": 282016,
      "step": 150
    },
    {
      "epoch": 0.025287543845338118,
      "grad_norm": 8.374654769897461,
      "learning_rate": 4.999817136965259e-05,
      "loss": 0.5077,
      "num_input_tokens_seen": 292416,
      "step": 155
    },
    {
      "epoch": 0.026103271066155478,
      "grad_norm": 2.4550392627716064,
      "learning_rate": 4.9998050701356794e-05,
      "loss": 0.4841,
      "num_input_tokens_seen": 302544,
      "step": 160
    },
    {
      "epoch": 0.026918998286972837,
      "grad_norm": 0.02012837491929531,
      "learning_rate": 4.999792617804744e-05,
      "loss": 0.2651,
      "num_input_tokens_seen": 312704,
      "step": 165
    },
    {
      "epoch": 0.027734725507790194,
      "grad_norm": 3.997022867202759,
      "learning_rate": 4.9997797799743724e-05,
      "loss": 0.2585,
      "num_input_tokens_seen": 322768,
      "step": 170
    },
    {
      "epoch": 0.028550452728607553,
      "grad_norm": 0.4285709857940674,
      "learning_rate": 4.999766556646545e-05,
      "loss": 0.0587,
      "num_input_tokens_seen": 331408,
      "step": 175
    },
    {
      "epoch": 0.029366179949424913,
      "grad_norm": 4.138998508453369,
      "learning_rate": 4.9997529478232996e-05,
      "loss": 0.2665,
      "num_input_tokens_seen": 340736,
      "step": 180
    },
    {
      "epoch": 0.030181907170242273,
      "grad_norm": 2.0220532417297363,
      "learning_rate": 4.9997389535067365e-05,
      "loss": 0.2245,
      "num_input_tokens_seen": 350416,
      "step": 185
    },
    {
      "epoch": 0.03099763439105963,
      "grad_norm": 1.753010869026184,
      "learning_rate": 4.999724573699012e-05,
      "loss": 0.3984,
      "num_input_tokens_seen": 359632,
      "step": 190
    },
    {
      "epoch": 0.031813361611876985,
      "grad_norm": 0.02388136461377144,
      "learning_rate": 4.9997098084023457e-05,
      "loss": 0.2189,
      "num_input_tokens_seen": 369600,
      "step": 195
    },
    {
      "epoch": 0.032629088832694345,
      "grad_norm": 1.3546841144561768,
      "learning_rate": 4.999694657619013e-05,
      "loss": 0.1201,
      "num_input_tokens_seen": 378944,
      "step": 200
    },
    {
      "epoch": 0.032629088832694345,
      "eval_loss": 0.3065047264099121,
      "eval_runtime": 133.1355,
      "eval_samples_per_second": 20.468,
      "eval_steps_per_second": 10.238,
      "num_input_tokens_seen": 378944,
      "step": 200
    },
    {
      "epoch": 0.033444816053511704,
      "grad_norm": 4.560217380523682,
      "learning_rate": 4.999679121351352e-05,
      "loss": 0.3496,
      "num_input_tokens_seen": 388816,
      "step": 205
    },
    {
      "epoch": 0.034260543274329064,
      "grad_norm": 0.8994247317314148,
      "learning_rate": 4.9996631996017565e-05,
      "loss": 0.0893,
      "num_input_tokens_seen": 399392,
      "step": 210
    },
    {
      "epoch": 0.035076270495146424,
      "grad_norm": 7.2808356285095215,
      "learning_rate": 4.9996468923726835e-05,
      "loss": 0.3573,
      "num_input_tokens_seen": 408832,
      "step": 215
    },
    {
      "epoch": 0.03589199771596378,
      "grad_norm": 4.1855082511901855,
      "learning_rate": 4.999630199666647e-05,
      "loss": 0.6332,
      "num_input_tokens_seen": 419456,
      "step": 220
    },
    {
      "epoch": 0.03670772493678114,
      "grad_norm": 1.1257834434509277,
      "learning_rate": 4.999613121486222e-05,
      "loss": 0.2045,
      "num_input_tokens_seen": 429280,
      "step": 225
    },
    {
      "epoch": 0.0375234521575985,
      "grad_norm": 0.019865121692419052,
      "learning_rate": 4.999595657834041e-05,
      "loss": 0.1993,
      "num_input_tokens_seen": 438416,
      "step": 230
    },
    {
      "epoch": 0.038339179378415855,
      "grad_norm": 0.7706409096717834,
      "learning_rate": 4.999577808712798e-05,
      "loss": 0.0836,
      "num_input_tokens_seen": 447600,
      "step": 235
    },
    {
      "epoch": 0.039154906599233215,
      "grad_norm": 0.06344639509916306,
      "learning_rate": 4.999559574125244e-05,
      "loss": 0.2334,
      "num_input_tokens_seen": 457680,
      "step": 240
    },
    {
      "epoch": 0.039970633820050575,
      "grad_norm": 3.6359057426452637,
      "learning_rate": 4.9995409540741934e-05,
      "loss": 0.4823,
      "num_input_tokens_seen": 466496,
      "step": 245
    },
    {
      "epoch": 0.040786361040867934,
      "grad_norm": 4.816012382507324,
      "learning_rate": 4.999521948562516e-05,
      "loss": 0.6169,
      "num_input_tokens_seen": 476400,
      "step": 250
    },
    {
      "epoch": 0.041602088261685294,
      "grad_norm": 1.4476065635681152,
      "learning_rate": 4.999502557593143e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 484960,
      "step": 255
    },
    {
      "epoch": 0.042417815482502653,
      "grad_norm": 2.2326223850250244,
      "learning_rate": 4.999482781169066e-05,
      "loss": 0.3858,
      "num_input_tokens_seen": 493760,
      "step": 260
    },
    {
      "epoch": 0.04323354270332001,
      "grad_norm": 3.306335687637329,
      "learning_rate": 4.9994626192933324e-05,
      "loss": 0.1977,
      "num_input_tokens_seen": 504400,
      "step": 265
    },
    {
      "epoch": 0.044049269924137366,
      "grad_norm": 0.03997604548931122,
      "learning_rate": 4.999442071969054e-05,
      "loss": 0.0719,
      "num_input_tokens_seen": 513264,
      "step": 270
    },
    {
      "epoch": 0.044864997144954726,
      "grad_norm": 0.4812820255756378,
      "learning_rate": 4.999421139199397e-05,
      "loss": 0.0111,
      "num_input_tokens_seen": 523216,
      "step": 275
    },
    {
      "epoch": 0.045680724365772085,
      "grad_norm": 0.34986284375190735,
      "learning_rate": 4.999399820987592e-05,
      "loss": 0.2565,
      "num_input_tokens_seen": 532832,
      "step": 280
    },
    {
      "epoch": 0.046496451586589445,
      "grad_norm": 3.6027774810791016,
      "learning_rate": 4.999378117336924e-05,
      "loss": 0.1066,
      "num_input_tokens_seen": 542272,
      "step": 285
    },
    {
      "epoch": 0.047312178807406804,
      "grad_norm": 3.6849751472473145,
      "learning_rate": 4.9993560282507415e-05,
      "loss": 0.4416,
      "num_input_tokens_seen": 551152,
      "step": 290
    },
    {
      "epoch": 0.048127906028224164,
      "grad_norm": 1.6279445886611938,
      "learning_rate": 4.9993335537324495e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 560880,
      "step": 295
    },
    {
      "epoch": 0.048943633249041524,
      "grad_norm": 0.34825941920280457,
      "learning_rate": 4.999310693785516e-05,
      "loss": 0.1137,
      "num_input_tokens_seen": 570320,
      "step": 300
    },
    {
      "epoch": 0.049759360469858877,
      "grad_norm": 6.723934173583984,
      "learning_rate": 4.9992874484134653e-05,
      "loss": 0.3389,
      "num_input_tokens_seen": 581328,
      "step": 305
    },
    {
      "epoch": 0.050575087690676236,
      "grad_norm": 2.423750638961792,
      "learning_rate": 4.999263817619882e-05,
      "loss": 0.1036,
      "num_input_tokens_seen": 590080,
      "step": 310
    },
    {
      "epoch": 0.051390814911493596,
      "grad_norm": 1.9903299808502197,
      "learning_rate": 4.9992398014084105e-05,
      "loss": 0.2871,
      "num_input_tokens_seen": 600064,
      "step": 315
    },
    {
      "epoch": 0.052206542132310955,
      "grad_norm": 1.8764643669128418,
      "learning_rate": 4.999215399782754e-05,
      "loss": 0.4458,
      "num_input_tokens_seen": 609920,
      "step": 320
    },
    {
      "epoch": 0.053022269353128315,
      "grad_norm": 1.736266851425171,
      "learning_rate": 4.999190612746675e-05,
      "loss": 0.3443,
      "num_input_tokens_seen": 618960,
      "step": 325
    },
    {
      "epoch": 0.053837996573945675,
      "grad_norm": 1.9124863147735596,
      "learning_rate": 4.999165440303998e-05,
      "loss": 0.345,
      "num_input_tokens_seen": 627680,
      "step": 330
    },
    {
      "epoch": 0.054653723794763034,
      "grad_norm": 0.16417156159877777,
      "learning_rate": 4.999139882458603e-05,
      "loss": 0.2275,
      "num_input_tokens_seen": 637520,
      "step": 335
    },
    {
      "epoch": 0.05546945101558039,
      "grad_norm": 0.10691526532173157,
      "learning_rate": 4.9991139392144314e-05,
      "loss": 0.2989,
      "num_input_tokens_seen": 646928,
      "step": 340
    },
    {
      "epoch": 0.05628517823639775,
      "grad_norm": 0.06232095882296562,
      "learning_rate": 4.999087610575485e-05,
      "loss": 0.3439,
      "num_input_tokens_seen": 655344,
      "step": 345
    },
    {
      "epoch": 0.057100905457215106,
      "grad_norm": 1.0880045890808105,
      "learning_rate": 4.999060896545824e-05,
      "loss": 0.0671,
      "num_input_tokens_seen": 664368,
      "step": 350
    },
    {
      "epoch": 0.057916632678032466,
      "grad_norm": 1.2919373512268066,
      "learning_rate": 4.999033797129568e-05,
      "loss": 0.1075,
      "num_input_tokens_seen": 675056,
      "step": 355
    },
    {
      "epoch": 0.058732359898849826,
      "grad_norm": 0.7277734875679016,
      "learning_rate": 4.999006312330894e-05,
      "loss": 0.1326,
      "num_input_tokens_seen": 683872,
      "step": 360
    },
    {
      "epoch": 0.059548087119667185,
      "grad_norm": 0.0148468604311347,
      "learning_rate": 4.998978442154043e-05,
      "loss": 0.4944,
      "num_input_tokens_seen": 694576,
      "step": 365
    },
    {
      "epoch": 0.060363814340484545,
      "grad_norm": 4.1964111328125,
      "learning_rate": 4.9989501866033125e-05,
      "loss": 0.1783,
      "num_input_tokens_seen": 704272,
      "step": 370
    },
    {
      "epoch": 0.0611795415613019,
      "grad_norm": 1.6984230279922485,
      "learning_rate": 4.998921545683059e-05,
      "loss": 0.3116,
      "num_input_tokens_seen": 713984,
      "step": 375
    },
    {
      "epoch": 0.06199526878211926,
      "grad_norm": 0.6425628066062927,
      "learning_rate": 4.9988925193976996e-05,
      "loss": 0.2184,
      "num_input_tokens_seen": 723360,
      "step": 380
    },
    {
      "epoch": 0.06281099600293662,
      "grad_norm": 3.616333484649658,
      "learning_rate": 4.998863107751711e-05,
      "loss": 0.3639,
      "num_input_tokens_seen": 732064,
      "step": 385
    },
    {
      "epoch": 0.06362672322375397,
      "grad_norm": 1.6806422472000122,
      "learning_rate": 4.998833310749629e-05,
      "loss": 0.2923,
      "num_input_tokens_seen": 740336,
      "step": 390
    },
    {
      "epoch": 0.06444245044457134,
      "grad_norm": 3.385082483291626,
      "learning_rate": 4.998803128396047e-05,
      "loss": 0.1751,
      "num_input_tokens_seen": 748960,
      "step": 395
    },
    {
      "epoch": 0.06525817766538869,
      "grad_norm": 0.6792603731155396,
      "learning_rate": 4.9987725606956215e-05,
      "loss": 0.275,
      "num_input_tokens_seen": 758192,
      "step": 400
    },
    {
      "epoch": 0.06525817766538869,
      "eval_loss": 0.25375238060951233,
      "eval_runtime": 133.0909,
      "eval_samples_per_second": 20.475,
      "eval_steps_per_second": 10.241,
      "num_input_tokens_seen": 758192,
      "step": 400
    },
    {
      "epoch": 0.06607390488620606,
      "grad_norm": 1.3958001136779785,
      "learning_rate": 4.998741607653066e-05,
      "loss": 0.1965,
      "num_input_tokens_seen": 767184,
      "step": 405
    },
    {
      "epoch": 0.06688963210702341,
      "grad_norm": 1.3058812618255615,
      "learning_rate": 4.9987102692731523e-05,
      "loss": 0.2204,
      "num_input_tokens_seen": 776784,
      "step": 410
    },
    {
      "epoch": 0.06770535932784078,
      "grad_norm": 0.515053391456604,
      "learning_rate": 4.9986785455607157e-05,
      "loss": 0.1962,
      "num_input_tokens_seen": 786640,
      "step": 415
    },
    {
      "epoch": 0.06852108654865813,
      "grad_norm": 0.36713916063308716,
      "learning_rate": 4.9986464365206456e-05,
      "loss": 0.053,
      "num_input_tokens_seen": 796624,
      "step": 420
    },
    {
      "epoch": 0.06933681376947548,
      "grad_norm": 0.18395482003688812,
      "learning_rate": 4.9986139421578956e-05,
      "loss": 0.2344,
      "num_input_tokens_seen": 805920,
      "step": 425
    },
    {
      "epoch": 0.07015254099029285,
      "grad_norm": 1.7785470485687256,
      "learning_rate": 4.998581062477477e-05,
      "loss": 0.237,
      "num_input_tokens_seen": 815664,
      "step": 430
    },
    {
      "epoch": 0.0709682682111102,
      "grad_norm": 0.5035198926925659,
      "learning_rate": 4.998547797484458e-05,
      "loss": 0.3144,
      "num_input_tokens_seen": 825328,
      "step": 435
    },
    {
      "epoch": 0.07178399543192757,
      "grad_norm": 1.0419203042984009,
      "learning_rate": 4.9985141471839706e-05,
      "loss": 0.4923,
      "num_input_tokens_seen": 834336,
      "step": 440
    },
    {
      "epoch": 0.07259972265274492,
      "grad_norm": 1.7538670301437378,
      "learning_rate": 4.998480111581203e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 843584,
      "step": 445
    },
    {
      "epoch": 0.07341544987356229,
      "grad_norm": 10.441343307495117,
      "learning_rate": 4.998445690681405e-05,
      "loss": 0.4999,
      "num_input_tokens_seen": 853040,
      "step": 450
    },
    {
      "epoch": 0.07423117709437964,
      "grad_norm": 0.2567311227321625,
      "learning_rate": 4.9984108844898834e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 862992,
      "step": 455
    },
    {
      "epoch": 0.075046904315197,
      "grad_norm": 3.522549629211426,
      "learning_rate": 4.9983756930120076e-05,
      "loss": 0.1405,
      "num_input_tokens_seen": 872048,
      "step": 460
    },
    {
      "epoch": 0.07586263153601436,
      "grad_norm": 0.3583533763885498,
      "learning_rate": 4.9983401162532025e-05,
      "loss": 0.2245,
      "num_input_tokens_seen": 882272,
      "step": 465
    },
    {
      "epoch": 0.07667835875683171,
      "grad_norm": 2.2436065673828125,
      "learning_rate": 4.998304154218955e-05,
      "loss": 0.2011,
      "num_input_tokens_seen": 892688,
      "step": 470
    },
    {
      "epoch": 0.07749408597764908,
      "grad_norm": 2.4565656185150146,
      "learning_rate": 4.998267806914812e-05,
      "loss": 0.2947,
      "num_input_tokens_seen": 902144,
      "step": 475
    },
    {
      "epoch": 0.07830981319846643,
      "grad_norm": 0.2723786532878876,
      "learning_rate": 4.998231074346378e-05,
      "loss": 0.3055,
      "num_input_tokens_seen": 911024,
      "step": 480
    },
    {
      "epoch": 0.0791255404192838,
      "grad_norm": 2.8036816120147705,
      "learning_rate": 4.998193956519317e-05,
      "loss": 0.2369,
      "num_input_tokens_seen": 921408,
      "step": 485
    },
    {
      "epoch": 0.07994126764010115,
      "grad_norm": 3.4678213596343994,
      "learning_rate": 4.9981564534393545e-05,
      "loss": 0.3742,
      "num_input_tokens_seen": 931440,
      "step": 490
    },
    {
      "epoch": 0.08075699486091852,
      "grad_norm": 0.09396962076425552,
      "learning_rate": 4.998118565112272e-05,
      "loss": 0.5301,
      "num_input_tokens_seen": 939280,
      "step": 495
    },
    {
      "epoch": 0.08157272208173587,
      "grad_norm": 2.5918099880218506,
      "learning_rate": 4.998080291543914e-05,
      "loss": 0.2524,
      "num_input_tokens_seen": 949360,
      "step": 500
    },
    {
      "epoch": 0.08238844930255322,
      "grad_norm": 0.7557159066200256,
      "learning_rate": 4.9980416327401826e-05,
      "loss": 0.4437,
      "num_input_tokens_seen": 958672,
      "step": 505
    },
    {
      "epoch": 0.08320417652337059,
      "grad_norm": 0.8074548244476318,
      "learning_rate": 4.998002588707038e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 967456,
      "step": 510
    },
    {
      "epoch": 0.08401990374418794,
      "grad_norm": 0.5129878520965576,
      "learning_rate": 4.997963159450503e-05,
      "loss": 0.1326,
      "num_input_tokens_seen": 977360,
      "step": 515
    },
    {
      "epoch": 0.08483563096500531,
      "grad_norm": 2.549395799636841,
      "learning_rate": 4.9979233449766575e-05,
      "loss": 0.3493,
      "num_input_tokens_seen": 986368,
      "step": 520
    },
    {
      "epoch": 0.08565135818582266,
      "grad_norm": 1.9045841693878174,
      "learning_rate": 4.997883145291641e-05,
      "loss": 0.31,
      "num_input_tokens_seen": 995776,
      "step": 525
    },
    {
      "epoch": 0.08646708540664003,
      "grad_norm": 0.21935094892978668,
      "learning_rate": 4.9978425604016536e-05,
      "loss": 0.2587,
      "num_input_tokens_seen": 1005280,
      "step": 530
    },
    {
      "epoch": 0.08728281262745738,
      "grad_norm": 1.903395175933838,
      "learning_rate": 4.9978015903129536e-05,
      "loss": 0.4182,
      "num_input_tokens_seen": 1015600,
      "step": 535
    },
    {
      "epoch": 0.08809853984827473,
      "grad_norm": 0.9490047097206116,
      "learning_rate": 4.997760235031859e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 1024912,
      "step": 540
    },
    {
      "epoch": 0.0889142670690921,
      "grad_norm": 0.717349112033844,
      "learning_rate": 4.9977184945647473e-05,
      "loss": 0.1957,
      "num_input_tokens_seen": 1034848,
      "step": 545
    },
    {
      "epoch": 0.08972999428990945,
      "grad_norm": 0.9406548142433167,
      "learning_rate": 4.997676368918055e-05,
      "loss": 0.3074,
      "num_input_tokens_seen": 1045648,
      "step": 550
    },
    {
      "epoch": 0.09054572151072682,
      "grad_norm": 3.009089946746826,
      "learning_rate": 4.9976338580982794e-05,
      "loss": 0.3554,
      "num_input_tokens_seen": 1056416,
      "step": 555
    },
    {
      "epoch": 0.09136144873154417,
      "grad_norm": 0.5453741550445557,
      "learning_rate": 4.9975909621119755e-05,
      "loss": 0.0557,
      "num_input_tokens_seen": 1066048,
      "step": 560
    },
    {
      "epoch": 0.09217717595236154,
      "grad_norm": 0.7799529433250427,
      "learning_rate": 4.997547680965758e-05,
      "loss": 0.158,
      "num_input_tokens_seen": 1075072,
      "step": 565
    },
    {
      "epoch": 0.09299290317317889,
      "grad_norm": 1.496537685394287,
      "learning_rate": 4.997504014666302e-05,
      "loss": 0.3264,
      "num_input_tokens_seen": 1083968,
      "step": 570
    },
    {
      "epoch": 0.09380863039399624,
      "grad_norm": 1.0193756818771362,
      "learning_rate": 4.997459963220342e-05,
      "loss": 0.1418,
      "num_input_tokens_seen": 1093728,
      "step": 575
    },
    {
      "epoch": 0.09462435761481361,
      "grad_norm": 1.8761909008026123,
      "learning_rate": 4.997415526634671e-05,
      "loss": 0.1515,
      "num_input_tokens_seen": 1103280,
      "step": 580
    },
    {
      "epoch": 0.09544008483563096,
      "grad_norm": 1.2879223823547363,
      "learning_rate": 4.99737070491614e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 1111744,
      "step": 585
    },
    {
      "epoch": 0.09625581205644833,
      "grad_norm": 0.8839752078056335,
      "learning_rate": 4.997325498071663e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 1122080,
      "step": 590
    },
    {
      "epoch": 0.09707153927726568,
      "grad_norm": 1.3635693788528442,
      "learning_rate": 4.997279906108211e-05,
      "loss": 0.0955,
      "num_input_tokens_seen": 1132096,
      "step": 595
    },
    {
      "epoch": 0.09788726649808305,
      "grad_norm": 0.7364757061004639,
      "learning_rate": 4.9972339290328155e-05,
      "loss": 0.3329,
      "num_input_tokens_seen": 1141408,
      "step": 600
    },
    {
      "epoch": 0.09788726649808305,
      "eval_loss": 0.2203824818134308,
      "eval_runtime": 132.9857,
      "eval_samples_per_second": 20.491,
      "eval_steps_per_second": 10.249,
      "num_input_tokens_seen": 1141408,
      "step": 600
    },
    {
      "epoch": 0.0987029937189004,
      "grad_norm": 0.9392332434654236,
      "learning_rate": 4.9971875668525646e-05,
      "loss": 0.2042,
      "num_input_tokens_seen": 1150912,
      "step": 605
    },
    {
      "epoch": 0.09951872093971775,
      "grad_norm": 0.06006195768713951,
      "learning_rate": 4.997140819574609e-05,
      "loss": 0.1925,
      "num_input_tokens_seen": 1160672,
      "step": 610
    },
    {
      "epoch": 0.10033444816053512,
      "grad_norm": 2.857642412185669,
      "learning_rate": 4.997093687206159e-05,
      "loss": 0.5693,
      "num_input_tokens_seen": 1170672,
      "step": 615
    },
    {
      "epoch": 0.10115017538135247,
      "grad_norm": 2.4577300548553467,
      "learning_rate": 4.997046169754482e-05,
      "loss": 0.3617,
      "num_input_tokens_seen": 1179792,
      "step": 620
    },
    {
      "epoch": 0.10196590260216984,
      "grad_norm": 0.1960863620042801,
      "learning_rate": 4.996998267226905e-05,
      "loss": 0.1217,
      "num_input_tokens_seen": 1189120,
      "step": 625
    },
    {
      "epoch": 0.10278162982298719,
      "grad_norm": 1.298007845878601,
      "learning_rate": 4.996949979630817e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 1198272,
      "step": 630
    },
    {
      "epoch": 0.10359735704380456,
      "grad_norm": 0.8937551379203796,
      "learning_rate": 4.996901306973663e-05,
      "loss": 0.4238,
      "num_input_tokens_seen": 1207216,
      "step": 635
    },
    {
      "epoch": 0.10441308426462191,
      "grad_norm": 0.6815471649169922,
      "learning_rate": 4.996852249262949e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 1215216,
      "step": 640
    },
    {
      "epoch": 0.10522881148543926,
      "grad_norm": 1.165092945098877,
      "learning_rate": 4.996802806506241e-05,
      "loss": 0.1945,
      "num_input_tokens_seen": 1224672,
      "step": 645
    },
    {
      "epoch": 0.10604453870625663,
      "grad_norm": 1.861132025718689,
      "learning_rate": 4.996752978711164e-05,
      "loss": 0.1963,
      "num_input_tokens_seen": 1234192,
      "step": 650
    },
    {
      "epoch": 0.10686026592707398,
      "grad_norm": 0.21975259482860565,
      "learning_rate": 4.996702765885401e-05,
      "loss": 0.0841,
      "num_input_tokens_seen": 1244192,
      "step": 655
    },
    {
      "epoch": 0.10767599314789135,
      "grad_norm": 0.5029909610748291,
      "learning_rate": 4.9966521680366964e-05,
      "loss": 0.5551,
      "num_input_tokens_seen": 1253024,
      "step": 660
    },
    {
      "epoch": 0.1084917203687087,
      "grad_norm": 4.120723247528076,
      "learning_rate": 4.9966011851728524e-05,
      "loss": 0.1799,
      "num_input_tokens_seen": 1261120,
      "step": 665
    },
    {
      "epoch": 0.10930744758952607,
      "grad_norm": 0.6678934693336487,
      "learning_rate": 4.996549817301731e-05,
      "loss": 0.2378,
      "num_input_tokens_seen": 1271168,
      "step": 670
    },
    {
      "epoch": 0.11012317481034342,
      "grad_norm": 0.6753316521644592,
      "learning_rate": 4.9964980644312544e-05,
      "loss": 0.1099,
      "num_input_tokens_seen": 1280848,
      "step": 675
    },
    {
      "epoch": 0.11093890203116077,
      "grad_norm": 1.0373022556304932,
      "learning_rate": 4.996445926569403e-05,
      "loss": 0.2054,
      "num_input_tokens_seen": 1290784,
      "step": 680
    },
    {
      "epoch": 0.11175462925197814,
      "grad_norm": 2.9150149822235107,
      "learning_rate": 4.996393403724218e-05,
      "loss": 0.2612,
      "num_input_tokens_seen": 1299568,
      "step": 685
    },
    {
      "epoch": 0.1125703564727955,
      "grad_norm": 1.9504966735839844,
      "learning_rate": 4.9963404959037985e-05,
      "loss": 0.1704,
      "num_input_tokens_seen": 1308960,
      "step": 690
    },
    {
      "epoch": 0.11338608369361286,
      "grad_norm": 3.0242388248443604,
      "learning_rate": 4.996287203116303e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 1319184,
      "step": 695
    },
    {
      "epoch": 0.11420181091443021,
      "grad_norm": 0.7070112824440002,
      "learning_rate": 4.996233525369951e-05,
      "loss": 0.2121,
      "num_input_tokens_seen": 1328304,
      "step": 700
    },
    {
      "epoch": 0.11501753813524758,
      "grad_norm": 0.17610368132591248,
      "learning_rate": 4.99617946267302e-05,
      "loss": 0.3084,
      "num_input_tokens_seen": 1337648,
      "step": 705
    },
    {
      "epoch": 0.11583326535606493,
      "grad_norm": 0.4266015887260437,
      "learning_rate": 4.996125015033846e-05,
      "loss": 0.1036,
      "num_input_tokens_seen": 1348384,
      "step": 710
    },
    {
      "epoch": 0.11664899257688228,
      "grad_norm": 0.7637392282485962,
      "learning_rate": 4.996070182460827e-05,
      "loss": 0.2107,
      "num_input_tokens_seen": 1358272,
      "step": 715
    },
    {
      "epoch": 0.11746471979769965,
      "grad_norm": 0.62162184715271,
      "learning_rate": 4.996014964962418e-05,
      "loss": 0.2261,
      "num_input_tokens_seen": 1366864,
      "step": 720
    },
    {
      "epoch": 0.118280447018517,
      "grad_norm": 0.6179205775260925,
      "learning_rate": 4.9959593625471344e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 1376384,
      "step": 725
    },
    {
      "epoch": 0.11909617423933437,
      "grad_norm": 1.414536476135254,
      "learning_rate": 4.995903375223552e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 1385088,
      "step": 730
    },
    {
      "epoch": 0.11991190146015172,
      "grad_norm": 4.484546184539795,
      "learning_rate": 4.995847003000302e-05,
      "loss": 0.4657,
      "num_input_tokens_seen": 1394816,
      "step": 735
    },
    {
      "epoch": 0.12072762868096909,
      "grad_norm": 2.969094753265381,
      "learning_rate": 4.9957902458860804e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 1404912,
      "step": 740
    },
    {
      "epoch": 0.12154335590178644,
      "grad_norm": 0.770622193813324,
      "learning_rate": 4.995733103889639e-05,
      "loss": 0.2698,
      "num_input_tokens_seen": 1415376,
      "step": 745
    },
    {
      "epoch": 0.1223590831226038,
      "grad_norm": 0.7591431140899658,
      "learning_rate": 4.99567557701979e-05,
      "loss": 0.1192,
      "num_input_tokens_seen": 1425456,
      "step": 750
    },
    {
      "epoch": 0.12317481034342116,
      "grad_norm": 0.9745700359344482,
      "learning_rate": 4.995617665285403e-05,
      "loss": 0.0811,
      "num_input_tokens_seen": 1435344,
      "step": 755
    },
    {
      "epoch": 0.12399053756423851,
      "grad_norm": 1.9898005723953247,
      "learning_rate": 4.99555936869541e-05,
      "loss": 0.3669,
      "num_input_tokens_seen": 1445584,
      "step": 760
    },
    {
      "epoch": 0.12480626478505588,
      "grad_norm": 0.08259454369544983,
      "learning_rate": 4.995500687258803e-05,
      "loss": 0.049,
      "num_input_tokens_seen": 1455056,
      "step": 765
    },
    {
      "epoch": 0.12562199200587323,
      "grad_norm": 1.548717737197876,
      "learning_rate": 4.995441620984628e-05,
      "loss": 0.2098,
      "num_input_tokens_seen": 1464448,
      "step": 770
    },
    {
      "epoch": 0.1264377192266906,
      "grad_norm": 2.613105297088623,
      "learning_rate": 4.995382169881996e-05,
      "loss": 0.2181,
      "num_input_tokens_seen": 1472672,
      "step": 775
    },
    {
      "epoch": 0.12725344644750794,
      "grad_norm": 1.215476155281067,
      "learning_rate": 4.9953223339600755e-05,
      "loss": 0.4414,
      "num_input_tokens_seen": 1481712,
      "step": 780
    },
    {
      "epoch": 0.12806917366832532,
      "grad_norm": 5.444200038909912,
      "learning_rate": 4.995262113228091e-05,
      "loss": 0.2238,
      "num_input_tokens_seen": 1490704,
      "step": 785
    },
    {
      "epoch": 0.12888490088914267,
      "grad_norm": 1.913511872291565,
      "learning_rate": 4.995201507695332e-05,
      "loss": 0.3211,
      "num_input_tokens_seen": 1500688,
      "step": 790
    },
    {
      "epoch": 0.12970062810996003,
      "grad_norm": 1.35863196849823,
      "learning_rate": 4.995140517371144e-05,
      "loss": 0.2364,
      "num_input_tokens_seen": 1508960,
      "step": 795
    },
    {
      "epoch": 0.13051635533077738,
      "grad_norm": 0.8200124502182007,
      "learning_rate": 4.995079142264932e-05,
      "loss": 0.3121,
      "num_input_tokens_seen": 1518336,
      "step": 800
    },
    {
      "epoch": 0.13051635533077738,
      "eval_loss": 0.20619988441467285,
      "eval_runtime": 133.162,
      "eval_samples_per_second": 20.464,
      "eval_steps_per_second": 10.236,
      "num_input_tokens_seen": 1518336,
      "step": 800
    },
    {
      "epoch": 0.13133208255159476,
      "grad_norm": 0.7079024314880371,
      "learning_rate": 4.995017382386162e-05,
      "loss": 0.2025,
      "num_input_tokens_seen": 1527888,
      "step": 805
    },
    {
      "epoch": 0.1321478097724121,
      "grad_norm": 1.13783597946167,
      "learning_rate": 4.994955237744356e-05,
      "loss": 0.2167,
      "num_input_tokens_seen": 1536976,
      "step": 810
    },
    {
      "epoch": 0.13296353699322946,
      "grad_norm": 1.2459145784378052,
      "learning_rate": 4.994892708349101e-05,
      "loss": 0.055,
      "num_input_tokens_seen": 1547152,
      "step": 815
    },
    {
      "epoch": 0.13377926421404682,
      "grad_norm": 1.643128752708435,
      "learning_rate": 4.994829794210035e-05,
      "loss": 0.1726,
      "num_input_tokens_seen": 1556704,
      "step": 820
    },
    {
      "epoch": 0.13459499143486417,
      "grad_norm": 0.06384608149528503,
      "learning_rate": 4.994766495336864e-05,
      "loss": 0.2073,
      "num_input_tokens_seen": 1565360,
      "step": 825
    },
    {
      "epoch": 0.13541071865568155,
      "grad_norm": 0.3464173972606659,
      "learning_rate": 4.994702811739348e-05,
      "loss": 0.1431,
      "num_input_tokens_seen": 1574640,
      "step": 830
    },
    {
      "epoch": 0.1362264458764989,
      "grad_norm": 3.037872076034546,
      "learning_rate": 4.994638743427308e-05,
      "loss": 0.2803,
      "num_input_tokens_seen": 1583936,
      "step": 835
    },
    {
      "epoch": 0.13704217309731626,
      "grad_norm": 0.34937795996665955,
      "learning_rate": 4.994574290410624e-05,
      "loss": 0.113,
      "num_input_tokens_seen": 1592992,
      "step": 840
    },
    {
      "epoch": 0.1378579003181336,
      "grad_norm": 0.7138415575027466,
      "learning_rate": 4.9945094526992364e-05,
      "loss": 0.4373,
      "num_input_tokens_seen": 1602160,
      "step": 845
    },
    {
      "epoch": 0.13867362753895096,
      "grad_norm": 2.137387990951538,
      "learning_rate": 4.994444230303142e-05,
      "loss": 0.1514,
      "num_input_tokens_seen": 1613280,
      "step": 850
    },
    {
      "epoch": 0.13948935475976834,
      "grad_norm": 0.3943535387516022,
      "learning_rate": 4.994378623232402e-05,
      "loss": 0.1534,
      "num_input_tokens_seen": 1623360,
      "step": 855
    },
    {
      "epoch": 0.1403050819805857,
      "grad_norm": 0.38238510489463806,
      "learning_rate": 4.99431263149713e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 1632960,
      "step": 860
    },
    {
      "epoch": 0.14112080920140305,
      "grad_norm": 1.4846388101577759,
      "learning_rate": 4.9942462551075056e-05,
      "loss": 0.2555,
      "num_input_tokens_seen": 1641936,
      "step": 865
    },
    {
      "epoch": 0.1419365364222204,
      "grad_norm": 1.212641716003418,
      "learning_rate": 4.994179494073764e-05,
      "loss": 0.3299,
      "num_input_tokens_seen": 1652896,
      "step": 870
    },
    {
      "epoch": 0.14275226364303778,
      "grad_norm": 2.2019104957580566,
      "learning_rate": 4.9941123484062e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 1664416,
      "step": 875
    },
    {
      "epoch": 0.14356799086385513,
      "grad_norm": 0.4717172682285309,
      "learning_rate": 4.99404481811517e-05,
      "loss": 0.1641,
      "num_input_tokens_seen": 1673680,
      "step": 880
    },
    {
      "epoch": 0.14438371808467249,
      "grad_norm": 0.7936664819717407,
      "learning_rate": 4.9939769032110864e-05,
      "loss": 0.0708,
      "num_input_tokens_seen": 1682256,
      "step": 885
    },
    {
      "epoch": 0.14519944530548984,
      "grad_norm": 2.5239264965057373,
      "learning_rate": 4.993908603704423e-05,
      "loss": 0.4489,
      "num_input_tokens_seen": 1691264,
      "step": 890
    },
    {
      "epoch": 0.1460151725263072,
      "grad_norm": 1.2555408477783203,
      "learning_rate": 4.9938399196057126e-05,
      "loss": 0.1943,
      "num_input_tokens_seen": 1699872,
      "step": 895
    },
    {
      "epoch": 0.14683089974712457,
      "grad_norm": 1.1134324073791504,
      "learning_rate": 4.993770850925547e-05,
      "loss": 0.3143,
      "num_input_tokens_seen": 1709008,
      "step": 900
    },
    {
      "epoch": 0.14764662696794192,
      "grad_norm": 1.846118450164795,
      "learning_rate": 4.993701397674577e-05,
      "loss": 0.1979,
      "num_input_tokens_seen": 1719216,
      "step": 905
    },
    {
      "epoch": 0.14846235418875928,
      "grad_norm": 0.06778592616319656,
      "learning_rate": 4.993631559863515e-05,
      "loss": 0.2822,
      "num_input_tokens_seen": 1729984,
      "step": 910
    },
    {
      "epoch": 0.14927808140957663,
      "grad_norm": 1.2722309827804565,
      "learning_rate": 4.9935613375031283e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 1739424,
      "step": 915
    },
    {
      "epoch": 0.150093808630394,
      "grad_norm": 0.410246878862381,
      "learning_rate": 4.993490730604248e-05,
      "loss": 0.2805,
      "num_input_tokens_seen": 1748496,
      "step": 920
    },
    {
      "epoch": 0.15090953585121136,
      "grad_norm": 2.301710367202759,
      "learning_rate": 4.993419739177761e-05,
      "loss": 0.1861,
      "num_input_tokens_seen": 1758336,
      "step": 925
    },
    {
      "epoch": 0.15172526307202872,
      "grad_norm": 0.1675238162279129,
      "learning_rate": 4.9933483632346164e-05,
      "loss": 0.0468,
      "num_input_tokens_seen": 1768208,
      "step": 930
    },
    {
      "epoch": 0.15254099029284607,
      "grad_norm": 2.4071149826049805,
      "learning_rate": 4.993276602785821e-05,
      "loss": 0.195,
      "num_input_tokens_seen": 1778672,
      "step": 935
    },
    {
      "epoch": 0.15335671751366342,
      "grad_norm": 1.4000799655914307,
      "learning_rate": 4.993204457842441e-05,
      "loss": 0.2891,
      "num_input_tokens_seen": 1788016,
      "step": 940
    },
    {
      "epoch": 0.1541724447344808,
      "grad_norm": 1.1289042234420776,
      "learning_rate": 4.993131928415602e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 1797072,
      "step": 945
    },
    {
      "epoch": 0.15498817195529815,
      "grad_norm": 0.3722313642501831,
      "learning_rate": 4.993059014516489e-05,
      "loss": 0.1321,
      "num_input_tokens_seen": 1807296,
      "step": 950
    },
    {
      "epoch": 0.1558038991761155,
      "grad_norm": 3.583338975906372,
      "learning_rate": 4.9929857161563464e-05,
      "loss": 0.2236,
      "num_input_tokens_seen": 1816160,
      "step": 955
    },
    {
      "epoch": 0.15661962639693286,
      "grad_norm": 2.3361334800720215,
      "learning_rate": 4.992912033346477e-05,
      "loss": 0.2496,
      "num_input_tokens_seen": 1827152,
      "step": 960
    },
    {
      "epoch": 0.1574353536177502,
      "grad_norm": 0.6830472350120544,
      "learning_rate": 4.992837966098245e-05,
      "loss": 0.0755,
      "num_input_tokens_seen": 1835968,
      "step": 965
    },
    {
      "epoch": 0.1582510808385676,
      "grad_norm": 0.742798388004303,
      "learning_rate": 4.992763514423071e-05,
      "loss": 0.1131,
      "num_input_tokens_seen": 1844800,
      "step": 970
    },
    {
      "epoch": 0.15906680805938495,
      "grad_norm": 1.39697265625,
      "learning_rate": 4.992688678332437e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 1854256,
      "step": 975
    },
    {
      "epoch": 0.1598825352802023,
      "grad_norm": 0.648740828037262,
      "learning_rate": 4.992613457837884e-05,
      "loss": 0.2254,
      "num_input_tokens_seen": 1863904,
      "step": 980
    },
    {
      "epoch": 0.16069826250101965,
      "grad_norm": 0.3987770974636078,
      "learning_rate": 4.992537852951011e-05,
      "loss": 0.0728,
      "num_input_tokens_seen": 1873168,
      "step": 985
    },
    {
      "epoch": 0.16151398972183703,
      "grad_norm": 1.7561815977096558,
      "learning_rate": 4.9924618636834785e-05,
      "loss": 0.1525,
      "num_input_tokens_seen": 1883328,
      "step": 990
    },
    {
      "epoch": 0.16232971694265438,
      "grad_norm": 2.4129958152770996,
      "learning_rate": 4.9923854900470046e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 1892528,
      "step": 995
    },
    {
      "epoch": 0.16314544416347174,
      "grad_norm": 0.4971231520175934,
      "learning_rate": 4.992308732053367e-05,
      "loss": 0.0258,
      "num_input_tokens_seen": 1901264,
      "step": 1000
    },
    {
      "epoch": 0.16314544416347174,
      "eval_loss": 0.19788126647472382,
      "eval_runtime": 133.1383,
      "eval_samples_per_second": 20.467,
      "eval_steps_per_second": 10.237,
      "num_input_tokens_seen": 1901264,
      "step": 1000
    },
    {
      "epoch": 0.1639611713842891,
      "grad_norm": 1.0221095085144043,
      "learning_rate": 4.992231589714402e-05,
      "loss": 0.2769,
      "num_input_tokens_seen": 1911680,
      "step": 1005
    },
    {
      "epoch": 0.16477689860510644,
      "grad_norm": 0.3021887242794037,
      "learning_rate": 4.992154063042007e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 1919872,
      "step": 1010
    },
    {
      "epoch": 0.16559262582592382,
      "grad_norm": 0.9342597723007202,
      "learning_rate": 4.992076152048136e-05,
      "loss": 0.1457,
      "num_input_tokens_seen": 1930480,
      "step": 1015
    },
    {
      "epoch": 0.16640835304674118,
      "grad_norm": 0.9198063015937805,
      "learning_rate": 4.991997856744807e-05,
      "loss": 0.3149,
      "num_input_tokens_seen": 1940048,
      "step": 1020
    },
    {
      "epoch": 0.16722408026755853,
      "grad_norm": 1.5209344625473022,
      "learning_rate": 4.9919191771440905e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 1949808,
      "step": 1025
    },
    {
      "epoch": 0.16803980748837588,
      "grad_norm": 0.5881098508834839,
      "learning_rate": 4.991840113258122e-05,
      "loss": 0.0843,
      "num_input_tokens_seen": 1959008,
      "step": 1030
    },
    {
      "epoch": 0.16885553470919323,
      "grad_norm": 2.087114095687866,
      "learning_rate": 4.9917606650990933e-05,
      "loss": 0.2454,
      "num_input_tokens_seen": 1968384,
      "step": 1035
    },
    {
      "epoch": 0.16967126193001061,
      "grad_norm": 1.5088270902633667,
      "learning_rate": 4.9916808326792566e-05,
      "loss": 0.2087,
      "num_input_tokens_seen": 1977904,
      "step": 1040
    },
    {
      "epoch": 0.17048698915082797,
      "grad_norm": 0.48583364486694336,
      "learning_rate": 4.9916006160109235e-05,
      "loss": 0.1026,
      "num_input_tokens_seen": 1987648,
      "step": 1045
    },
    {
      "epoch": 0.17130271637164532,
      "grad_norm": 0.548417866230011,
      "learning_rate": 4.991520015106464e-05,
      "loss": 0.078,
      "num_input_tokens_seen": 1997568,
      "step": 1050
    },
    {
      "epoch": 0.17211844359246267,
      "grad_norm": 0.14395496249198914,
      "learning_rate": 4.991439029978308e-05,
      "loss": 0.0371,
      "num_input_tokens_seen": 2005408,
      "step": 1055
    },
    {
      "epoch": 0.17293417081328005,
      "grad_norm": 1.7565174102783203,
      "learning_rate": 4.9913576606389434e-05,
      "loss": 0.2695,
      "num_input_tokens_seen": 2015232,
      "step": 1060
    },
    {
      "epoch": 0.1737498980340974,
      "grad_norm": 0.20170333981513977,
      "learning_rate": 4.991275907100919e-05,
      "loss": 0.2325,
      "num_input_tokens_seen": 2024800,
      "step": 1065
    },
    {
      "epoch": 0.17456562525491476,
      "grad_norm": 2.820068120956421,
      "learning_rate": 4.9911937693768434e-05,
      "loss": 0.2208,
      "num_input_tokens_seen": 2033712,
      "step": 1070
    },
    {
      "epoch": 0.1753813524757321,
      "grad_norm": 0.27787595987319946,
      "learning_rate": 4.991111247479382e-05,
      "loss": 0.1653,
      "num_input_tokens_seen": 2043264,
      "step": 1075
    },
    {
      "epoch": 0.17619707969654946,
      "grad_norm": 0.058811984956264496,
      "learning_rate": 4.9910283414212605e-05,
      "loss": 0.1143,
      "num_input_tokens_seen": 2052720,
      "step": 1080
    },
    {
      "epoch": 0.17701280691736684,
      "grad_norm": 0.6202765703201294,
      "learning_rate": 4.990945051215265e-05,
      "loss": 0.0982,
      "num_input_tokens_seen": 2062064,
      "step": 1085
    },
    {
      "epoch": 0.1778285341381842,
      "grad_norm": 0.3782899081707001,
      "learning_rate": 4.99086137687424e-05,
      "loss": 0.1046,
      "num_input_tokens_seen": 2071504,
      "step": 1090
    },
    {
      "epoch": 0.17864426135900155,
      "grad_norm": 0.2507684826850891,
      "learning_rate": 4.9907773184110874e-05,
      "loss": 0.0575,
      "num_input_tokens_seen": 2083440,
      "step": 1095
    },
    {
      "epoch": 0.1794599885798189,
      "grad_norm": 1.9338828325271606,
      "learning_rate": 4.9906928758387715e-05,
      "loss": 0.3049,
      "num_input_tokens_seen": 2092912,
      "step": 1100
    },
    {
      "epoch": 0.18027571580063625,
      "grad_norm": 0.3833279609680176,
      "learning_rate": 4.9906080491703146e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 2102160,
      "step": 1105
    },
    {
      "epoch": 0.18109144302145364,
      "grad_norm": 0.4162094295024872,
      "learning_rate": 4.990522838418797e-05,
      "loss": 0.3223,
      "num_input_tokens_seen": 2111808,
      "step": 1110
    },
    {
      "epoch": 0.181907170242271,
      "grad_norm": 1.3689614534378052,
      "learning_rate": 4.9904372435973604e-05,
      "loss": 0.2376,
      "num_input_tokens_seen": 2122704,
      "step": 1115
    },
    {
      "epoch": 0.18272289746308834,
      "grad_norm": 0.06482353806495667,
      "learning_rate": 4.990351264719203e-05,
      "loss": 0.0768,
      "num_input_tokens_seen": 2132128,
      "step": 1120
    },
    {
      "epoch": 0.1835386246839057,
      "grad_norm": 0.24577397108078003,
      "learning_rate": 4.990264901797586e-05,
      "loss": 0.1095,
      "num_input_tokens_seen": 2141584,
      "step": 1125
    },
    {
      "epoch": 0.18435435190472307,
      "grad_norm": 0.5546548962593079,
      "learning_rate": 4.990178154845826e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 2151168,
      "step": 1130
    },
    {
      "epoch": 0.18517007912554043,
      "grad_norm": 0.35332629084587097,
      "learning_rate": 4.9900910238773014e-05,
      "loss": 0.096,
      "num_input_tokens_seen": 2159744,
      "step": 1135
    },
    {
      "epoch": 0.18598580634635778,
      "grad_norm": 0.17641513049602509,
      "learning_rate": 4.990003508905448e-05,
      "loss": 0.1076,
      "num_input_tokens_seen": 2168464,
      "step": 1140
    },
    {
      "epoch": 0.18680153356717513,
      "grad_norm": 0.6305069327354431,
      "learning_rate": 4.989915609943763e-05,
      "loss": 0.0857,
      "num_input_tokens_seen": 2179104,
      "step": 1145
    },
    {
      "epoch": 0.18761726078799248,
      "grad_norm": 2.1355135440826416,
      "learning_rate": 4.9898273270058e-05,
      "loss": 0.1918,
      "num_input_tokens_seen": 2186992,
      "step": 1150
    },
    {
      "epoch": 0.18843298800880987,
      "grad_norm": 0.7163896560668945,
      "learning_rate": 4.989738660105174e-05,
      "loss": 0.1925,
      "num_input_tokens_seen": 2196080,
      "step": 1155
    },
    {
      "epoch": 0.18924871522962722,
      "grad_norm": 2.627490520477295,
      "learning_rate": 4.989649609255559e-05,
      "loss": 0.4059,
      "num_input_tokens_seen": 2204832,
      "step": 1160
    },
    {
      "epoch": 0.19006444245044457,
      "grad_norm": 0.6810788512229919,
      "learning_rate": 4.989560174470687e-05,
      "loss": 0.1699,
      "num_input_tokens_seen": 2214736,
      "step": 1165
    },
    {
      "epoch": 0.19088016967126192,
      "grad_norm": 2.392770767211914,
      "learning_rate": 4.989470355764351e-05,
      "loss": 0.3844,
      "num_input_tokens_seen": 2224800,
      "step": 1170
    },
    {
      "epoch": 0.19169589689207928,
      "grad_norm": 2.1730425357818604,
      "learning_rate": 4.9893801531504e-05,
      "loss": 0.2409,
      "num_input_tokens_seen": 2234848,
      "step": 1175
    },
    {
      "epoch": 0.19251162411289666,
      "grad_norm": 0.9319165945053101,
      "learning_rate": 4.9892895666427475e-05,
      "loss": 0.206,
      "num_input_tokens_seen": 2244544,
      "step": 1180
    },
    {
      "epoch": 0.193327351333714,
      "grad_norm": 0.3399196267127991,
      "learning_rate": 4.9891985962553606e-05,
      "loss": 0.1465,
      "num_input_tokens_seen": 2254480,
      "step": 1185
    },
    {
      "epoch": 0.19414307855453136,
      "grad_norm": 0.4017165005207062,
      "learning_rate": 4.989107242002269e-05,
      "loss": 0.2238,
      "num_input_tokens_seen": 2262656,
      "step": 1190
    },
    {
      "epoch": 0.19495880577534871,
      "grad_norm": 2.664108991622925,
      "learning_rate": 4.989015503897561e-05,
      "loss": 0.1086,
      "num_input_tokens_seen": 2271184,
      "step": 1195
    },
    {
      "epoch": 0.1957745329961661,
      "grad_norm": 0.5749203562736511,
      "learning_rate": 4.988923381955383e-05,
      "loss": 0.3025,
      "num_input_tokens_seen": 2279552,
      "step": 1200
    },
    {
      "epoch": 0.1957745329961661,
      "eval_loss": 0.19734390079975128,
      "eval_runtime": 133.1372,
      "eval_samples_per_second": 20.468,
      "eval_steps_per_second": 10.238,
      "num_input_tokens_seen": 2279552,
      "step": 1200
    },
    {
      "epoch": 0.19659026021698345,
      "grad_norm": 0.5350419878959656,
      "learning_rate": 4.988830876189942e-05,
      "loss": 0.2766,
      "num_input_tokens_seen": 2289280,
      "step": 1205
    },
    {
      "epoch": 0.1974059874378008,
      "grad_norm": 0.4800526201725006,
      "learning_rate": 4.988737986615503e-05,
      "loss": 0.1256,
      "num_input_tokens_seen": 2298480,
      "step": 1210
    },
    {
      "epoch": 0.19822171465861815,
      "grad_norm": 0.1763894110918045,
      "learning_rate": 4.988644713246391e-05,
      "loss": 0.0501,
      "num_input_tokens_seen": 2308784,
      "step": 1215
    },
    {
      "epoch": 0.1990374418794355,
      "grad_norm": 1.8767367601394653,
      "learning_rate": 4.988551056096991e-05,
      "loss": 0.2197,
      "num_input_tokens_seen": 2318640,
      "step": 1220
    },
    {
      "epoch": 0.1998531691002529,
      "grad_norm": 2.8569447994232178,
      "learning_rate": 4.988457015181743e-05,
      "loss": 0.192,
      "num_input_tokens_seen": 2327664,
      "step": 1225
    },
    {
      "epoch": 0.20066889632107024,
      "grad_norm": 2.0794010162353516,
      "learning_rate": 4.988362590515153e-05,
      "loss": 0.2771,
      "num_input_tokens_seen": 2338432,
      "step": 1230
    },
    {
      "epoch": 0.2014846235418876,
      "grad_norm": 1.1200827360153198,
      "learning_rate": 4.9882677821117805e-05,
      "loss": 0.2168,
      "num_input_tokens_seen": 2349072,
      "step": 1235
    },
    {
      "epoch": 0.20230035076270494,
      "grad_norm": 1.6671627759933472,
      "learning_rate": 4.988172589986246e-05,
      "loss": 0.3286,
      "num_input_tokens_seen": 2358576,
      "step": 1240
    },
    {
      "epoch": 0.2031160779835223,
      "grad_norm": 2.9104976654052734,
      "learning_rate": 4.9880770141532304e-05,
      "loss": 0.2699,
      "num_input_tokens_seen": 2367456,
      "step": 1245
    },
    {
      "epoch": 0.20393180520433968,
      "grad_norm": 2.7752978801727295,
      "learning_rate": 4.987981054627472e-05,
      "loss": 0.2669,
      "num_input_tokens_seen": 2377104,
      "step": 1250
    },
    {
      "epoch": 0.20474753242515703,
      "grad_norm": 2.289884567260742,
      "learning_rate": 4.987884711423769e-05,
      "loss": 0.4519,
      "num_input_tokens_seen": 2385824,
      "step": 1255
    },
    {
      "epoch": 0.20556325964597438,
      "grad_norm": 0.13279598951339722,
      "learning_rate": 4.9877879845569784e-05,
      "loss": 0.1212,
      "num_input_tokens_seen": 2396528,
      "step": 1260
    },
    {
      "epoch": 0.20637898686679174,
      "grad_norm": 1.1996725797653198,
      "learning_rate": 4.9876908740420175e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 2406288,
      "step": 1265
    },
    {
      "epoch": 0.20719471408760912,
      "grad_norm": 0.943985104560852,
      "learning_rate": 4.987593379893861e-05,
      "loss": 0.2125,
      "num_input_tokens_seen": 2415296,
      "step": 1270
    },
    {
      "epoch": 0.20801044130842647,
      "grad_norm": 0.9118590950965881,
      "learning_rate": 4.987495502127545e-05,
      "loss": 0.2085,
      "num_input_tokens_seen": 2425168,
      "step": 1275
    },
    {
      "epoch": 0.20882616852924382,
      "grad_norm": 0.6291457414627075,
      "learning_rate": 4.987397240758162e-05,
      "loss": 0.1204,
      "num_input_tokens_seen": 2435040,
      "step": 1280
    },
    {
      "epoch": 0.20964189575006117,
      "grad_norm": 2.6148977279663086,
      "learning_rate": 4.9872985958008664e-05,
      "loss": 0.195,
      "num_input_tokens_seen": 2444224,
      "step": 1285
    },
    {
      "epoch": 0.21045762297087853,
      "grad_norm": 0.06040278449654579,
      "learning_rate": 4.987199567270871e-05,
      "loss": 0.1778,
      "num_input_tokens_seen": 2453984,
      "step": 1290
    },
    {
      "epoch": 0.2112733501916959,
      "grad_norm": 0.06124170869588852,
      "learning_rate": 4.9871001551834444e-05,
      "loss": 0.0958,
      "num_input_tokens_seen": 2464336,
      "step": 1295
    },
    {
      "epoch": 0.21208907741251326,
      "grad_norm": 2.1283533573150635,
      "learning_rate": 4.98700035955392e-05,
      "loss": 0.2226,
      "num_input_tokens_seen": 2472688,
      "step": 1300
    },
    {
      "epoch": 0.2129048046333306,
      "grad_norm": 0.588334321975708,
      "learning_rate": 4.986900180397686e-05,
      "loss": 0.3194,
      "num_input_tokens_seen": 2483168,
      "step": 1305
    },
    {
      "epoch": 0.21372053185414797,
      "grad_norm": 0.23913580179214478,
      "learning_rate": 4.9867996177301926e-05,
      "loss": 0.0957,
      "num_input_tokens_seen": 2492752,
      "step": 1310
    },
    {
      "epoch": 0.21453625907496532,
      "grad_norm": 1.7628105878829956,
      "learning_rate": 4.9866986715669464e-05,
      "loss": 0.2053,
      "num_input_tokens_seen": 2501424,
      "step": 1315
    },
    {
      "epoch": 0.2153519862957827,
      "grad_norm": 0.022423215210437775,
      "learning_rate": 4.9865973419235155e-05,
      "loss": 0.1128,
      "num_input_tokens_seen": 2511648,
      "step": 1320
    },
    {
      "epoch": 0.21616771351660005,
      "grad_norm": 0.7439555525779724,
      "learning_rate": 4.986495628815526e-05,
      "loss": 0.2079,
      "num_input_tokens_seen": 2521552,
      "step": 1325
    },
    {
      "epoch": 0.2169834407374174,
      "grad_norm": 2.0982391834259033,
      "learning_rate": 4.986393532258663e-05,
      "loss": 0.3992,
      "num_input_tokens_seen": 2530816,
      "step": 1330
    },
    {
      "epoch": 0.21779916795823476,
      "grad_norm": 0.1741734743118286,
      "learning_rate": 4.986291052268671e-05,
      "loss": 0.146,
      "num_input_tokens_seen": 2540880,
      "step": 1335
    },
    {
      "epoch": 0.21861489517905214,
      "grad_norm": 1.13291597366333,
      "learning_rate": 4.986188188861355e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 2550160,
      "step": 1340
    },
    {
      "epoch": 0.2194306223998695,
      "grad_norm": 0.5688232183456421,
      "learning_rate": 4.9860849420525766e-05,
      "loss": 0.0953,
      "num_input_tokens_seen": 2560704,
      "step": 1345
    },
    {
      "epoch": 0.22024634962068684,
      "grad_norm": 1.1214573383331299,
      "learning_rate": 4.9859813118582575e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 2570592,
      "step": 1350
    },
    {
      "epoch": 0.2210620768415042,
      "grad_norm": 0.6352591514587402,
      "learning_rate": 4.98587729829438e-05,
      "loss": 0.2448,
      "num_input_tokens_seen": 2579472,
      "step": 1355
    },
    {
      "epoch": 0.22187780406232155,
      "grad_norm": 2.045254945755005,
      "learning_rate": 4.985772901376983e-05,
      "loss": 0.1788,
      "num_input_tokens_seen": 2590016,
      "step": 1360
    },
    {
      "epoch": 0.22269353128313893,
      "grad_norm": 1.104307770729065,
      "learning_rate": 4.9856681211221666e-05,
      "loss": 0.2547,
      "num_input_tokens_seen": 2599472,
      "step": 1365
    },
    {
      "epoch": 0.22350925850395628,
      "grad_norm": 1.5595825910568237,
      "learning_rate": 4.985562957546089e-05,
      "loss": 0.1877,
      "num_input_tokens_seen": 2609632,
      "step": 1370
    },
    {
      "epoch": 0.22432498572477363,
      "grad_norm": 0.8995135426521301,
      "learning_rate": 4.9854574106649686e-05,
      "loss": 0.1267,
      "num_input_tokens_seen": 2620560,
      "step": 1375
    },
    {
      "epoch": 0.225140712945591,
      "grad_norm": 1.9361729621887207,
      "learning_rate": 4.985351480495081e-05,
      "loss": 0.1973,
      "num_input_tokens_seen": 2630192,
      "step": 1380
    },
    {
      "epoch": 0.22595644016640834,
      "grad_norm": 0.42119917273521423,
      "learning_rate": 4.985245167052762e-05,
      "loss": 0.1853,
      "num_input_tokens_seen": 2639824,
      "step": 1385
    },
    {
      "epoch": 0.22677216738722572,
      "grad_norm": 0.19467391073703766,
      "learning_rate": 4.9851384703544066e-05,
      "loss": 0.4058,
      "num_input_tokens_seen": 2648992,
      "step": 1390
    },
    {
      "epoch": 0.22758789460804307,
      "grad_norm": 0.8588306307792664,
      "learning_rate": 4.985031390416469e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 2659296,
      "step": 1395
    },
    {
      "epoch": 0.22840362182886043,
      "grad_norm": 0.2162671834230423,
      "learning_rate": 4.984923927255461e-05,
      "loss": 0.2009,
      "num_input_tokens_seen": 2668256,
      "step": 1400
    },
    {
      "epoch": 0.22840362182886043,
      "eval_loss": 0.18615153431892395,
      "eval_runtime": 133.0657,
      "eval_samples_per_second": 20.479,
      "eval_steps_per_second": 10.243,
      "num_input_tokens_seen": 2668256,
      "step": 1400
    },
    {
      "epoch": 0.22921934904967778,
      "grad_norm": 0.17404496669769287,
      "learning_rate": 4.984816080887958e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 2679536,
      "step": 1405
    },
    {
      "epoch": 0.23003507627049516,
      "grad_norm": 2.455467462539673,
      "learning_rate": 4.9847078513305875e-05,
      "loss": 0.3279,
      "num_input_tokens_seen": 2688768,
      "step": 1410
    },
    {
      "epoch": 0.2308508034913125,
      "grad_norm": 1.3208752870559692,
      "learning_rate": 4.984599238600043e-05,
      "loss": 0.2042,
      "num_input_tokens_seen": 2697808,
      "step": 1415
    },
    {
      "epoch": 0.23166653071212986,
      "grad_norm": 0.25537386536598206,
      "learning_rate": 4.9844902427130716e-05,
      "loss": 0.0851,
      "num_input_tokens_seen": 2706848,
      "step": 1420
    },
    {
      "epoch": 0.23248225793294722,
      "grad_norm": 1.7820391654968262,
      "learning_rate": 4.984380863686482e-05,
      "loss": 0.3088,
      "num_input_tokens_seen": 2714816,
      "step": 1425
    },
    {
      "epoch": 0.23329798515376457,
      "grad_norm": 2.1848132610321045,
      "learning_rate": 4.984271101537143e-05,
      "loss": 0.1742,
      "num_input_tokens_seen": 2722512,
      "step": 1430
    },
    {
      "epoch": 0.23411371237458195,
      "grad_norm": 0.9686208367347717,
      "learning_rate": 4.9841609562819816e-05,
      "loss": 0.268,
      "num_input_tokens_seen": 2732192,
      "step": 1435
    },
    {
      "epoch": 0.2349294395953993,
      "grad_norm": 0.7435345649719238,
      "learning_rate": 4.984050427937983e-05,
      "loss": 0.0795,
      "num_input_tokens_seen": 2741856,
      "step": 1440
    },
    {
      "epoch": 0.23574516681621666,
      "grad_norm": 1.0544617176055908,
      "learning_rate": 4.983939516522191e-05,
      "loss": 0.1175,
      "num_input_tokens_seen": 2751104,
      "step": 1445
    },
    {
      "epoch": 0.236560894037034,
      "grad_norm": 1.0596301555633545,
      "learning_rate": 4.983828222051711e-05,
      "loss": 0.2653,
      "num_input_tokens_seen": 2759840,
      "step": 1450
    },
    {
      "epoch": 0.23737662125785136,
      "grad_norm": 0.6432933807373047,
      "learning_rate": 4.983716544543705e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 2769072,
      "step": 1455
    },
    {
      "epoch": 0.23819234847866874,
      "grad_norm": 0.9761828184127808,
      "learning_rate": 4.983604484015395e-05,
      "loss": 0.1078,
      "num_input_tokens_seen": 2778848,
      "step": 1460
    },
    {
      "epoch": 0.2390080756994861,
      "grad_norm": 0.4752540588378906,
      "learning_rate": 4.983492040484064e-05,
      "loss": 0.2027,
      "num_input_tokens_seen": 2789136,
      "step": 1465
    },
    {
      "epoch": 0.23982380292030345,
      "grad_norm": 1.6701676845550537,
      "learning_rate": 4.98337921396705e-05,
      "loss": 0.086,
      "num_input_tokens_seen": 2800352,
      "step": 1470
    },
    {
      "epoch": 0.2406395301411208,
      "grad_norm": 2.649320363998413,
      "learning_rate": 4.983266004481753e-05,
      "loss": 0.3556,
      "num_input_tokens_seen": 2810080,
      "step": 1475
    },
    {
      "epoch": 0.24145525736193818,
      "grad_norm": 0.7282776236534119,
      "learning_rate": 4.9831524120456316e-05,
      "loss": 0.27,
      "num_input_tokens_seen": 2819664,
      "step": 1480
    },
    {
      "epoch": 0.24227098458275553,
      "grad_norm": 1.244693636894226,
      "learning_rate": 4.9830384366762026e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 2828304,
      "step": 1485
    },
    {
      "epoch": 0.24308671180357289,
      "grad_norm": 0.5055590271949768,
      "learning_rate": 4.9829240783910436e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 2836992,
      "step": 1490
    },
    {
      "epoch": 0.24390243902439024,
      "grad_norm": 0.4892026484012604,
      "learning_rate": 4.982809337207789e-05,
      "loss": 0.2181,
      "num_input_tokens_seen": 2847584,
      "step": 1495
    },
    {
      "epoch": 0.2447181662452076,
      "grad_norm": 1.2856323719024658,
      "learning_rate": 4.9826942131441337e-05,
      "loss": 0.3004,
      "num_input_tokens_seen": 2856752,
      "step": 1500
    },
    {
      "epoch": 0.24553389346602497,
      "grad_norm": 1.2780262231826782,
      "learning_rate": 4.9825787062178315e-05,
      "loss": 0.1203,
      "num_input_tokens_seen": 2866656,
      "step": 1505
    },
    {
      "epoch": 0.24634962068684232,
      "grad_norm": 0.2715601623058319,
      "learning_rate": 4.9824628164466945e-05,
      "loss": 0.0694,
      "num_input_tokens_seen": 2875824,
      "step": 1510
    },
    {
      "epoch": 0.24716534790765968,
      "grad_norm": 0.5848760604858398,
      "learning_rate": 4.982346543848595e-05,
      "loss": 0.2108,
      "num_input_tokens_seen": 2885568,
      "step": 1515
    },
    {
      "epoch": 0.24798107512847703,
      "grad_norm": 0.16274535655975342,
      "learning_rate": 4.9822298884414626e-05,
      "loss": 0.2488,
      "num_input_tokens_seen": 2895536,
      "step": 1520
    },
    {
      "epoch": 0.24879680234929438,
      "grad_norm": 0.857880711555481,
      "learning_rate": 4.982112850243288e-05,
      "loss": 0.2658,
      "num_input_tokens_seen": 2905184,
      "step": 1525
    },
    {
      "epoch": 0.24961252957011176,
      "grad_norm": 0.45512938499450684,
      "learning_rate": 4.98199542927212e-05,
      "loss": 0.1172,
      "num_input_tokens_seen": 2913808,
      "step": 1530
    },
    {
      "epoch": 0.2504282567909291,
      "grad_norm": 0.9529207944869995,
      "learning_rate": 4.981877625546066e-05,
      "loss": 0.263,
      "num_input_tokens_seen": 2924064,
      "step": 1535
    },
    {
      "epoch": 0.25124398401174647,
      "grad_norm": 2.625666618347168,
      "learning_rate": 4.981759439083293e-05,
      "loss": 0.2024,
      "num_input_tokens_seen": 2934576,
      "step": 1540
    },
    {
      "epoch": 0.2520597112325638,
      "grad_norm": 0.8878980278968811,
      "learning_rate": 4.981640869902027e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 2944304,
      "step": 1545
    },
    {
      "epoch": 0.2528754384533812,
      "grad_norm": 1.2263116836547852,
      "learning_rate": 4.9815219180205517e-05,
      "loss": 0.1432,
      "num_input_tokens_seen": 2953216,
      "step": 1550
    },
    {
      "epoch": 0.2536911656741985,
      "grad_norm": 0.5461456775665283,
      "learning_rate": 4.9814025834572126e-05,
      "loss": 0.1987,
      "num_input_tokens_seen": 2962832,
      "step": 1555
    },
    {
      "epoch": 0.2545068928950159,
      "grad_norm": 2.1943440437316895,
      "learning_rate": 4.981282866230411e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 2971952,
      "step": 1560
    },
    {
      "epoch": 0.2553226201158333,
      "grad_norm": 1.69133722782135,
      "learning_rate": 4.981162766358611e-05,
      "loss": 0.1691,
      "num_input_tokens_seen": 2981328,
      "step": 1565
    },
    {
      "epoch": 0.25613834733665064,
      "grad_norm": 0.32694849371910095,
      "learning_rate": 4.9810422838603316e-05,
      "loss": 0.3704,
      "num_input_tokens_seen": 2990848,
      "step": 1570
    },
    {
      "epoch": 0.256954074557468,
      "grad_norm": 0.5244548320770264,
      "learning_rate": 4.9809214187541533e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 2998880,
      "step": 1575
    },
    {
      "epoch": 0.25776980177828535,
      "grad_norm": 0.2062338888645172,
      "learning_rate": 4.980800171058715e-05,
      "loss": 0.0792,
      "num_input_tokens_seen": 3006864,
      "step": 1580
    },
    {
      "epoch": 0.2585855289991027,
      "grad_norm": 0.552874743938446,
      "learning_rate": 4.980678540792715e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 3017408,
      "step": 1585
    },
    {
      "epoch": 0.25940125621992005,
      "grad_norm": 0.5734285116195679,
      "learning_rate": 4.980556527974909e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 3027968,
      "step": 1590
    },
    {
      "epoch": 0.2602169834407374,
      "grad_norm": 0.39380908012390137,
      "learning_rate": 4.980434132624114e-05,
      "loss": 0.0821,
      "num_input_tokens_seen": 3038480,
      "step": 1595
    },
    {
      "epoch": 0.26103271066155476,
      "grad_norm": 0.49615171551704407,
      "learning_rate": 4.980311354759205e-05,
      "loss": 0.1066,
      "num_input_tokens_seen": 3047328,
      "step": 1600
    },
    {
      "epoch": 0.26103271066155476,
      "eval_loss": 0.1832430362701416,
      "eval_runtime": 133.0644,
      "eval_samples_per_second": 20.479,
      "eval_steps_per_second": 10.243,
      "num_input_tokens_seen": 3047328,
      "step": 1600
    },
    {
      "epoch": 0.2618484378823721,
      "grad_norm": 0.3004530072212219,
      "learning_rate": 4.980188194399116e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 3057072,
      "step": 1605
    },
    {
      "epoch": 0.2626641651031895,
      "grad_norm": 0.7719806432723999,
      "learning_rate": 4.9800646515628384e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 3067920,
      "step": 1610
    },
    {
      "epoch": 0.26347989232400687,
      "grad_norm": 1.3854910135269165,
      "learning_rate": 4.979940726269426e-05,
      "loss": 0.2146,
      "num_input_tokens_seen": 3077808,
      "step": 1615
    },
    {
      "epoch": 0.2642956195448242,
      "grad_norm": 0.17119133472442627,
      "learning_rate": 4.979816418537988e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 3086832,
      "step": 1620
    },
    {
      "epoch": 0.2651113467656416,
      "grad_norm": 1.3540164232254028,
      "learning_rate": 4.979691728387696e-05,
      "loss": 0.219,
      "num_input_tokens_seen": 3096928,
      "step": 1625
    },
    {
      "epoch": 0.26592707398645893,
      "grad_norm": 0.04990815743803978,
      "learning_rate": 4.979566655837776e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 3107120,
      "step": 1630
    },
    {
      "epoch": 0.2667428012072763,
      "grad_norm": 1.487789511680603,
      "learning_rate": 4.9794412009075184e-05,
      "loss": 0.1901,
      "num_input_tokens_seen": 3115824,
      "step": 1635
    },
    {
      "epoch": 0.26755852842809363,
      "grad_norm": 1.640666127204895,
      "learning_rate": 4.979315363616269e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 3125824,
      "step": 1640
    },
    {
      "epoch": 0.268374255648911,
      "grad_norm": 0.5021424293518066,
      "learning_rate": 4.979189143983434e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 3135616,
      "step": 1645
    },
    {
      "epoch": 0.26918998286972834,
      "grad_norm": 0.35913169384002686,
      "learning_rate": 4.979062542028478e-05,
      "loss": 0.1125,
      "num_input_tokens_seen": 3146112,
      "step": 1650
    },
    {
      "epoch": 0.27000571009054575,
      "grad_norm": 2.8174521923065186,
      "learning_rate": 4.978935557770923e-05,
      "loss": 0.1927,
      "num_input_tokens_seen": 3155648,
      "step": 1655
    },
    {
      "epoch": 0.2708214373113631,
      "grad_norm": 0.41898030042648315,
      "learning_rate": 4.978808191230353e-05,
      "loss": 0.0793,
      "num_input_tokens_seen": 3165408,
      "step": 1660
    },
    {
      "epoch": 0.27163716453218045,
      "grad_norm": 0.3538326025009155,
      "learning_rate": 4.9786804424264085e-05,
      "loss": 0.0826,
      "num_input_tokens_seen": 3175088,
      "step": 1665
    },
    {
      "epoch": 0.2724528917529978,
      "grad_norm": 1.8629847764968872,
      "learning_rate": 4.978552311378792e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 3183984,
      "step": 1670
    },
    {
      "epoch": 0.27326861897381516,
      "grad_norm": 0.42656397819519043,
      "learning_rate": 4.978423798107261e-05,
      "loss": 0.0883,
      "num_input_tokens_seen": 3194048,
      "step": 1675
    },
    {
      "epoch": 0.2740843461946325,
      "grad_norm": 3.3663218021392822,
      "learning_rate": 4.978294902631635e-05,
      "loss": 0.2616,
      "num_input_tokens_seen": 3203840,
      "step": 1680
    },
    {
      "epoch": 0.27490007341544986,
      "grad_norm": 1.3997677564620972,
      "learning_rate": 4.9781656249717914e-05,
      "loss": 0.1903,
      "num_input_tokens_seen": 3213856,
      "step": 1685
    },
    {
      "epoch": 0.2757158006362672,
      "grad_norm": 3.7515499591827393,
      "learning_rate": 4.9780359651476645e-05,
      "loss": 0.375,
      "num_input_tokens_seen": 3222816,
      "step": 1690
    },
    {
      "epoch": 0.27653152785708457,
      "grad_norm": 1.195969820022583,
      "learning_rate": 4.977905923179251e-05,
      "loss": 0.1132,
      "num_input_tokens_seen": 3232832,
      "step": 1695
    },
    {
      "epoch": 0.2773472550779019,
      "grad_norm": 0.3568428158760071,
      "learning_rate": 4.977775499086606e-05,
      "loss": 0.296,
      "num_input_tokens_seen": 3242016,
      "step": 1700
    },
    {
      "epoch": 0.27816298229871933,
      "grad_norm": 0.4855261445045471,
      "learning_rate": 4.97764469288984e-05,
      "loss": 0.2193,
      "num_input_tokens_seen": 3251248,
      "step": 1705
    },
    {
      "epoch": 0.2789787095195367,
      "grad_norm": 0.7400346398353577,
      "learning_rate": 4.977513504609127e-05,
      "loss": 0.0637,
      "num_input_tokens_seen": 3261264,
      "step": 1710
    },
    {
      "epoch": 0.27979443674035404,
      "grad_norm": 0.36527255177497864,
      "learning_rate": 4.9773819342646965e-05,
      "loss": 0.0651,
      "num_input_tokens_seen": 3271088,
      "step": 1715
    },
    {
      "epoch": 0.2806101639611714,
      "grad_norm": 0.22418376803398132,
      "learning_rate": 4.97724998187684e-05,
      "loss": 0.2348,
      "num_input_tokens_seen": 3280176,
      "step": 1720
    },
    {
      "epoch": 0.28142589118198874,
      "grad_norm": 2.3049864768981934,
      "learning_rate": 4.9771176474659045e-05,
      "loss": 0.243,
      "num_input_tokens_seen": 3290048,
      "step": 1725
    },
    {
      "epoch": 0.2822416184028061,
      "grad_norm": 0.9105322957038879,
      "learning_rate": 4.976984931052299e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 3299504,
      "step": 1730
    },
    {
      "epoch": 0.28305734562362345,
      "grad_norm": 1.7574862241744995,
      "learning_rate": 4.976851832656489e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 3308560,
      "step": 1735
    },
    {
      "epoch": 0.2838730728444408,
      "grad_norm": 2.0970025062561035,
      "learning_rate": 4.9767183522990004e-05,
      "loss": 0.5413,
      "num_input_tokens_seen": 3316848,
      "step": 1740
    },
    {
      "epoch": 0.28468880006525815,
      "grad_norm": 1.5421417951583862,
      "learning_rate": 4.9765844900004176e-05,
      "loss": 0.1999,
      "num_input_tokens_seen": 3325536,
      "step": 1745
    },
    {
      "epoch": 0.28550452728607556,
      "grad_norm": 1.170609712600708,
      "learning_rate": 4.9764502457813834e-05,
      "loss": 0.2042,
      "num_input_tokens_seen": 3336176,
      "step": 1750
    },
    {
      "epoch": 0.2863202545068929,
      "grad_norm": 0.5363856554031372,
      "learning_rate": 4.9763156196626005e-05,
      "loss": 0.0556,
      "num_input_tokens_seen": 3345968,
      "step": 1755
    },
    {
      "epoch": 0.28713598172771027,
      "grad_norm": 1.042510747909546,
      "learning_rate": 4.97618061166483e-05,
      "loss": 0.1046,
      "num_input_tokens_seen": 3354640,
      "step": 1760
    },
    {
      "epoch": 0.2879517089485276,
      "grad_norm": 0.8330152034759521,
      "learning_rate": 4.9760452218088915e-05,
      "loss": 0.0768,
      "num_input_tokens_seen": 3363936,
      "step": 1765
    },
    {
      "epoch": 0.28876743616934497,
      "grad_norm": 1.4867428541183472,
      "learning_rate": 4.975909450115663e-05,
      "loss": 0.4947,
      "num_input_tokens_seen": 3373168,
      "step": 1770
    },
    {
      "epoch": 0.2895831633901623,
      "grad_norm": 0.057241011410951614,
      "learning_rate": 4.975773296606084e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 3383232,
      "step": 1775
    },
    {
      "epoch": 0.2903988906109797,
      "grad_norm": 0.3030959963798523,
      "learning_rate": 4.97563676130115e-05,
      "loss": 0.1364,
      "num_input_tokens_seen": 3393328,
      "step": 1780
    },
    {
      "epoch": 0.29121461783179703,
      "grad_norm": 1.0506988763809204,
      "learning_rate": 4.9754998442219166e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 3402192,
      "step": 1785
    },
    {
      "epoch": 0.2920303450526144,
      "grad_norm": 0.07536742091178894,
      "learning_rate": 4.9753625453894984e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 3410928,
      "step": 1790
    },
    {
      "epoch": 0.2928460722734318,
      "grad_norm": 1.0747013092041016,
      "learning_rate": 4.975224864825068e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 3420160,
      "step": 1795
    },
    {
      "epoch": 0.29366179949424914,
      "grad_norm": 0.7137489318847656,
      "learning_rate": 4.9750868025498576e-05,
      "loss": 0.0954,
      "num_input_tokens_seen": 3429984,
      "step": 1800
    },
    {
      "epoch": 0.29366179949424914,
      "eval_loss": 0.1835136115550995,
      "eval_runtime": 133.2743,
      "eval_samples_per_second": 20.447,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 3429984,
      "step": 1800
    },
    {
      "epoch": 0.2944775267150665,
      "grad_norm": 0.9617941379547119,
      "learning_rate": 4.974948358585158e-05,
      "loss": 0.0686,
      "num_input_tokens_seen": 3440304,
      "step": 1805
    },
    {
      "epoch": 0.29529325393588385,
      "grad_norm": 1.782692790031433,
      "learning_rate": 4.9748095329523205e-05,
      "loss": 0.1799,
      "num_input_tokens_seen": 3449136,
      "step": 1810
    },
    {
      "epoch": 0.2961089811567012,
      "grad_norm": 0.9914631247520447,
      "learning_rate": 4.974670325672752e-05,
      "loss": 0.1822,
      "num_input_tokens_seen": 3459568,
      "step": 1815
    },
    {
      "epoch": 0.29692470837751855,
      "grad_norm": 2.0792791843414307,
      "learning_rate": 4.974530736767921e-05,
      "loss": 0.2421,
      "num_input_tokens_seen": 3466976,
      "step": 1820
    },
    {
      "epoch": 0.2977404355983359,
      "grad_norm": 0.7215911149978638,
      "learning_rate": 4.9743907662593524e-05,
      "loss": 0.146,
      "num_input_tokens_seen": 3477360,
      "step": 1825
    },
    {
      "epoch": 0.29855616281915326,
      "grad_norm": 0.22244583070278168,
      "learning_rate": 4.974250414168633e-05,
      "loss": 0.2837,
      "num_input_tokens_seen": 3488048,
      "step": 1830
    },
    {
      "epoch": 0.2993718900399706,
      "grad_norm": 0.1749742031097412,
      "learning_rate": 4.974109680517407e-05,
      "loss": 0.2005,
      "num_input_tokens_seen": 3497120,
      "step": 1835
    },
    {
      "epoch": 0.300187617260788,
      "grad_norm": 2.327420949935913,
      "learning_rate": 4.973968565327376e-05,
      "loss": 0.367,
      "num_input_tokens_seen": 3506656,
      "step": 1840
    },
    {
      "epoch": 0.3010033444816054,
      "grad_norm": 0.8419979214668274,
      "learning_rate": 4.973827068620303e-05,
      "loss": 0.1843,
      "num_input_tokens_seen": 3515024,
      "step": 1845
    },
    {
      "epoch": 0.3018190717024227,
      "grad_norm": 0.7486478090286255,
      "learning_rate": 4.973685190418008e-05,
      "loss": 0.3151,
      "num_input_tokens_seen": 3524304,
      "step": 1850
    },
    {
      "epoch": 0.3026347989232401,
      "grad_norm": 0.8558363318443298,
      "learning_rate": 4.97354293074237e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 3533280,
      "step": 1855
    },
    {
      "epoch": 0.30345052614405743,
      "grad_norm": 0.5546383261680603,
      "learning_rate": 4.9734002896153276e-05,
      "loss": 0.0791,
      "num_input_tokens_seen": 3541984,
      "step": 1860
    },
    {
      "epoch": 0.3042662533648748,
      "grad_norm": 0.7137537002563477,
      "learning_rate": 4.973257267058877e-05,
      "loss": 0.234,
      "num_input_tokens_seen": 3551168,
      "step": 1865
    },
    {
      "epoch": 0.30508198058569214,
      "grad_norm": 0.7604904174804688,
      "learning_rate": 4.973113863095076e-05,
      "loss": 0.2723,
      "num_input_tokens_seen": 3560880,
      "step": 1870
    },
    {
      "epoch": 0.3058977078065095,
      "grad_norm": 0.26781779527664185,
      "learning_rate": 4.9729700777460384e-05,
      "loss": 0.1238,
      "num_input_tokens_seen": 3570704,
      "step": 1875
    },
    {
      "epoch": 0.30671343502732684,
      "grad_norm": 2.2202072143554688,
      "learning_rate": 4.972825911033937e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 3580608,
      "step": 1880
    },
    {
      "epoch": 0.3075291622481442,
      "grad_norm": 0.2502926290035248,
      "learning_rate": 4.9726813629810056e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 3589808,
      "step": 1885
    },
    {
      "epoch": 0.3083448894689616,
      "grad_norm": 1.0356255769729614,
      "learning_rate": 4.9725364336095326e-05,
      "loss": 0.0844,
      "num_input_tokens_seen": 3599392,
      "step": 1890
    },
    {
      "epoch": 0.30916061668977896,
      "grad_norm": 0.4660522937774658,
      "learning_rate": 4.972391122941871e-05,
      "loss": 0.166,
      "num_input_tokens_seen": 3608208,
      "step": 1895
    },
    {
      "epoch": 0.3099763439105963,
      "grad_norm": 1.4490690231323242,
      "learning_rate": 4.972245431000428e-05,
      "loss": 0.2681,
      "num_input_tokens_seen": 3617920,
      "step": 1900
    },
    {
      "epoch": 0.31079207113141366,
      "grad_norm": 0.30918022990226746,
      "learning_rate": 4.972099357807671e-05,
      "loss": 0.1779,
      "num_input_tokens_seen": 3628688,
      "step": 1905
    },
    {
      "epoch": 0.311607798352231,
      "grad_norm": 3.0635056495666504,
      "learning_rate": 4.971952903386127e-05,
      "loss": 0.1166,
      "num_input_tokens_seen": 3638784,
      "step": 1910
    },
    {
      "epoch": 0.31242352557304837,
      "grad_norm": 0.7872055172920227,
      "learning_rate": 4.971806067758381e-05,
      "loss": 0.1704,
      "num_input_tokens_seen": 3648192,
      "step": 1915
    },
    {
      "epoch": 0.3132392527938657,
      "grad_norm": 0.8401194214820862,
      "learning_rate": 4.971658850947076e-05,
      "loss": 0.365,
      "num_input_tokens_seen": 3657936,
      "step": 1920
    },
    {
      "epoch": 0.31405498001468307,
      "grad_norm": 3.0678842067718506,
      "learning_rate": 4.9715112529749165e-05,
      "loss": 0.2802,
      "num_input_tokens_seen": 3668384,
      "step": 1925
    },
    {
      "epoch": 0.3148707072355004,
      "grad_norm": 1.1222971677780151,
      "learning_rate": 4.9713632738646624e-05,
      "loss": 0.1183,
      "num_input_tokens_seen": 3678480,
      "step": 1930
    },
    {
      "epoch": 0.31568643445631783,
      "grad_norm": 0.911420464515686,
      "learning_rate": 4.971214913639134e-05,
      "loss": 0.1062,
      "num_input_tokens_seen": 3688624,
      "step": 1935
    },
    {
      "epoch": 0.3165021616771352,
      "grad_norm": 0.316581666469574,
      "learning_rate": 4.9710661723212104e-05,
      "loss": 0.1327,
      "num_input_tokens_seen": 3699184,
      "step": 1940
    },
    {
      "epoch": 0.31731788889795254,
      "grad_norm": 0.6569178700447083,
      "learning_rate": 4.9709170499338295e-05,
      "loss": 0.1404,
      "num_input_tokens_seen": 3707760,
      "step": 1945
    },
    {
      "epoch": 0.3181336161187699,
      "grad_norm": 0.9238899946212769,
      "learning_rate": 4.9707675464999895e-05,
      "loss": 0.2611,
      "num_input_tokens_seen": 3717360,
      "step": 1950
    },
    {
      "epoch": 0.31894934333958724,
      "grad_norm": 0.7648108601570129,
      "learning_rate": 4.970617662042743e-05,
      "loss": 0.1095,
      "num_input_tokens_seen": 3727680,
      "step": 1955
    },
    {
      "epoch": 0.3197650705604046,
      "grad_norm": 0.03765552118420601,
      "learning_rate": 4.970467396585206e-05,
      "loss": 0.094,
      "num_input_tokens_seen": 3736720,
      "step": 1960
    },
    {
      "epoch": 0.32058079778122195,
      "grad_norm": 0.6361733675003052,
      "learning_rate": 4.97031675015055e-05,
      "loss": 0.4073,
      "num_input_tokens_seen": 3745632,
      "step": 1965
    },
    {
      "epoch": 0.3213965250020393,
      "grad_norm": 0.4498339295387268,
      "learning_rate": 4.9701657227620075e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 3755184,
      "step": 1970
    },
    {
      "epoch": 0.32221225222285665,
      "grad_norm": 0.785466194152832,
      "learning_rate": 4.9700143144428685e-05,
      "loss": 0.0977,
      "num_input_tokens_seen": 3764784,
      "step": 1975
    },
    {
      "epoch": 0.32302797944367406,
      "grad_norm": 0.2200617492198944,
      "learning_rate": 4.969862525216482e-05,
      "loss": 0.1226,
      "num_input_tokens_seen": 3774256,
      "step": 1980
    },
    {
      "epoch": 0.3238437066644914,
      "grad_norm": 1.9601080417633057,
      "learning_rate": 4.9697103551062556e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 3783552,
      "step": 1985
    },
    {
      "epoch": 0.32465943388530877,
      "grad_norm": 1.4570412635803223,
      "learning_rate": 4.9695578041356565e-05,
      "loss": 0.256,
      "num_input_tokens_seen": 3793216,
      "step": 1990
    },
    {
      "epoch": 0.3254751611061261,
      "grad_norm": 0.2757014036178589,
      "learning_rate": 4.969404872328209e-05,
      "loss": 0.1245,
      "num_input_tokens_seen": 3803456,
      "step": 1995
    },
    {
      "epoch": 0.3262908883269435,
      "grad_norm": 0.511043131351471,
      "learning_rate": 4.969251559707498e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 3814576,
      "step": 2000
    },
    {
      "epoch": 0.3262908883269435,
      "eval_loss": 0.17937684059143066,
      "eval_runtime": 133.1951,
      "eval_samples_per_second": 20.459,
      "eval_steps_per_second": 10.233,
      "num_input_tokens_seen": 3814576,
      "step": 2000
    },
    {
      "epoch": 0.3271066155477608,
      "grad_norm": 0.2921624183654785,
      "learning_rate": 4.9690978662971674e-05,
      "loss": 0.2455,
      "num_input_tokens_seen": 3824384,
      "step": 2005
    },
    {
      "epoch": 0.3279223427685782,
      "grad_norm": 0.995122492313385,
      "learning_rate": 4.968943792120916e-05,
      "loss": 0.2009,
      "num_input_tokens_seen": 3834048,
      "step": 2010
    },
    {
      "epoch": 0.32873806998939553,
      "grad_norm": 3.7687785625457764,
      "learning_rate": 4.9687893372025046e-05,
      "loss": 0.1952,
      "num_input_tokens_seen": 3844048,
      "step": 2015
    },
    {
      "epoch": 0.3295537972102129,
      "grad_norm": 0.25339728593826294,
      "learning_rate": 4.9686345015657535e-05,
      "loss": 0.3267,
      "num_input_tokens_seen": 3854544,
      "step": 2020
    },
    {
      "epoch": 0.33036952443103024,
      "grad_norm": 1.8406693935394287,
      "learning_rate": 4.968479285234538e-05,
      "loss": 0.1172,
      "num_input_tokens_seen": 3864240,
      "step": 2025
    },
    {
      "epoch": 0.33118525165184765,
      "grad_norm": 2.015005111694336,
      "learning_rate": 4.9683236882327974e-05,
      "loss": 0.1865,
      "num_input_tokens_seen": 3873776,
      "step": 2030
    },
    {
      "epoch": 0.332000978872665,
      "grad_norm": 0.42686721682548523,
      "learning_rate": 4.968167710584526e-05,
      "loss": 0.2139,
      "num_input_tokens_seen": 3882560,
      "step": 2035
    },
    {
      "epoch": 0.33281670609348235,
      "grad_norm": 0.16534170508384705,
      "learning_rate": 4.968011352313775e-05,
      "loss": 0.0889,
      "num_input_tokens_seen": 3892144,
      "step": 2040
    },
    {
      "epoch": 0.3336324333142997,
      "grad_norm": 0.5639033317565918,
      "learning_rate": 4.967854613444659e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 3902000,
      "step": 2045
    },
    {
      "epoch": 0.33444816053511706,
      "grad_norm": 1.5559618473052979,
      "learning_rate": 4.967697494001349e-05,
      "loss": 0.2768,
      "num_input_tokens_seen": 3910976,
      "step": 2050
    },
    {
      "epoch": 0.3352638877559344,
      "grad_norm": 1.3341150283813477,
      "learning_rate": 4.9675399940080736e-05,
      "loss": 0.2252,
      "num_input_tokens_seen": 3920608,
      "step": 2055
    },
    {
      "epoch": 0.33607961497675176,
      "grad_norm": 1.0763622522354126,
      "learning_rate": 4.9673821134891226e-05,
      "loss": 0.0891,
      "num_input_tokens_seen": 3929408,
      "step": 2060
    },
    {
      "epoch": 0.3368953421975691,
      "grad_norm": 0.7243898510932922,
      "learning_rate": 4.967223852468842e-05,
      "loss": 0.0797,
      "num_input_tokens_seen": 3938928,
      "step": 2065
    },
    {
      "epoch": 0.33771106941838647,
      "grad_norm": 2.63608717918396,
      "learning_rate": 4.967065210971639e-05,
      "loss": 0.1183,
      "num_input_tokens_seen": 3949248,
      "step": 2070
    },
    {
      "epoch": 0.3385267966392039,
      "grad_norm": 1.9440754652023315,
      "learning_rate": 4.966906189021977e-05,
      "loss": 0.3067,
      "num_input_tokens_seen": 3959088,
      "step": 2075
    },
    {
      "epoch": 0.33934252386002123,
      "grad_norm": 2.053971529006958,
      "learning_rate": 4.966746786644379e-05,
      "loss": 0.368,
      "num_input_tokens_seen": 3967456,
      "step": 2080
    },
    {
      "epoch": 0.3401582510808386,
      "grad_norm": 0.1921541690826416,
      "learning_rate": 4.966587003863429e-05,
      "loss": 0.0906,
      "num_input_tokens_seen": 3976544,
      "step": 2085
    },
    {
      "epoch": 0.34097397830165593,
      "grad_norm": 1.4287446737289429,
      "learning_rate": 4.966426840703765e-05,
      "loss": 0.2275,
      "num_input_tokens_seen": 3986992,
      "step": 2090
    },
    {
      "epoch": 0.3417897055224733,
      "grad_norm": 0.46612975001335144,
      "learning_rate": 4.9662662971900875e-05,
      "loss": 0.0798,
      "num_input_tokens_seen": 3996432,
      "step": 2095
    },
    {
      "epoch": 0.34260543274329064,
      "grad_norm": 0.31626221537590027,
      "learning_rate": 4.9661053733471534e-05,
      "loss": 0.1993,
      "num_input_tokens_seen": 4005472,
      "step": 2100
    },
    {
      "epoch": 0.343421159964108,
      "grad_norm": 1.940926194190979,
      "learning_rate": 4.965944069199781e-05,
      "loss": 0.2901,
      "num_input_tokens_seen": 4015456,
      "step": 2105
    },
    {
      "epoch": 0.34423688718492534,
      "grad_norm": 0.6949278712272644,
      "learning_rate": 4.965782384772842e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 4024336,
      "step": 2110
    },
    {
      "epoch": 0.3450526144057427,
      "grad_norm": 0.6315200328826904,
      "learning_rate": 4.9656203200912734e-05,
      "loss": 0.1144,
      "num_input_tokens_seen": 4033920,
      "step": 2115
    },
    {
      "epoch": 0.3458683416265601,
      "grad_norm": 0.26436859369277954,
      "learning_rate": 4.965457875180067e-05,
      "loss": 0.0634,
      "num_input_tokens_seen": 4043984,
      "step": 2120
    },
    {
      "epoch": 0.34668406884737746,
      "grad_norm": 1.5094294548034668,
      "learning_rate": 4.9652950500642724e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 4053680,
      "step": 2125
    },
    {
      "epoch": 0.3474997960681948,
      "grad_norm": 0.11046683043241501,
      "learning_rate": 4.965131844769001e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 4062864,
      "step": 2130
    },
    {
      "epoch": 0.34831552328901216,
      "grad_norm": 2.4206724166870117,
      "learning_rate": 4.96496825931942e-05,
      "loss": 0.3682,
      "num_input_tokens_seen": 4071984,
      "step": 2135
    },
    {
      "epoch": 0.3491312505098295,
      "grad_norm": 0.6027148365974426,
      "learning_rate": 4.9648042937407566e-05,
      "loss": 0.0867,
      "num_input_tokens_seen": 4080528,
      "step": 2140
    },
    {
      "epoch": 0.34994697773064687,
      "grad_norm": 0.16424928605556488,
      "learning_rate": 4.964639948058297e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 4089056,
      "step": 2145
    },
    {
      "epoch": 0.3507627049514642,
      "grad_norm": 0.6969748139381409,
      "learning_rate": 4.9644752222973846e-05,
      "loss": 0.2027,
      "num_input_tokens_seen": 4098544,
      "step": 2150
    },
    {
      "epoch": 0.3515784321722816,
      "grad_norm": 0.49607938528060913,
      "learning_rate": 4.964310116483422e-05,
      "loss": 0.1946,
      "num_input_tokens_seen": 4108544,
      "step": 2155
    },
    {
      "epoch": 0.3523941593930989,
      "grad_norm": 0.7357487082481384,
      "learning_rate": 4.964144630641872e-05,
      "loss": 0.2974,
      "num_input_tokens_seen": 4117328,
      "step": 2160
    },
    {
      "epoch": 0.3532098866139163,
      "grad_norm": 0.8226310014724731,
      "learning_rate": 4.9639787647982525e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 4125904,
      "step": 2165
    },
    {
      "epoch": 0.3540256138347337,
      "grad_norm": 0.25057363510131836,
      "learning_rate": 4.963812518978143e-05,
      "loss": 0.0985,
      "num_input_tokens_seen": 4136336,
      "step": 2170
    },
    {
      "epoch": 0.35484134105555104,
      "grad_norm": 0.8438670635223389,
      "learning_rate": 4.963645893207182e-05,
      "loss": 0.1891,
      "num_input_tokens_seen": 4145360,
      "step": 2175
    },
    {
      "epoch": 0.3556570682763684,
      "grad_norm": 0.43560925126075745,
      "learning_rate": 4.963478887511063e-05,
      "loss": 0.246,
      "num_input_tokens_seen": 4154144,
      "step": 2180
    },
    {
      "epoch": 0.35647279549718575,
      "grad_norm": 0.4939209520816803,
      "learning_rate": 4.963311501915542e-05,
      "loss": 0.07,
      "num_input_tokens_seen": 4162880,
      "step": 2185
    },
    {
      "epoch": 0.3572885227180031,
      "grad_norm": 1.442290186882019,
      "learning_rate": 4.963143736446432e-05,
      "loss": 0.378,
      "num_input_tokens_seen": 4171488,
      "step": 2190
    },
    {
      "epoch": 0.35810424993882045,
      "grad_norm": 0.7450513243675232,
      "learning_rate": 4.962975591129603e-05,
      "loss": 0.0736,
      "num_input_tokens_seen": 4181072,
      "step": 2195
    },
    {
      "epoch": 0.3589199771596378,
      "grad_norm": 0.39639630913734436,
      "learning_rate": 4.962807065990986e-05,
      "loss": 0.192,
      "num_input_tokens_seen": 4190352,
      "step": 2200
    },
    {
      "epoch": 0.3589199771596378,
      "eval_loss": 0.18029749393463135,
      "eval_runtime": 133.2248,
      "eval_samples_per_second": 20.454,
      "eval_steps_per_second": 10.231,
      "num_input_tokens_seen": 4190352,
      "step": 2200
    },
    {
      "epoch": 0.35973570438045516,
      "grad_norm": 2.520526170730591,
      "learning_rate": 4.9626381610565714e-05,
      "loss": 0.2461,
      "num_input_tokens_seen": 4199200,
      "step": 2205
    },
    {
      "epoch": 0.3605514316012725,
      "grad_norm": 0.22166985273361206,
      "learning_rate": 4.9624688763524043e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 4209472,
      "step": 2210
    },
    {
      "epoch": 0.3613671588220899,
      "grad_norm": 1.9402036666870117,
      "learning_rate": 4.962299211904591e-05,
      "loss": 0.1921,
      "num_input_tokens_seen": 4218880,
      "step": 2215
    },
    {
      "epoch": 0.36218288604290727,
      "grad_norm": 2.921316385269165,
      "learning_rate": 4.962129167739296e-05,
      "loss": 0.3698,
      "num_input_tokens_seen": 4229008,
      "step": 2220
    },
    {
      "epoch": 0.3629986132637246,
      "grad_norm": 0.3046807646751404,
      "learning_rate": 4.961958743882742e-05,
      "loss": 0.3656,
      "num_input_tokens_seen": 4238640,
      "step": 2225
    },
    {
      "epoch": 0.363814340484542,
      "grad_norm": 0.5515411496162415,
      "learning_rate": 4.961787940361211e-05,
      "loss": 0.1961,
      "num_input_tokens_seen": 4248400,
      "step": 2230
    },
    {
      "epoch": 0.36463006770535933,
      "grad_norm": 0.9371939301490784,
      "learning_rate": 4.961616757201043e-05,
      "loss": 0.12,
      "num_input_tokens_seen": 4258480,
      "step": 2235
    },
    {
      "epoch": 0.3654457949261767,
      "grad_norm": 0.2411399483680725,
      "learning_rate": 4.961445194428637e-05,
      "loss": 0.0909,
      "num_input_tokens_seen": 4267712,
      "step": 2240
    },
    {
      "epoch": 0.36626152214699403,
      "grad_norm": 2.141577959060669,
      "learning_rate": 4.9612732520704486e-05,
      "loss": 0.2796,
      "num_input_tokens_seen": 4277952,
      "step": 2245
    },
    {
      "epoch": 0.3670772493678114,
      "grad_norm": 0.5420566201210022,
      "learning_rate": 4.961100930152994e-05,
      "loss": 0.3852,
      "num_input_tokens_seen": 4287072,
      "step": 2250
    },
    {
      "epoch": 0.36789297658862874,
      "grad_norm": 0.570094108581543,
      "learning_rate": 4.960928228702849e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 4297088,
      "step": 2255
    },
    {
      "epoch": 0.36870870380944615,
      "grad_norm": 0.8473966121673584,
      "learning_rate": 4.960755147746645e-05,
      "loss": 0.2406,
      "num_input_tokens_seen": 4304000,
      "step": 2260
    },
    {
      "epoch": 0.3695244310302635,
      "grad_norm": 0.2298835664987564,
      "learning_rate": 4.9605816873110736e-05,
      "loss": 0.1531,
      "num_input_tokens_seen": 4313344,
      "step": 2265
    },
    {
      "epoch": 0.37034015825108085,
      "grad_norm": 0.6671218276023865,
      "learning_rate": 4.960407847422883e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 4322000,
      "step": 2270
    },
    {
      "epoch": 0.3711558854718982,
      "grad_norm": 0.7266539931297302,
      "learning_rate": 4.960233628108885e-05,
      "loss": 0.117,
      "num_input_tokens_seen": 4332064,
      "step": 2275
    },
    {
      "epoch": 0.37197161269271556,
      "grad_norm": 0.22930797934532166,
      "learning_rate": 4.960059029395942e-05,
      "loss": 0.1525,
      "num_input_tokens_seen": 4341520,
      "step": 2280
    },
    {
      "epoch": 0.3727873399135329,
      "grad_norm": 1.375317096710205,
      "learning_rate": 4.959884051310983e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 4351856,
      "step": 2285
    },
    {
      "epoch": 0.37360306713435026,
      "grad_norm": 0.30471429228782654,
      "learning_rate": 4.959708693880991e-05,
      "loss": 0.2808,
      "num_input_tokens_seen": 4360528,
      "step": 2290
    },
    {
      "epoch": 0.3744187943551676,
      "grad_norm": 0.8434260487556458,
      "learning_rate": 4.9595329571330074e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 4368992,
      "step": 2295
    },
    {
      "epoch": 0.37523452157598497,
      "grad_norm": 0.6489983201026917,
      "learning_rate": 4.9593568410941326e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 4378192,
      "step": 2300
    },
    {
      "epoch": 0.3760502487968023,
      "grad_norm": 0.04139533266425133,
      "learning_rate": 4.959180345791528e-05,
      "loss": 0.1046,
      "num_input_tokens_seen": 4387328,
      "step": 2305
    },
    {
      "epoch": 0.37686597601761973,
      "grad_norm": 0.3170318007469177,
      "learning_rate": 4.9590034712524086e-05,
      "loss": 0.07,
      "num_input_tokens_seen": 4397120,
      "step": 2310
    },
    {
      "epoch": 0.3776817032384371,
      "grad_norm": 0.8678091168403625,
      "learning_rate": 4.958826217504053e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 4406832,
      "step": 2315
    },
    {
      "epoch": 0.37849743045925444,
      "grad_norm": 0.5145722031593323,
      "learning_rate": 4.958648584573795e-05,
      "loss": 0.0873,
      "num_input_tokens_seen": 4416288,
      "step": 2320
    },
    {
      "epoch": 0.3793131576800718,
      "grad_norm": 1.0412065982818604,
      "learning_rate": 4.958470572489028e-05,
      "loss": 0.2128,
      "num_input_tokens_seen": 4425344,
      "step": 2325
    },
    {
      "epoch": 0.38012888490088914,
      "grad_norm": 0.32064706087112427,
      "learning_rate": 4.958292181277203e-05,
      "loss": 0.1983,
      "num_input_tokens_seen": 4434992,
      "step": 2330
    },
    {
      "epoch": 0.3809446121217065,
      "grad_norm": 0.37774601578712463,
      "learning_rate": 4.958113410965832e-05,
      "loss": 0.1965,
      "num_input_tokens_seen": 4444384,
      "step": 2335
    },
    {
      "epoch": 0.38176033934252385,
      "grad_norm": 0.4508562386035919,
      "learning_rate": 4.957934261582481e-05,
      "loss": 0.201,
      "num_input_tokens_seen": 4455584,
      "step": 2340
    },
    {
      "epoch": 0.3825760665633412,
      "grad_norm": 0.7914713025093079,
      "learning_rate": 4.95775473315478e-05,
      "loss": 0.3964,
      "num_input_tokens_seen": 4465168,
      "step": 2345
    },
    {
      "epoch": 0.38339179378415855,
      "grad_norm": 0.6675577163696289,
      "learning_rate": 4.9575748257104124e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 4474464,
      "step": 2350
    },
    {
      "epoch": 0.38420752100497596,
      "grad_norm": 2.632662773132324,
      "learning_rate": 4.9573945392771224e-05,
      "loss": 0.2488,
      "num_input_tokens_seen": 4483424,
      "step": 2355
    },
    {
      "epoch": 0.3850232482257933,
      "grad_norm": 0.9043718576431274,
      "learning_rate": 4.9572138738827134e-05,
      "loss": 0.1668,
      "num_input_tokens_seen": 4492144,
      "step": 2360
    },
    {
      "epoch": 0.38583897544661067,
      "grad_norm": 0.5289804339408875,
      "learning_rate": 4.957032829555046e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 4500624,
      "step": 2365
    },
    {
      "epoch": 0.386654702667428,
      "grad_norm": 1.464279294013977,
      "learning_rate": 4.956851406322039e-05,
      "loss": 0.2833,
      "num_input_tokens_seen": 4509920,
      "step": 2370
    },
    {
      "epoch": 0.38747042988824537,
      "grad_norm": 0.09804787486791611,
      "learning_rate": 4.9566696042116704e-05,
      "loss": 0.1932,
      "num_input_tokens_seen": 4519440,
      "step": 2375
    },
    {
      "epoch": 0.3882861571090627,
      "grad_norm": 0.7891910076141357,
      "learning_rate": 4.9564874232519766e-05,
      "loss": 0.1021,
      "num_input_tokens_seen": 4529472,
      "step": 2380
    },
    {
      "epoch": 0.3891018843298801,
      "grad_norm": 2.3793752193450928,
      "learning_rate": 4.9563048634710516e-05,
      "loss": 0.3199,
      "num_input_tokens_seen": 4539104,
      "step": 2385
    },
    {
      "epoch": 0.38991761155069743,
      "grad_norm": 1.0403443574905396,
      "learning_rate": 4.956121924897049e-05,
      "loss": 0.2744,
      "num_input_tokens_seen": 4548592,
      "step": 2390
    },
    {
      "epoch": 0.3907333387715148,
      "grad_norm": 0.29414650797843933,
      "learning_rate": 4.955938607558181e-05,
      "loss": 0.08,
      "num_input_tokens_seen": 4558096,
      "step": 2395
    },
    {
      "epoch": 0.3915490659923322,
      "grad_norm": 0.33984729647636414,
      "learning_rate": 4.955754911482715e-05,
      "loss": 0.1108,
      "num_input_tokens_seen": 4567440,
      "step": 2400
    },
    {
      "epoch": 0.3915490659923322,
      "eval_loss": 0.17265379428863525,
      "eval_runtime": 133.2202,
      "eval_samples_per_second": 20.455,
      "eval_steps_per_second": 10.231,
      "num_input_tokens_seen": 4567440,
      "step": 2400
    },
    {
      "epoch": 0.39236479321314954,
      "grad_norm": 1.9674694538116455,
      "learning_rate": 4.9555708366989804e-05,
      "loss": 0.1986,
      "num_input_tokens_seen": 4576832,
      "step": 2405
    },
    {
      "epoch": 0.3931805204339669,
      "grad_norm": 1.901394248008728,
      "learning_rate": 4.9553863832353655e-05,
      "loss": 0.1871,
      "num_input_tokens_seen": 4585792,
      "step": 2410
    },
    {
      "epoch": 0.39399624765478425,
      "grad_norm": 1.250996708869934,
      "learning_rate": 4.955201551120313e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 4595744,
      "step": 2415
    },
    {
      "epoch": 0.3948119748756016,
      "grad_norm": 0.6665579676628113,
      "learning_rate": 4.955016340382328e-05,
      "loss": 0.2528,
      "num_input_tokens_seen": 4606160,
      "step": 2420
    },
    {
      "epoch": 0.39562770209641895,
      "grad_norm": 1.0738847255706787,
      "learning_rate": 4.954830751049972e-05,
      "loss": 0.1279,
      "num_input_tokens_seen": 4615360,
      "step": 2425
    },
    {
      "epoch": 0.3964434293172363,
      "grad_norm": 1.3382548093795776,
      "learning_rate": 4.954644783151864e-05,
      "loss": 0.1998,
      "num_input_tokens_seen": 4625216,
      "step": 2430
    },
    {
      "epoch": 0.39725915653805366,
      "grad_norm": 0.6155682802200317,
      "learning_rate": 4.954458436716684e-05,
      "loss": 0.1422,
      "num_input_tokens_seen": 4634128,
      "step": 2435
    },
    {
      "epoch": 0.398074883758871,
      "grad_norm": 0.8232117891311646,
      "learning_rate": 4.954271711773168e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 4644032,
      "step": 2440
    },
    {
      "epoch": 0.39889061097968836,
      "grad_norm": 0.7818354368209839,
      "learning_rate": 4.9540846083501115e-05,
      "loss": 0.2056,
      "num_input_tokens_seen": 4654192,
      "step": 2445
    },
    {
      "epoch": 0.3997063382005058,
      "grad_norm": 0.4962497353553772,
      "learning_rate": 4.953897126476369e-05,
      "loss": 0.0867,
      "num_input_tokens_seen": 4664032,
      "step": 2450
    },
    {
      "epoch": 0.4005220654213231,
      "grad_norm": 0.7887216806411743,
      "learning_rate": 4.9537092661808514e-05,
      "loss": 0.2525,
      "num_input_tokens_seen": 4673888,
      "step": 2455
    },
    {
      "epoch": 0.4013377926421405,
      "grad_norm": 0.39339733123779297,
      "learning_rate": 4.9535210274925306e-05,
      "loss": 0.0913,
      "num_input_tokens_seen": 4684064,
      "step": 2460
    },
    {
      "epoch": 0.40215351986295783,
      "grad_norm": 0.132538840174675,
      "learning_rate": 4.953332410440435e-05,
      "loss": 0.1097,
      "num_input_tokens_seen": 4694560,
      "step": 2465
    },
    {
      "epoch": 0.4029692470837752,
      "grad_norm": 0.27074068784713745,
      "learning_rate": 4.9531434150536496e-05,
      "loss": 0.0631,
      "num_input_tokens_seen": 4703856,
      "step": 2470
    },
    {
      "epoch": 0.40378497430459254,
      "grad_norm": 0.40296033024787903,
      "learning_rate": 4.952954041361322e-05,
      "loss": 0.221,
      "num_input_tokens_seen": 4714528,
      "step": 2475
    },
    {
      "epoch": 0.4046007015254099,
      "grad_norm": 0.8925316333770752,
      "learning_rate": 4.952764289392655e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 4724480,
      "step": 2480
    },
    {
      "epoch": 0.40541642874622724,
      "grad_norm": 1.139454960823059,
      "learning_rate": 4.952574159176912e-05,
      "loss": 0.0868,
      "num_input_tokens_seen": 4732448,
      "step": 2485
    },
    {
      "epoch": 0.4062321559670446,
      "grad_norm": 0.18602700531482697,
      "learning_rate": 4.952383650743413e-05,
      "loss": 0.0748,
      "num_input_tokens_seen": 4740752,
      "step": 2490
    },
    {
      "epoch": 0.407047883187862,
      "grad_norm": 1.8236151933670044,
      "learning_rate": 4.952192764121536e-05,
      "loss": 0.2057,
      "num_input_tokens_seen": 4750336,
      "step": 2495
    },
    {
      "epoch": 0.40786361040867936,
      "grad_norm": 0.2704469561576843,
      "learning_rate": 4.9520014993407185e-05,
      "loss": 0.1182,
      "num_input_tokens_seen": 4759168,
      "step": 2500
    },
    {
      "epoch": 0.4086793376294967,
      "grad_norm": 0.22705025970935822,
      "learning_rate": 4.951809856430456e-05,
      "loss": 0.1769,
      "num_input_tokens_seen": 4768928,
      "step": 2505
    },
    {
      "epoch": 0.40949506485031406,
      "grad_norm": 0.5947977900505066,
      "learning_rate": 4.951617835420303e-05,
      "loss": 0.2521,
      "num_input_tokens_seen": 4778080,
      "step": 2510
    },
    {
      "epoch": 0.4103107920711314,
      "grad_norm": 0.6008138060569763,
      "learning_rate": 4.951425436339869e-05,
      "loss": 0.1749,
      "num_input_tokens_seen": 4786432,
      "step": 2515
    },
    {
      "epoch": 0.41112651929194877,
      "grad_norm": 2.607851982116699,
      "learning_rate": 4.9512326592188274e-05,
      "loss": 0.324,
      "num_input_tokens_seen": 4795920,
      "step": 2520
    },
    {
      "epoch": 0.4119422465127661,
      "grad_norm": 2.9825048446655273,
      "learning_rate": 4.9510395040869054e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 4805072,
      "step": 2525
    },
    {
      "epoch": 0.41275797373358347,
      "grad_norm": 0.1748061627149582,
      "learning_rate": 4.9508459709738905e-05,
      "loss": 0.2078,
      "num_input_tokens_seen": 4814896,
      "step": 2530
    },
    {
      "epoch": 0.4135737009544008,
      "grad_norm": 1.325439214706421,
      "learning_rate": 4.950652059909627e-05,
      "loss": 0.1217,
      "num_input_tokens_seen": 4822288,
      "step": 2535
    },
    {
      "epoch": 0.41438942817521823,
      "grad_norm": 0.4742916524410248,
      "learning_rate": 4.95045777092402e-05,
      "loss": 0.1814,
      "num_input_tokens_seen": 4831152,
      "step": 2540
    },
    {
      "epoch": 0.4152051553960356,
      "grad_norm": 2.422130823135376,
      "learning_rate": 4.950263104047031e-05,
      "loss": 0.3813,
      "num_input_tokens_seen": 4840512,
      "step": 2545
    },
    {
      "epoch": 0.41602088261685294,
      "grad_norm": 0.15169715881347656,
      "learning_rate": 4.9500680593086775e-05,
      "loss": 0.0511,
      "num_input_tokens_seen": 4850176,
      "step": 2550
    },
    {
      "epoch": 0.4168366098376703,
      "grad_norm": 0.23792028427124023,
      "learning_rate": 4.94987263673904e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 4859520,
      "step": 2555
    },
    {
      "epoch": 0.41765233705848764,
      "grad_norm": 0.37704676389694214,
      "learning_rate": 4.949676836368256e-05,
      "loss": 0.1389,
      "num_input_tokens_seen": 4869232,
      "step": 2560
    },
    {
      "epoch": 0.418468064279305,
      "grad_norm": 1.039276123046875,
      "learning_rate": 4.949480658226518e-05,
      "loss": 0.1256,
      "num_input_tokens_seen": 4879792,
      "step": 2565
    },
    {
      "epoch": 0.41928379150012235,
      "grad_norm": 0.9675346612930298,
      "learning_rate": 4.949284102344082e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 4888192,
      "step": 2570
    },
    {
      "epoch": 0.4200995187209397,
      "grad_norm": 1.3656415939331055,
      "learning_rate": 4.9490871687512565e-05,
      "loss": 0.2551,
      "num_input_tokens_seen": 4898416,
      "step": 2575
    },
    {
      "epoch": 0.42091524594175705,
      "grad_norm": 0.9151525497436523,
      "learning_rate": 4.948889857478413e-05,
      "loss": 0.0757,
      "num_input_tokens_seen": 4907280,
      "step": 2580
    },
    {
      "epoch": 0.4217309731625744,
      "grad_norm": 0.8305755853652954,
      "learning_rate": 4.948692168555978e-05,
      "loss": 0.1292,
      "num_input_tokens_seen": 4916160,
      "step": 2585
    },
    {
      "epoch": 0.4225467003833918,
      "grad_norm": 0.2908661961555481,
      "learning_rate": 4.94849410201444e-05,
      "loss": 0.1164,
      "num_input_tokens_seen": 4924720,
      "step": 2590
    },
    {
      "epoch": 0.42336242760420917,
      "grad_norm": 3.3345088958740234,
      "learning_rate": 4.948295657884341e-05,
      "loss": 0.2282,
      "num_input_tokens_seen": 4934784,
      "step": 2595
    },
    {
      "epoch": 0.4241781548250265,
      "grad_norm": 2.5341038703918457,
      "learning_rate": 4.9480968361962835e-05,
      "loss": 0.486,
      "num_input_tokens_seen": 4944384,
      "step": 2600
    },
    {
      "epoch": 0.4241781548250265,
      "eval_loss": 0.17167143523693085,
      "eval_runtime": 133.1698,
      "eval_samples_per_second": 20.463,
      "eval_steps_per_second": 10.235,
      "num_input_tokens_seen": 4944384,
      "step": 2600
    },
    {
      "epoch": 0.4249938820458439,
      "grad_norm": 1.9377453327178955,
      "learning_rate": 4.9478976369809305e-05,
      "loss": 0.2348,
      "num_input_tokens_seen": 4954240,
      "step": 2605
    },
    {
      "epoch": 0.4258096092666612,
      "grad_norm": 0.32576560974121094,
      "learning_rate": 4.947698060268999e-05,
      "loss": 0.0652,
      "num_input_tokens_seen": 4963168,
      "step": 2610
    },
    {
      "epoch": 0.4266253364874786,
      "grad_norm": 1.3894671201705933,
      "learning_rate": 4.9474981060912665e-05,
      "loss": 0.2343,
      "num_input_tokens_seen": 4972016,
      "step": 2615
    },
    {
      "epoch": 0.42744106370829593,
      "grad_norm": 1.5759952068328857,
      "learning_rate": 4.94729777447857e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 4980992,
      "step": 2620
    },
    {
      "epoch": 0.4282567909291133,
      "grad_norm": 0.9019586443901062,
      "learning_rate": 4.947097065461801e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 4990784,
      "step": 2625
    },
    {
      "epoch": 0.42907251814993064,
      "grad_norm": 0.31558263301849365,
      "learning_rate": 4.9468959790719125e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 5000464,
      "step": 2630
    },
    {
      "epoch": 0.42988824537074805,
      "grad_norm": 0.7953096032142639,
      "learning_rate": 4.9466945153399146e-05,
      "loss": 0.2128,
      "num_input_tokens_seen": 5008960,
      "step": 2635
    },
    {
      "epoch": 0.4307039725915654,
      "grad_norm": 0.1391826570034027,
      "learning_rate": 4.9464926742968755e-05,
      "loss": 0.0634,
      "num_input_tokens_seen": 5018944,
      "step": 2640
    },
    {
      "epoch": 0.43151969981238275,
      "grad_norm": 0.6601292490959167,
      "learning_rate": 4.946290455973921e-05,
      "loss": 0.1737,
      "num_input_tokens_seen": 5029088,
      "step": 2645
    },
    {
      "epoch": 0.4323354270332001,
      "grad_norm": 2.2608039379119873,
      "learning_rate": 4.9460878604022365e-05,
      "loss": 0.4215,
      "num_input_tokens_seen": 5038528,
      "step": 2650
    },
    {
      "epoch": 0.43315115425401746,
      "grad_norm": 0.527417778968811,
      "learning_rate": 4.945884887613065e-05,
      "loss": 0.0763,
      "num_input_tokens_seen": 5047648,
      "step": 2655
    },
    {
      "epoch": 0.4339668814748348,
      "grad_norm": 0.2824595868587494,
      "learning_rate": 4.9456815376377055e-05,
      "loss": 0.1983,
      "num_input_tokens_seen": 5057168,
      "step": 2660
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 0.6329284310340881,
      "learning_rate": 4.9454778105075195e-05,
      "loss": 0.1094,
      "num_input_tokens_seen": 5066560,
      "step": 2665
    },
    {
      "epoch": 0.4355983359164695,
      "grad_norm": 0.3288673460483551,
      "learning_rate": 4.945273706253924e-05,
      "loss": 0.0588,
      "num_input_tokens_seen": 5076400,
      "step": 2670
    },
    {
      "epoch": 0.43641406313728687,
      "grad_norm": 0.6333422660827637,
      "learning_rate": 4.9450692249083925e-05,
      "loss": 0.1856,
      "num_input_tokens_seen": 5085632,
      "step": 2675
    },
    {
      "epoch": 0.4372297903581043,
      "grad_norm": 2.390387535095215,
      "learning_rate": 4.9448643665024605e-05,
      "loss": 0.184,
      "num_input_tokens_seen": 5094832,
      "step": 2680
    },
    {
      "epoch": 0.43804551757892163,
      "grad_norm": 0.17397256195545197,
      "learning_rate": 4.944659131067719e-05,
      "loss": 0.0727,
      "num_input_tokens_seen": 5105168,
      "step": 2685
    },
    {
      "epoch": 0.438861244799739,
      "grad_norm": 0.09544269740581512,
      "learning_rate": 4.944453518635818e-05,
      "loss": 0.2017,
      "num_input_tokens_seen": 5114240,
      "step": 2690
    },
    {
      "epoch": 0.43967697202055633,
      "grad_norm": 0.6979895830154419,
      "learning_rate": 4.944247529238465e-05,
      "loss": 0.2429,
      "num_input_tokens_seen": 5122960,
      "step": 2695
    },
    {
      "epoch": 0.4404926992413737,
      "grad_norm": 0.883449137210846,
      "learning_rate": 4.944041162907427e-05,
      "loss": 0.1224,
      "num_input_tokens_seen": 5130896,
      "step": 2700
    },
    {
      "epoch": 0.44130842646219104,
      "grad_norm": 0.9254229068756104,
      "learning_rate": 4.943834419674529e-05,
      "loss": 0.3298,
      "num_input_tokens_seen": 5141072,
      "step": 2705
    },
    {
      "epoch": 0.4421241536830084,
      "grad_norm": 0.8441610336303711,
      "learning_rate": 4.9436272995716506e-05,
      "loss": 0.1826,
      "num_input_tokens_seen": 5149536,
      "step": 2710
    },
    {
      "epoch": 0.44293988090382574,
      "grad_norm": 0.44043850898742676,
      "learning_rate": 4.943419802630735e-05,
      "loss": 0.1196,
      "num_input_tokens_seen": 5158944,
      "step": 2715
    },
    {
      "epoch": 0.4437556081246431,
      "grad_norm": 0.5315830707550049,
      "learning_rate": 4.94321192888378e-05,
      "loss": 0.2078,
      "num_input_tokens_seen": 5169472,
      "step": 2720
    },
    {
      "epoch": 0.4445713353454605,
      "grad_norm": 1.1286325454711914,
      "learning_rate": 4.943003678362842e-05,
      "loss": 0.375,
      "num_input_tokens_seen": 5179376,
      "step": 2725
    },
    {
      "epoch": 0.44538706256627786,
      "grad_norm": 1.201703667640686,
      "learning_rate": 4.942795051100036e-05,
      "loss": 0.1164,
      "num_input_tokens_seen": 5188192,
      "step": 2730
    },
    {
      "epoch": 0.4462027897870952,
      "grad_norm": 1.2158184051513672,
      "learning_rate": 4.942586047127536e-05,
      "loss": 0.1277,
      "num_input_tokens_seen": 5198320,
      "step": 2735
    },
    {
      "epoch": 0.44701851700791256,
      "grad_norm": 0.22909873723983765,
      "learning_rate": 4.942376666477571e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 5207872,
      "step": 2740
    },
    {
      "epoch": 0.4478342442287299,
      "grad_norm": 0.9519137144088745,
      "learning_rate": 4.9421669091824304e-05,
      "loss": 0.1141,
      "num_input_tokens_seen": 5216432,
      "step": 2745
    },
    {
      "epoch": 0.44864997144954727,
      "grad_norm": 1.0054973363876343,
      "learning_rate": 4.9419567752744634e-05,
      "loss": 0.2736,
      "num_input_tokens_seen": 5227024,
      "step": 2750
    },
    {
      "epoch": 0.4494656986703646,
      "grad_norm": 1.865236520767212,
      "learning_rate": 4.941746264786074e-05,
      "loss": 0.3958,
      "num_input_tokens_seen": 5236928,
      "step": 2755
    },
    {
      "epoch": 0.450281425891182,
      "grad_norm": 0.6959102153778076,
      "learning_rate": 4.9415353777497254e-05,
      "loss": 0.3032,
      "num_input_tokens_seen": 5247120,
      "step": 2760
    },
    {
      "epoch": 0.4510971531119993,
      "grad_norm": 0.3129928410053253,
      "learning_rate": 4.9413241141979394e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 5257136,
      "step": 2765
    },
    {
      "epoch": 0.4519128803328167,
      "grad_norm": 1.7251510620117188,
      "learning_rate": 4.9411124741632956e-05,
      "loss": 0.1366,
      "num_input_tokens_seen": 5266528,
      "step": 2770
    },
    {
      "epoch": 0.4527286075536341,
      "grad_norm": 1.0015630722045898,
      "learning_rate": 4.940900457678431e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 5276640,
      "step": 2775
    },
    {
      "epoch": 0.45354433477445144,
      "grad_norm": 1.1887389421463013,
      "learning_rate": 4.9406880647760425e-05,
      "loss": 0.1248,
      "num_input_tokens_seen": 5286816,
      "step": 2780
    },
    {
      "epoch": 0.4543600619952688,
      "grad_norm": 1.7474429607391357,
      "learning_rate": 4.9404752954888824e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 5296752,
      "step": 2785
    },
    {
      "epoch": 0.45517578921608615,
      "grad_norm": 0.07182968407869339,
      "learning_rate": 4.940262149849762e-05,
      "loss": 0.1134,
      "num_input_tokens_seen": 5305216,
      "step": 2790
    },
    {
      "epoch": 0.4559915164369035,
      "grad_norm": 2.2939987182617188,
      "learning_rate": 4.9400486278915526e-05,
      "loss": 0.2848,
      "num_input_tokens_seen": 5316032,
      "step": 2795
    },
    {
      "epoch": 0.45680724365772085,
      "grad_norm": 0.7449731230735779,
      "learning_rate": 4.939834729647181e-05,
      "loss": 0.2059,
      "num_input_tokens_seen": 5325216,
      "step": 2800
    },
    {
      "epoch": 0.45680724365772085,
      "eval_loss": 0.16885004937648773,
      "eval_runtime": 132.9855,
      "eval_samples_per_second": 20.491,
      "eval_steps_per_second": 10.249,
      "num_input_tokens_seen": 5325216,
      "step": 2800
    },
    {
      "epoch": 0.4576229708785382,
      "grad_norm": 2.261556386947632,
      "learning_rate": 4.9396204551496326e-05,
      "loss": 0.4111,
      "num_input_tokens_seen": 5334016,
      "step": 2805
    },
    {
      "epoch": 0.45843869809935556,
      "grad_norm": 0.9332560896873474,
      "learning_rate": 4.939405804431952e-05,
      "loss": 0.0944,
      "num_input_tokens_seen": 5342672,
      "step": 2810
    },
    {
      "epoch": 0.4592544253201729,
      "grad_norm": 2.2029013633728027,
      "learning_rate": 4.9391907775272414e-05,
      "loss": 0.1394,
      "num_input_tokens_seen": 5351328,
      "step": 2815
    },
    {
      "epoch": 0.4600701525409903,
      "grad_norm": 0.8381232619285583,
      "learning_rate": 4.9389753744686604e-05,
      "loss": 0.1109,
      "num_input_tokens_seen": 5361520,
      "step": 2820
    },
    {
      "epoch": 0.46088587976180767,
      "grad_norm": 1.053671956062317,
      "learning_rate": 4.938759595289426e-05,
      "loss": 0.0605,
      "num_input_tokens_seen": 5371184,
      "step": 2825
    },
    {
      "epoch": 0.461701606982625,
      "grad_norm": 1.5793319940567017,
      "learning_rate": 4.938543440022815e-05,
      "loss": 0.3441,
      "num_input_tokens_seen": 5380704,
      "step": 2830
    },
    {
      "epoch": 0.4625173342034424,
      "grad_norm": 0.4004461169242859,
      "learning_rate": 4.938326908702161e-05,
      "loss": 0.0703,
      "num_input_tokens_seen": 5389584,
      "step": 2835
    },
    {
      "epoch": 0.46333306142425973,
      "grad_norm": 0.7076089978218079,
      "learning_rate": 4.9381100013608554e-05,
      "loss": 0.1981,
      "num_input_tokens_seen": 5399312,
      "step": 2840
    },
    {
      "epoch": 0.4641487886450771,
      "grad_norm": 2.258239984512329,
      "learning_rate": 4.9378927180323485e-05,
      "loss": 0.2166,
      "num_input_tokens_seen": 5408912,
      "step": 2845
    },
    {
      "epoch": 0.46496451586589443,
      "grad_norm": 1.038957953453064,
      "learning_rate": 4.937675058750148e-05,
      "loss": 0.2275,
      "num_input_tokens_seen": 5418512,
      "step": 2850
    },
    {
      "epoch": 0.4657802430867118,
      "grad_norm": 2.1498725414276123,
      "learning_rate": 4.937457023547819e-05,
      "loss": 0.2027,
      "num_input_tokens_seen": 5428288,
      "step": 2855
    },
    {
      "epoch": 0.46659597030752914,
      "grad_norm": 1.8528378009796143,
      "learning_rate": 4.9372386124589876e-05,
      "loss": 0.1685,
      "num_input_tokens_seen": 5437888,
      "step": 2860
    },
    {
      "epoch": 0.46741169752834655,
      "grad_norm": 0.42744091153144836,
      "learning_rate": 4.937019825517333e-05,
      "loss": 0.0635,
      "num_input_tokens_seen": 5446656,
      "step": 2865
    },
    {
      "epoch": 0.4682274247491639,
      "grad_norm": 0.905279278755188,
      "learning_rate": 4.9368006627565954e-05,
      "loss": 0.3679,
      "num_input_tokens_seen": 5455696,
      "step": 2870
    },
    {
      "epoch": 0.46904315196998125,
      "grad_norm": 0.9215461611747742,
      "learning_rate": 4.936581124210573e-05,
      "loss": 0.2623,
      "num_input_tokens_seen": 5465056,
      "step": 2875
    },
    {
      "epoch": 0.4698588791907986,
      "grad_norm": 0.5466522574424744,
      "learning_rate": 4.9363612099131216e-05,
      "loss": 0.4248,
      "num_input_tokens_seen": 5474560,
      "step": 2880
    },
    {
      "epoch": 0.47067460641161596,
      "grad_norm": 1.3446576595306396,
      "learning_rate": 4.936140919898155e-05,
      "loss": 0.193,
      "num_input_tokens_seen": 5483584,
      "step": 2885
    },
    {
      "epoch": 0.4714903336324333,
      "grad_norm": 0.1332315355539322,
      "learning_rate": 4.9359202541996426e-05,
      "loss": 0.0646,
      "num_input_tokens_seen": 5492112,
      "step": 2890
    },
    {
      "epoch": 0.47230606085325066,
      "grad_norm": 0.8549277186393738,
      "learning_rate": 4.935699212851616e-05,
      "loss": 0.1946,
      "num_input_tokens_seen": 5501536,
      "step": 2895
    },
    {
      "epoch": 0.473121788074068,
      "grad_norm": 0.7489109635353088,
      "learning_rate": 4.935477795888162e-05,
      "loss": 0.1629,
      "num_input_tokens_seen": 5511280,
      "step": 2900
    },
    {
      "epoch": 0.47393751529488537,
      "grad_norm": 0.13561555743217468,
      "learning_rate": 4.935256003343426e-05,
      "loss": 0.1237,
      "num_input_tokens_seen": 5520016,
      "step": 2905
    },
    {
      "epoch": 0.4747532425157027,
      "grad_norm": 0.41387489438056946,
      "learning_rate": 4.93503383525161e-05,
      "loss": 0.2181,
      "num_input_tokens_seen": 5529808,
      "step": 2910
    },
    {
      "epoch": 0.47556896973652013,
      "grad_norm": 0.15072408318519592,
      "learning_rate": 4.934811291646977e-05,
      "loss": 0.1988,
      "num_input_tokens_seen": 5538896,
      "step": 2915
    },
    {
      "epoch": 0.4763846969573375,
      "grad_norm": 0.5867430567741394,
      "learning_rate": 4.934588372563845e-05,
      "loss": 0.078,
      "num_input_tokens_seen": 5548112,
      "step": 2920
    },
    {
      "epoch": 0.47720042417815484,
      "grad_norm": 2.030919313430786,
      "learning_rate": 4.93436507803659e-05,
      "loss": 0.2079,
      "num_input_tokens_seen": 5557504,
      "step": 2925
    },
    {
      "epoch": 0.4780161513989722,
      "grad_norm": 0.7732830047607422,
      "learning_rate": 4.934141408099649e-05,
      "loss": 0.2551,
      "num_input_tokens_seen": 5566208,
      "step": 2930
    },
    {
      "epoch": 0.47883187861978954,
      "grad_norm": 1.238082766532898,
      "learning_rate": 4.9339173627875135e-05,
      "loss": 0.1476,
      "num_input_tokens_seen": 5576176,
      "step": 2935
    },
    {
      "epoch": 0.4796476058406069,
      "grad_norm": 1.909623384475708,
      "learning_rate": 4.9336929421347335e-05,
      "loss": 0.3744,
      "num_input_tokens_seen": 5584688,
      "step": 2940
    },
    {
      "epoch": 0.48046333306142425,
      "grad_norm": 1.5802147388458252,
      "learning_rate": 4.933468146175918e-05,
      "loss": 0.091,
      "num_input_tokens_seen": 5592784,
      "step": 2945
    },
    {
      "epoch": 0.4812790602822416,
      "grad_norm": 0.3804842531681061,
      "learning_rate": 4.933242974945734e-05,
      "loss": 0.1186,
      "num_input_tokens_seen": 5602336,
      "step": 2950
    },
    {
      "epoch": 0.48209478750305895,
      "grad_norm": 0.08182328939437866,
      "learning_rate": 4.933017428478906e-05,
      "loss": 0.0514,
      "num_input_tokens_seen": 5612160,
      "step": 2955
    },
    {
      "epoch": 0.48291051472387636,
      "grad_norm": 1.0956923961639404,
      "learning_rate": 4.932791506810214e-05,
      "loss": 0.135,
      "num_input_tokens_seen": 5621392,
      "step": 2960
    },
    {
      "epoch": 0.4837262419446937,
      "grad_norm": 0.5830453038215637,
      "learning_rate": 4.932565209974499e-05,
      "loss": 0.0771,
      "num_input_tokens_seen": 5631696,
      "step": 2965
    },
    {
      "epoch": 0.48454196916551107,
      "grad_norm": 0.60622239112854,
      "learning_rate": 4.93233853800666e-05,
      "loss": 0.2074,
      "num_input_tokens_seen": 5640880,
      "step": 2970
    },
    {
      "epoch": 0.4853576963863284,
      "grad_norm": 0.9151844382286072,
      "learning_rate": 4.932111490941651e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 5650368,
      "step": 2975
    },
    {
      "epoch": 0.48617342360714577,
      "grad_norm": 0.13530747592449188,
      "learning_rate": 4.9318840688144876e-05,
      "loss": 0.167,
      "num_input_tokens_seen": 5660960,
      "step": 2980
    },
    {
      "epoch": 0.4869891508279631,
      "grad_norm": 0.7268181443214417,
      "learning_rate": 4.9316562716602387e-05,
      "loss": 0.2634,
      "num_input_tokens_seen": 5670592,
      "step": 2985
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 2.0541136264801025,
      "learning_rate": 4.9314280995140346e-05,
      "loss": 0.2369,
      "num_input_tokens_seen": 5679360,
      "step": 2990
    },
    {
      "epoch": 0.48862060526959783,
      "grad_norm": 0.8217318058013916,
      "learning_rate": 4.931199552411063e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 5689264,
      "step": 2995
    },
    {
      "epoch": 0.4894363324904152,
      "grad_norm": 0.7274150252342224,
      "learning_rate": 4.930970630386568e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 5698896,
      "step": 3000
    },
    {
      "epoch": 0.4894363324904152,
      "eval_loss": 0.16604842245578766,
      "eval_runtime": 133.2055,
      "eval_samples_per_second": 20.457,
      "eval_steps_per_second": 10.232,
      "num_input_tokens_seen": 5698896,
      "step": 3000
    },
    {
      "epoch": 0.4902520597112326,
      "grad_norm": 1.1137927770614624,
      "learning_rate": 4.9307413334758524e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 5708816,
      "step": 3005
    },
    {
      "epoch": 0.49106778693204994,
      "grad_norm": 1.5817726850509644,
      "learning_rate": 4.930511661714276e-05,
      "loss": 0.1721,
      "num_input_tokens_seen": 5718384,
      "step": 3010
    },
    {
      "epoch": 0.4918835141528673,
      "grad_norm": 0.9721799492835999,
      "learning_rate": 4.9302816151372576e-05,
      "loss": 0.2113,
      "num_input_tokens_seen": 5727120,
      "step": 3015
    },
    {
      "epoch": 0.49269924137368465,
      "grad_norm": 2.396566390991211,
      "learning_rate": 4.930051193780274e-05,
      "loss": 0.2339,
      "num_input_tokens_seen": 5735888,
      "step": 3020
    },
    {
      "epoch": 0.493514968594502,
      "grad_norm": 0.30275943875312805,
      "learning_rate": 4.929820397678858e-05,
      "loss": 0.0978,
      "num_input_tokens_seen": 5744672,
      "step": 3025
    },
    {
      "epoch": 0.49433069581531935,
      "grad_norm": 0.8180168271064758,
      "learning_rate": 4.9295892268686015e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 5753984,
      "step": 3030
    },
    {
      "epoch": 0.4951464230361367,
      "grad_norm": 0.3686399459838867,
      "learning_rate": 4.9293576813851536e-05,
      "loss": 0.0724,
      "num_input_tokens_seen": 5763472,
      "step": 3035
    },
    {
      "epoch": 0.49596215025695406,
      "grad_norm": 0.4319579601287842,
      "learning_rate": 4.929125761264223e-05,
      "loss": 0.2559,
      "num_input_tokens_seen": 5773440,
      "step": 3040
    },
    {
      "epoch": 0.4967778774777714,
      "grad_norm": 0.3013916611671448,
      "learning_rate": 4.928893466541573e-05,
      "loss": 0.2478,
      "num_input_tokens_seen": 5783664,
      "step": 3045
    },
    {
      "epoch": 0.49759360469858877,
      "grad_norm": 1.5588116645812988,
      "learning_rate": 4.928660797253027e-05,
      "loss": 0.212,
      "num_input_tokens_seen": 5793872,
      "step": 3050
    },
    {
      "epoch": 0.4984093319194062,
      "grad_norm": 1.1791826486587524,
      "learning_rate": 4.928427753434467e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 5803600,
      "step": 3055
    },
    {
      "epoch": 0.4992250591402235,
      "grad_norm": 1.4717106819152832,
      "learning_rate": 4.9281943351218286e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 5812320,
      "step": 3060
    },
    {
      "epoch": 0.5000407863610409,
      "grad_norm": 1.6231176853179932,
      "learning_rate": 4.9279605423511095e-05,
      "loss": 0.2726,
      "num_input_tokens_seen": 5821408,
      "step": 3065
    },
    {
      "epoch": 0.5008565135818582,
      "grad_norm": 1.5469762086868286,
      "learning_rate": 4.927726375158363e-05,
      "loss": 0.2938,
      "num_input_tokens_seen": 5830752,
      "step": 3070
    },
    {
      "epoch": 0.5016722408026756,
      "grad_norm": 0.46597009897232056,
      "learning_rate": 4.9274918335797004e-05,
      "loss": 0.0966,
      "num_input_tokens_seen": 5839616,
      "step": 3075
    },
    {
      "epoch": 0.5024879680234929,
      "grad_norm": 1.2342963218688965,
      "learning_rate": 4.927256917651292e-05,
      "loss": 0.1724,
      "num_input_tokens_seen": 5848800,
      "step": 3080
    },
    {
      "epoch": 0.5033036952443103,
      "grad_norm": 1.0319805145263672,
      "learning_rate": 4.927021627409364e-05,
      "loss": 0.2422,
      "num_input_tokens_seen": 5857920,
      "step": 3085
    },
    {
      "epoch": 0.5041194224651276,
      "grad_norm": 1.2228456735610962,
      "learning_rate": 4.9267859628902005e-05,
      "loss": 0.1729,
      "num_input_tokens_seen": 5866624,
      "step": 3090
    },
    {
      "epoch": 0.504935149685945,
      "grad_norm": 1.7883638143539429,
      "learning_rate": 4.9265499241301454e-05,
      "loss": 0.2997,
      "num_input_tokens_seen": 5876240,
      "step": 3095
    },
    {
      "epoch": 0.5057508769067623,
      "grad_norm": 1.6699402332305908,
      "learning_rate": 4.926313511165598e-05,
      "loss": 0.2604,
      "num_input_tokens_seen": 5885472,
      "step": 3100
    },
    {
      "epoch": 0.5065666041275797,
      "grad_norm": 0.44467076659202576,
      "learning_rate": 4.926076724033016e-05,
      "loss": 0.0713,
      "num_input_tokens_seen": 5895488,
      "step": 3105
    },
    {
      "epoch": 0.507382331348397,
      "grad_norm": 0.2455458790063858,
      "learning_rate": 4.9258395627689146e-05,
      "loss": 0.2755,
      "num_input_tokens_seen": 5905840,
      "step": 3110
    },
    {
      "epoch": 0.5081980585692144,
      "grad_norm": 0.3314821124076843,
      "learning_rate": 4.925602027409868e-05,
      "loss": 0.1058,
      "num_input_tokens_seen": 5915280,
      "step": 3115
    },
    {
      "epoch": 0.5090137857900318,
      "grad_norm": 1.3515701293945312,
      "learning_rate": 4.925364117992507e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 5924784,
      "step": 3120
    },
    {
      "epoch": 0.5098295130108492,
      "grad_norm": 0.8027592897415161,
      "learning_rate": 4.92512583455352e-05,
      "loss": 0.2939,
      "num_input_tokens_seen": 5934432,
      "step": 3125
    },
    {
      "epoch": 0.5106452402316666,
      "grad_norm": 0.07312072813510895,
      "learning_rate": 4.9248871771296536e-05,
      "loss": 0.1972,
      "num_input_tokens_seen": 5943360,
      "step": 3130
    },
    {
      "epoch": 0.5114609674524839,
      "grad_norm": 2.159242630004883,
      "learning_rate": 4.924648145757711e-05,
      "loss": 0.2581,
      "num_input_tokens_seen": 5952016,
      "step": 3135
    },
    {
      "epoch": 0.5122766946733013,
      "grad_norm": 0.8258094191551208,
      "learning_rate": 4.924408740474554e-05,
      "loss": 0.1547,
      "num_input_tokens_seen": 5961568,
      "step": 3140
    },
    {
      "epoch": 0.5130924218941186,
      "grad_norm": 0.20302478969097137,
      "learning_rate": 4.924168961317103e-05,
      "loss": 0.0996,
      "num_input_tokens_seen": 5970544,
      "step": 3145
    },
    {
      "epoch": 0.513908149114936,
      "grad_norm": 0.36450979113578796,
      "learning_rate": 4.9239288083223334e-05,
      "loss": 0.0749,
      "num_input_tokens_seen": 5978896,
      "step": 3150
    },
    {
      "epoch": 0.5147238763357533,
      "grad_norm": 0.7469759583473206,
      "learning_rate": 4.9236882815272803e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 5986976,
      "step": 3155
    },
    {
      "epoch": 0.5155396035565707,
      "grad_norm": 1.3396360874176025,
      "learning_rate": 4.9234473809690365e-05,
      "loss": 0.1895,
      "num_input_tokens_seen": 5995632,
      "step": 3160
    },
    {
      "epoch": 0.516355330777388,
      "grad_norm": 0.4467015564441681,
      "learning_rate": 4.923206106684752e-05,
      "loss": 0.1831,
      "num_input_tokens_seen": 6005728,
      "step": 3165
    },
    {
      "epoch": 0.5171710579982054,
      "grad_norm": 0.31315749883651733,
      "learning_rate": 4.922964458711634e-05,
      "loss": 0.1467,
      "num_input_tokens_seen": 6015856,
      "step": 3170
    },
    {
      "epoch": 0.5179867852190227,
      "grad_norm": 0.829138457775116,
      "learning_rate": 4.9227224370869474e-05,
      "loss": 0.1414,
      "num_input_tokens_seen": 6025088,
      "step": 3175
    },
    {
      "epoch": 0.5188025124398401,
      "grad_norm": 1.0441538095474243,
      "learning_rate": 4.9224800418480155e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 6034352,
      "step": 3180
    },
    {
      "epoch": 0.5196182396606575,
      "grad_norm": 1.3986988067626953,
      "learning_rate": 4.9222372730322176e-05,
      "loss": 0.245,
      "num_input_tokens_seen": 6044320,
      "step": 3185
    },
    {
      "epoch": 0.5204339668814748,
      "grad_norm": 2.111682653427124,
      "learning_rate": 4.921994130676993e-05,
      "loss": 0.2237,
      "num_input_tokens_seen": 6055072,
      "step": 3190
    },
    {
      "epoch": 0.5212496941022922,
      "grad_norm": 0.36753329634666443,
      "learning_rate": 4.9217506148198366e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 6064688,
      "step": 3195
    },
    {
      "epoch": 0.5220654213231095,
      "grad_norm": 2.348046064376831,
      "learning_rate": 4.921506725498302e-05,
      "loss": 0.3348,
      "num_input_tokens_seen": 6074432,
      "step": 3200
    },
    {
      "epoch": 0.5220654213231095,
      "eval_loss": 0.1693449467420578,
      "eval_runtime": 133.2124,
      "eval_samples_per_second": 20.456,
      "eval_steps_per_second": 10.232,
      "num_input_tokens_seen": 6074432,
      "step": 3200
    },
    {
      "epoch": 0.5228811485439269,
      "grad_norm": 1.219756007194519,
      "learning_rate": 4.9212624627499994e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 6084256,
      "step": 3205
    },
    {
      "epoch": 0.5236968757647442,
      "grad_norm": 0.38175299763679504,
      "learning_rate": 4.921017826612597e-05,
      "loss": 0.0793,
      "num_input_tokens_seen": 6093616,
      "step": 3210
    },
    {
      "epoch": 0.5245126029855617,
      "grad_norm": 2.0236339569091797,
      "learning_rate": 4.9207728171238223e-05,
      "loss": 0.2353,
      "num_input_tokens_seen": 6103184,
      "step": 3215
    },
    {
      "epoch": 0.525328330206379,
      "grad_norm": 0.4813328683376312,
      "learning_rate": 4.920527434321458e-05,
      "loss": 0.0574,
      "num_input_tokens_seen": 6111728,
      "step": 3220
    },
    {
      "epoch": 0.5261440574271964,
      "grad_norm": 2.3659305572509766,
      "learning_rate": 4.920281678243345e-05,
      "loss": 0.3504,
      "num_input_tokens_seen": 6121328,
      "step": 3225
    },
    {
      "epoch": 0.5269597846480137,
      "grad_norm": 0.8389526605606079,
      "learning_rate": 4.920035548927381e-05,
      "loss": 0.0974,
      "num_input_tokens_seen": 6130496,
      "step": 3230
    },
    {
      "epoch": 0.5277755118688311,
      "grad_norm": 0.7969070672988892,
      "learning_rate": 4.919789046411525e-05,
      "loss": 0.2235,
      "num_input_tokens_seen": 6139344,
      "step": 3235
    },
    {
      "epoch": 0.5285912390896484,
      "grad_norm": 1.3612346649169922,
      "learning_rate": 4.919542170733787e-05,
      "loss": 0.1945,
      "num_input_tokens_seen": 6149280,
      "step": 3240
    },
    {
      "epoch": 0.5294069663104658,
      "grad_norm": 0.8073938488960266,
      "learning_rate": 4.919294921932242e-05,
      "loss": 0.0714,
      "num_input_tokens_seen": 6158224,
      "step": 3245
    },
    {
      "epoch": 0.5302226935312832,
      "grad_norm": 0.43264302611351013,
      "learning_rate": 4.919047300045016e-05,
      "loss": 0.342,
      "num_input_tokens_seen": 6167648,
      "step": 3250
    },
    {
      "epoch": 0.5310384207521005,
      "grad_norm": 1.828895926475525,
      "learning_rate": 4.918799305110299e-05,
      "loss": 0.2437,
      "num_input_tokens_seen": 6177600,
      "step": 3255
    },
    {
      "epoch": 0.5318541479729179,
      "grad_norm": 1.4274793863296509,
      "learning_rate": 4.918550937166331e-05,
      "loss": 0.283,
      "num_input_tokens_seen": 6186688,
      "step": 3260
    },
    {
      "epoch": 0.5326698751937352,
      "grad_norm": 2.0096983909606934,
      "learning_rate": 4.918302196251415e-05,
      "loss": 0.2534,
      "num_input_tokens_seen": 6196384,
      "step": 3265
    },
    {
      "epoch": 0.5334856024145526,
      "grad_norm": 1.6929148435592651,
      "learning_rate": 4.91805308240391e-05,
      "loss": 0.2343,
      "num_input_tokens_seen": 6205760,
      "step": 3270
    },
    {
      "epoch": 0.5343013296353699,
      "grad_norm": 1.6670657396316528,
      "learning_rate": 4.9178035956622326e-05,
      "loss": 0.158,
      "num_input_tokens_seen": 6215584,
      "step": 3275
    },
    {
      "epoch": 0.5351170568561873,
      "grad_norm": 0.21768498420715332,
      "learning_rate": 4.917553736064857e-05,
      "loss": 0.2021,
      "num_input_tokens_seen": 6226016,
      "step": 3280
    },
    {
      "epoch": 0.5359327840770046,
      "grad_norm": 0.4138501286506653,
      "learning_rate": 4.917303503650314e-05,
      "loss": 0.0815,
      "num_input_tokens_seen": 6236080,
      "step": 3285
    },
    {
      "epoch": 0.536748511297822,
      "grad_norm": 0.7821457982063293,
      "learning_rate": 4.917052898457194e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 6245056,
      "step": 3290
    },
    {
      "epoch": 0.5375642385186393,
      "grad_norm": 1.7103575468063354,
      "learning_rate": 4.916801920524141e-05,
      "loss": 0.2109,
      "num_input_tokens_seen": 6254224,
      "step": 3295
    },
    {
      "epoch": 0.5383799657394567,
      "grad_norm": 0.6247990131378174,
      "learning_rate": 4.916550569889862e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 6263952,
      "step": 3300
    },
    {
      "epoch": 0.539195692960274,
      "grad_norm": 0.404748797416687,
      "learning_rate": 4.916298846593116e-05,
      "loss": 0.0812,
      "num_input_tokens_seen": 6274096,
      "step": 3305
    },
    {
      "epoch": 0.5400114201810915,
      "grad_norm": 0.4049340486526489,
      "learning_rate": 4.916046750672722e-05,
      "loss": 0.1025,
      "num_input_tokens_seen": 6283904,
      "step": 3310
    },
    {
      "epoch": 0.5408271474019088,
      "grad_norm": 0.12082573771476746,
      "learning_rate": 4.915794282167559e-05,
      "loss": 0.2813,
      "num_input_tokens_seen": 6293648,
      "step": 3315
    },
    {
      "epoch": 0.5416428746227262,
      "grad_norm": 0.9795013666152954,
      "learning_rate": 4.915541441116558e-05,
      "loss": 0.2058,
      "num_input_tokens_seen": 6302720,
      "step": 3320
    },
    {
      "epoch": 0.5424586018435436,
      "grad_norm": 0.10642383247613907,
      "learning_rate": 4.915288227558711e-05,
      "loss": 0.2051,
      "num_input_tokens_seen": 6312240,
      "step": 3325
    },
    {
      "epoch": 0.5432743290643609,
      "grad_norm": 0.693794310092926,
      "learning_rate": 4.915034641533066e-05,
      "loss": 0.0908,
      "num_input_tokens_seen": 6321312,
      "step": 3330
    },
    {
      "epoch": 0.5440900562851783,
      "grad_norm": 0.232740581035614,
      "learning_rate": 4.914780683078731e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 6330512,
      "step": 3335
    },
    {
      "epoch": 0.5449057835059956,
      "grad_norm": 0.10620350390672684,
      "learning_rate": 4.9145263522348695e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 6339872,
      "step": 3340
    },
    {
      "epoch": 0.545721510726813,
      "grad_norm": 1.0265778303146362,
      "learning_rate": 4.9142716490407e-05,
      "loss": 0.3032,
      "num_input_tokens_seen": 6349488,
      "step": 3345
    },
    {
      "epoch": 0.5465372379476303,
      "grad_norm": 1.2603627443313599,
      "learning_rate": 4.914016573535504e-05,
      "loss": 0.2416,
      "num_input_tokens_seen": 6358480,
      "step": 3350
    },
    {
      "epoch": 0.5473529651684477,
      "grad_norm": 0.15514923632144928,
      "learning_rate": 4.9137611257586154e-05,
      "loss": 0.2064,
      "num_input_tokens_seen": 6367264,
      "step": 3355
    },
    {
      "epoch": 0.548168692389265,
      "grad_norm": 0.2952350378036499,
      "learning_rate": 4.9135053057494274e-05,
      "loss": 0.1106,
      "num_input_tokens_seen": 6376784,
      "step": 3360
    },
    {
      "epoch": 0.5489844196100824,
      "grad_norm": 1.0135488510131836,
      "learning_rate": 4.913249113547392e-05,
      "loss": 0.1174,
      "num_input_tokens_seen": 6386336,
      "step": 3365
    },
    {
      "epoch": 0.5498001468308997,
      "grad_norm": 0.46586403250694275,
      "learning_rate": 4.912992549192016e-05,
      "loss": 0.0907,
      "num_input_tokens_seen": 6395600,
      "step": 3370
    },
    {
      "epoch": 0.5506158740517171,
      "grad_norm": 1.547369360923767,
      "learning_rate": 4.9127356127228665e-05,
      "loss": 0.1869,
      "num_input_tokens_seen": 6405280,
      "step": 3375
    },
    {
      "epoch": 0.5514316012725344,
      "grad_norm": 0.4394802749156952,
      "learning_rate": 4.912478304179564e-05,
      "loss": 0.062,
      "num_input_tokens_seen": 6415184,
      "step": 3380
    },
    {
      "epoch": 0.5522473284933518,
      "grad_norm": 1.246511459350586,
      "learning_rate": 4.9122206236017896e-05,
      "loss": 0.1701,
      "num_input_tokens_seen": 6425984,
      "step": 3385
    },
    {
      "epoch": 0.5530630557141691,
      "grad_norm": 2.249781608581543,
      "learning_rate": 4.911962571029282e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 6435952,
      "step": 3390
    },
    {
      "epoch": 0.5538787829349865,
      "grad_norm": 0.9046935439109802,
      "learning_rate": 4.9117041465018353e-05,
      "loss": 0.243,
      "num_input_tokens_seen": 6445472,
      "step": 3395
    },
    {
      "epoch": 0.5546945101558038,
      "grad_norm": 0.2740926444530487,
      "learning_rate": 4.911445350059302e-05,
      "loss": 0.0563,
      "num_input_tokens_seen": 6454208,
      "step": 3400
    },
    {
      "epoch": 0.5546945101558038,
      "eval_loss": 0.16440188884735107,
      "eval_runtime": 133.2737,
      "eval_samples_per_second": 20.447,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 6454208,
      "step": 3400
    },
    {
      "epoch": 0.5555102373766213,
      "grad_norm": 2.013253927230835,
      "learning_rate": 4.9111861817415905e-05,
      "loss": 0.2784,
      "num_input_tokens_seen": 6463312,
      "step": 3405
    },
    {
      "epoch": 0.5563259645974387,
      "grad_norm": 0.1894233375787735,
      "learning_rate": 4.91092664158867e-05,
      "loss": 0.0998,
      "num_input_tokens_seen": 6471632,
      "step": 3410
    },
    {
      "epoch": 0.557141691818256,
      "grad_norm": 1.3361207246780396,
      "learning_rate": 4.910666729640563e-05,
      "loss": 0.2163,
      "num_input_tokens_seen": 6480976,
      "step": 3415
    },
    {
      "epoch": 0.5579574190390734,
      "grad_norm": 1.0692594051361084,
      "learning_rate": 4.910406445937353e-05,
      "loss": 0.2089,
      "num_input_tokens_seen": 6490832,
      "step": 3420
    },
    {
      "epoch": 0.5587731462598907,
      "grad_norm": 0.059294115751981735,
      "learning_rate": 4.9101457905191774e-05,
      "loss": 0.135,
      "num_input_tokens_seen": 6500112,
      "step": 3425
    },
    {
      "epoch": 0.5595888734807081,
      "grad_norm": 0.9320427179336548,
      "learning_rate": 4.909884763426233e-05,
      "loss": 0.101,
      "num_input_tokens_seen": 6508992,
      "step": 3430
    },
    {
      "epoch": 0.5604046007015254,
      "grad_norm": 0.49356314539909363,
      "learning_rate": 4.9096233646987736e-05,
      "loss": 0.2773,
      "num_input_tokens_seen": 6518976,
      "step": 3435
    },
    {
      "epoch": 0.5612203279223428,
      "grad_norm": 0.1395374983549118,
      "learning_rate": 4.9093615943771104e-05,
      "loss": 0.1893,
      "num_input_tokens_seen": 6528160,
      "step": 3440
    },
    {
      "epoch": 0.5620360551431601,
      "grad_norm": 0.08500301837921143,
      "learning_rate": 4.909099452501611e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 6536128,
      "step": 3445
    },
    {
      "epoch": 0.5628517823639775,
      "grad_norm": 0.44670185446739197,
      "learning_rate": 4.908836939112702e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 6545504,
      "step": 3450
    },
    {
      "epoch": 0.5636675095847948,
      "grad_norm": 1.05312180519104,
      "learning_rate": 4.908574054250865e-05,
      "loss": 0.2602,
      "num_input_tokens_seen": 6556048,
      "step": 3455
    },
    {
      "epoch": 0.5644832368056122,
      "grad_norm": 0.24583542346954346,
      "learning_rate": 4.9083107979566414e-05,
      "loss": 0.0853,
      "num_input_tokens_seen": 6565616,
      "step": 3460
    },
    {
      "epoch": 0.5652989640264295,
      "grad_norm": 0.8435521721839905,
      "learning_rate": 4.908047170270628e-05,
      "loss": 0.1152,
      "num_input_tokens_seen": 6575600,
      "step": 3465
    },
    {
      "epoch": 0.5661146912472469,
      "grad_norm": 0.7685081362724304,
      "learning_rate": 4.9077831712334784e-05,
      "loss": 0.2061,
      "num_input_tokens_seen": 6585008,
      "step": 3470
    },
    {
      "epoch": 0.5669304184680642,
      "grad_norm": 0.22187961637973785,
      "learning_rate": 4.907518800885907e-05,
      "loss": 0.1412,
      "num_input_tokens_seen": 6594896,
      "step": 3475
    },
    {
      "epoch": 0.5677461456888816,
      "grad_norm": 1.8919501304626465,
      "learning_rate": 4.907254059268681e-05,
      "loss": 0.3595,
      "num_input_tokens_seen": 6603632,
      "step": 3480
    },
    {
      "epoch": 0.568561872909699,
      "grad_norm": 1.742069959640503,
      "learning_rate": 4.906988946422628e-05,
      "loss": 0.2535,
      "num_input_tokens_seen": 6613264,
      "step": 3485
    },
    {
      "epoch": 0.5693776001305163,
      "grad_norm": 0.6573871970176697,
      "learning_rate": 4.9067234623886315e-05,
      "loss": 0.088,
      "num_input_tokens_seen": 6623040,
      "step": 3490
    },
    {
      "epoch": 0.5701933273513338,
      "grad_norm": 0.15846553444862366,
      "learning_rate": 4.9064576072076316e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 6633200,
      "step": 3495
    },
    {
      "epoch": 0.5710090545721511,
      "grad_norm": 1.1575722694396973,
      "learning_rate": 4.906191380920628e-05,
      "loss": 0.2543,
      "num_input_tokens_seen": 6642576,
      "step": 3500
    },
    {
      "epoch": 0.5718247817929685,
      "grad_norm": 0.5509277582168579,
      "learning_rate": 4.905924783568675e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 6650736,
      "step": 3505
    },
    {
      "epoch": 0.5726405090137858,
      "grad_norm": 0.9171512722969055,
      "learning_rate": 4.905657815192886e-05,
      "loss": 0.3394,
      "num_input_tokens_seen": 6660416,
      "step": 3510
    },
    {
      "epoch": 0.5734562362346032,
      "grad_norm": 0.2156369835138321,
      "learning_rate": 4.90539047583443e-05,
      "loss": 0.0756,
      "num_input_tokens_seen": 6670576,
      "step": 3515
    },
    {
      "epoch": 0.5742719634554205,
      "grad_norm": 1.8034660816192627,
      "learning_rate": 4.905122765534534e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 6679680,
      "step": 3520
    },
    {
      "epoch": 0.5750876906762379,
      "grad_norm": 1.1042964458465576,
      "learning_rate": 4.9048546843344846e-05,
      "loss": 0.0844,
      "num_input_tokens_seen": 6688544,
      "step": 3525
    },
    {
      "epoch": 0.5759034178970552,
      "grad_norm": 0.30032989382743835,
      "learning_rate": 4.9045862322756206e-05,
      "loss": 0.052,
      "num_input_tokens_seen": 6698560,
      "step": 3530
    },
    {
      "epoch": 0.5767191451178726,
      "grad_norm": 0.3302914798259735,
      "learning_rate": 4.904317409399342e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 6708816,
      "step": 3535
    },
    {
      "epoch": 0.5775348723386899,
      "grad_norm": 0.3727857172489166,
      "learning_rate": 4.904048215747104e-05,
      "loss": 0.1061,
      "num_input_tokens_seen": 6717984,
      "step": 3540
    },
    {
      "epoch": 0.5783505995595073,
      "grad_norm": 0.40824654698371887,
      "learning_rate": 4.90377865136042e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 6727232,
      "step": 3545
    },
    {
      "epoch": 0.5791663267803246,
      "grad_norm": 1.1156797409057617,
      "learning_rate": 4.90350871628086e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 6736832,
      "step": 3550
    },
    {
      "epoch": 0.579982054001142,
      "grad_norm": 1.2025023698806763,
      "learning_rate": 4.903238410550052e-05,
      "loss": 0.1126,
      "num_input_tokens_seen": 6746640,
      "step": 3555
    },
    {
      "epoch": 0.5807977812219594,
      "grad_norm": 0.820178747177124,
      "learning_rate": 4.90296773420968e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 6756800,
      "step": 3560
    },
    {
      "epoch": 0.5816135084427767,
      "grad_norm": 0.6440359354019165,
      "learning_rate": 4.902696687301486e-05,
      "loss": 0.082,
      "num_input_tokens_seen": 6765840,
      "step": 3565
    },
    {
      "epoch": 0.5824292356635941,
      "grad_norm": 1.295622706413269,
      "learning_rate": 4.902425269867268e-05,
      "loss": 0.1389,
      "num_input_tokens_seen": 6774912,
      "step": 3570
    },
    {
      "epoch": 0.5832449628844114,
      "grad_norm": 0.4805803596973419,
      "learning_rate": 4.902153481948883e-05,
      "loss": 0.1131,
      "num_input_tokens_seen": 6784832,
      "step": 3575
    },
    {
      "epoch": 0.5840606901052288,
      "grad_norm": 0.05258537456393242,
      "learning_rate": 4.901881323588244e-05,
      "loss": 0.1723,
      "num_input_tokens_seen": 6793504,
      "step": 3580
    },
    {
      "epoch": 0.5848764173260461,
      "grad_norm": 0.06167425960302353,
      "learning_rate": 4.90160879482732e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 6802000,
      "step": 3585
    },
    {
      "epoch": 0.5856921445468636,
      "grad_norm": 0.29396775364875793,
      "learning_rate": 4.9013358957081405e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 6811680,
      "step": 3590
    },
    {
      "epoch": 0.5865078717676809,
      "grad_norm": 0.6537598371505737,
      "learning_rate": 4.901062626272789e-05,
      "loss": 0.099,
      "num_input_tokens_seen": 6822112,
      "step": 3595
    },
    {
      "epoch": 0.5873235989884983,
      "grad_norm": 0.6097482442855835,
      "learning_rate": 4.900788986563406e-05,
      "loss": 0.0621,
      "num_input_tokens_seen": 6831056,
      "step": 3600
    },
    {
      "epoch": 0.5873235989884983,
      "eval_loss": 0.16681234538555145,
      "eval_runtime": 133.2803,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 6831056,
      "step": 3600
    },
    {
      "epoch": 0.5881393262093156,
      "grad_norm": 0.19379457831382751,
      "learning_rate": 4.9005149766221915e-05,
      "loss": 0.2226,
      "num_input_tokens_seen": 6840528,
      "step": 3605
    },
    {
      "epoch": 0.588955053430133,
      "grad_norm": 0.3810690641403198,
      "learning_rate": 4.9002405964914e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 6850016,
      "step": 3610
    },
    {
      "epoch": 0.5897707806509503,
      "grad_norm": 1.0904264450073242,
      "learning_rate": 4.899965846213346e-05,
      "loss": 0.2412,
      "num_input_tokens_seen": 6859552,
      "step": 3615
    },
    {
      "epoch": 0.5905865078717677,
      "grad_norm": 0.06072712317109108,
      "learning_rate": 4.899690725830399e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 6868656,
      "step": 3620
    },
    {
      "epoch": 0.591402235092585,
      "grad_norm": 0.6752508878707886,
      "learning_rate": 4.899415235384985e-05,
      "loss": 0.0665,
      "num_input_tokens_seen": 6879600,
      "step": 3625
    },
    {
      "epoch": 0.5922179623134024,
      "grad_norm": 1.720642328262329,
      "learning_rate": 4.899139374919589e-05,
      "loss": 0.1868,
      "num_input_tokens_seen": 6888944,
      "step": 3630
    },
    {
      "epoch": 0.5930336895342198,
      "grad_norm": 1.0869008302688599,
      "learning_rate": 4.898863144476752e-05,
      "loss": 0.1934,
      "num_input_tokens_seen": 6898288,
      "step": 3635
    },
    {
      "epoch": 0.5938494167550371,
      "grad_norm": 1.5018373727798462,
      "learning_rate": 4.898586544099072e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 6906880,
      "step": 3640
    },
    {
      "epoch": 0.5946651439758545,
      "grad_norm": 0.22033117711544037,
      "learning_rate": 4.898309573829204e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 6915824,
      "step": 3645
    },
    {
      "epoch": 0.5954808711966718,
      "grad_norm": 2.502547025680542,
      "learning_rate": 4.898032233709862e-05,
      "loss": 0.1947,
      "num_input_tokens_seen": 6925488,
      "step": 3650
    },
    {
      "epoch": 0.5962965984174892,
      "grad_norm": 1.1929550170898438,
      "learning_rate": 4.8977545237838123e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 6936112,
      "step": 3655
    },
    {
      "epoch": 0.5971123256383065,
      "grad_norm": 0.4178708791732788,
      "learning_rate": 4.8974764440938836e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 6946480,
      "step": 3660
    },
    {
      "epoch": 0.5979280528591239,
      "grad_norm": 1.918763279914856,
      "learning_rate": 4.897197994682959e-05,
      "loss": 0.2602,
      "num_input_tokens_seen": 6955568,
      "step": 3665
    },
    {
      "epoch": 0.5987437800799412,
      "grad_norm": 1.1491422653198242,
      "learning_rate": 4.8969191755939786e-05,
      "loss": 0.1765,
      "num_input_tokens_seen": 6964400,
      "step": 3670
    },
    {
      "epoch": 0.5995595073007586,
      "grad_norm": 0.5969305038452148,
      "learning_rate": 4.8966399868699396e-05,
      "loss": 0.1234,
      "num_input_tokens_seen": 6974288,
      "step": 3675
    },
    {
      "epoch": 0.600375234521576,
      "grad_norm": 1.0924203395843506,
      "learning_rate": 4.8963604285538965e-05,
      "loss": 0.1431,
      "num_input_tokens_seen": 6985776,
      "step": 3680
    },
    {
      "epoch": 0.6011909617423934,
      "grad_norm": 0.9503542184829712,
      "learning_rate": 4.8960805006889604e-05,
      "loss": 0.101,
      "num_input_tokens_seen": 6993440,
      "step": 3685
    },
    {
      "epoch": 0.6020066889632107,
      "grad_norm": 1.7106534242630005,
      "learning_rate": 4.8958002033183004e-05,
      "loss": 0.2621,
      "num_input_tokens_seen": 7003648,
      "step": 3690
    },
    {
      "epoch": 0.6028224161840281,
      "grad_norm": 1.1617361307144165,
      "learning_rate": 4.8955195364851414e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 7011776,
      "step": 3695
    },
    {
      "epoch": 0.6036381434048455,
      "grad_norm": 0.1271304488182068,
      "learning_rate": 4.895238500232766e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 7020816,
      "step": 3700
    },
    {
      "epoch": 0.6044538706256628,
      "grad_norm": 1.0061200857162476,
      "learning_rate": 4.8949570946045143e-05,
      "loss": 0.2115,
      "num_input_tokens_seen": 7031104,
      "step": 3705
    },
    {
      "epoch": 0.6052695978464802,
      "grad_norm": 0.18648511171340942,
      "learning_rate": 4.89467531964378e-05,
      "loss": 0.0943,
      "num_input_tokens_seen": 7039776,
      "step": 3710
    },
    {
      "epoch": 0.6060853250672975,
      "grad_norm": 2.227543830871582,
      "learning_rate": 4.894393175394019e-05,
      "loss": 0.2984,
      "num_input_tokens_seen": 7048976,
      "step": 3715
    },
    {
      "epoch": 0.6069010522881149,
      "grad_norm": 0.4651886522769928,
      "learning_rate": 4.8941106618987406e-05,
      "loss": 0.1072,
      "num_input_tokens_seen": 7057888,
      "step": 3720
    },
    {
      "epoch": 0.6077167795089322,
      "grad_norm": 0.5756481289863586,
      "learning_rate": 4.893827779201512e-05,
      "loss": 0.1834,
      "num_input_tokens_seen": 7067632,
      "step": 3725
    },
    {
      "epoch": 0.6085325067297496,
      "grad_norm": 0.9644002914428711,
      "learning_rate": 4.893544527345957e-05,
      "loss": 0.0716,
      "num_input_tokens_seen": 7076672,
      "step": 3730
    },
    {
      "epoch": 0.6093482339505669,
      "grad_norm": 0.5274430513381958,
      "learning_rate": 4.8932609063757563e-05,
      "loss": 0.2806,
      "num_input_tokens_seen": 7087680,
      "step": 3735
    },
    {
      "epoch": 0.6101639611713843,
      "grad_norm": 1.1169406175613403,
      "learning_rate": 4.8929769163346484e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 7097536,
      "step": 3740
    },
    {
      "epoch": 0.6109796883922016,
      "grad_norm": 0.5220376253128052,
      "learning_rate": 4.892692557266429e-05,
      "loss": 0.0607,
      "num_input_tokens_seen": 7106768,
      "step": 3745
    },
    {
      "epoch": 0.611795415613019,
      "grad_norm": 1.5127822160720825,
      "learning_rate": 4.8924078292149464e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 7116320,
      "step": 3750
    },
    {
      "epoch": 0.6126111428338363,
      "grad_norm": 0.9245728850364685,
      "learning_rate": 4.892122732224114e-05,
      "loss": 0.1117,
      "num_input_tokens_seen": 7125648,
      "step": 3755
    },
    {
      "epoch": 0.6134268700546537,
      "grad_norm": 0.31788119673728943,
      "learning_rate": 4.8918372663378944e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 7134352,
      "step": 3760
    },
    {
      "epoch": 0.614242597275471,
      "grad_norm": 0.35281702876091003,
      "learning_rate": 4.89155143160031e-05,
      "loss": 0.1152,
      "num_input_tokens_seen": 7144256,
      "step": 3765
    },
    {
      "epoch": 0.6150583244962884,
      "grad_norm": 2.0213918685913086,
      "learning_rate": 4.891265228055441e-05,
      "loss": 0.2889,
      "num_input_tokens_seen": 7154592,
      "step": 3770
    },
    {
      "epoch": 0.6158740517171059,
      "grad_norm": 0.39464059472084045,
      "learning_rate": 4.890978655747424e-05,
      "loss": 0.2361,
      "num_input_tokens_seen": 7164512,
      "step": 3775
    },
    {
      "epoch": 0.6166897789379232,
      "grad_norm": 0.42300790548324585,
      "learning_rate": 4.89069171472045e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 7173792,
      "step": 3780
    },
    {
      "epoch": 0.6175055061587406,
      "grad_norm": 0.8289487361907959,
      "learning_rate": 4.890404405018772e-05,
      "loss": 0.187,
      "num_input_tokens_seen": 7182480,
      "step": 3785
    },
    {
      "epoch": 0.6183212333795579,
      "grad_norm": 0.9306904077529907,
      "learning_rate": 4.8901167266866934e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 7191840,
      "step": 3790
    },
    {
      "epoch": 0.6191369606003753,
      "grad_norm": 2.606008529663086,
      "learning_rate": 4.88982867976858e-05,
      "loss": 0.1747,
      "num_input_tokens_seen": 7201152,
      "step": 3795
    },
    {
      "epoch": 0.6199526878211926,
      "grad_norm": 0.9524069428443909,
      "learning_rate": 4.889540264308852e-05,
      "loss": 0.099,
      "num_input_tokens_seen": 7209536,
      "step": 3800
    },
    {
      "epoch": 0.6199526878211926,
      "eval_loss": 0.1654559224843979,
      "eval_runtime": 133.224,
      "eval_samples_per_second": 20.454,
      "eval_steps_per_second": 10.231,
      "num_input_tokens_seen": 7209536,
      "step": 3800
    },
    {
      "epoch": 0.62076841504201,
      "grad_norm": 0.8904304504394531,
      "learning_rate": 4.889251480351986e-05,
      "loss": 0.1366,
      "num_input_tokens_seen": 7218736,
      "step": 3805
    },
    {
      "epoch": 0.6215841422628273,
      "grad_norm": 0.8531232476234436,
      "learning_rate": 4.888962327942517e-05,
      "loss": 0.1529,
      "num_input_tokens_seen": 7227952,
      "step": 3810
    },
    {
      "epoch": 0.6223998694836447,
      "grad_norm": 1.260356068611145,
      "learning_rate": 4.8886728071250356e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 7239056,
      "step": 3815
    },
    {
      "epoch": 0.623215596704462,
      "grad_norm": 0.6758674383163452,
      "learning_rate": 4.8883829179441884e-05,
      "loss": 0.0759,
      "num_input_tokens_seen": 7248816,
      "step": 3820
    },
    {
      "epoch": 0.6240313239252794,
      "grad_norm": 2.396521806716919,
      "learning_rate": 4.888092660444682e-05,
      "loss": 0.3302,
      "num_input_tokens_seen": 7258448,
      "step": 3825
    },
    {
      "epoch": 0.6248470511460967,
      "grad_norm": 0.3039580285549164,
      "learning_rate": 4.887802034671276e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 7268128,
      "step": 3830
    },
    {
      "epoch": 0.6256627783669141,
      "grad_norm": 0.0624595582485199,
      "learning_rate": 4.88751104066879e-05,
      "loss": 0.0692,
      "num_input_tokens_seen": 7276912,
      "step": 3835
    },
    {
      "epoch": 0.6264785055877314,
      "grad_norm": 1.867896318435669,
      "learning_rate": 4.887219678482098e-05,
      "loss": 0.2761,
      "num_input_tokens_seen": 7287328,
      "step": 3840
    },
    {
      "epoch": 0.6272942328085488,
      "grad_norm": 0.037623148411512375,
      "learning_rate": 4.8869279481561316e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 7295568,
      "step": 3845
    },
    {
      "epoch": 0.6281099600293661,
      "grad_norm": 0.24948376417160034,
      "learning_rate": 4.88663584973588e-05,
      "loss": 0.0491,
      "num_input_tokens_seen": 7305920,
      "step": 3850
    },
    {
      "epoch": 0.6289256872501835,
      "grad_norm": 1.1958732604980469,
      "learning_rate": 4.8863433832663874e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 7313616,
      "step": 3855
    },
    {
      "epoch": 0.6297414144710008,
      "grad_norm": 0.2967202067375183,
      "learning_rate": 4.886050548792757e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 7323328,
      "step": 3860
    },
    {
      "epoch": 0.6305571416918182,
      "grad_norm": 0.7706910371780396,
      "learning_rate": 4.8857573463601465e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 7333248,
      "step": 3865
    },
    {
      "epoch": 0.6313728689126357,
      "grad_norm": 0.075591541826725,
      "learning_rate": 4.885463776013772e-05,
      "loss": 0.2378,
      "num_input_tokens_seen": 7343408,
      "step": 3870
    },
    {
      "epoch": 0.632188596133453,
      "grad_norm": 1.160649299621582,
      "learning_rate": 4.8851698377989056e-05,
      "loss": 0.3019,
      "num_input_tokens_seen": 7353216,
      "step": 3875
    },
    {
      "epoch": 0.6330043233542704,
      "grad_norm": 0.13721877336502075,
      "learning_rate": 4.884875531760876e-05,
      "loss": 0.1739,
      "num_input_tokens_seen": 7362768,
      "step": 3880
    },
    {
      "epoch": 0.6338200505750877,
      "grad_norm": 0.42988768219947815,
      "learning_rate": 4.88458085794507e-05,
      "loss": 0.109,
      "num_input_tokens_seen": 7371488,
      "step": 3885
    },
    {
      "epoch": 0.6346357777959051,
      "grad_norm": 0.747150182723999,
      "learning_rate": 4.884285816396929e-05,
      "loss": 0.355,
      "num_input_tokens_seen": 7380464,
      "step": 3890
    },
    {
      "epoch": 0.6354515050167224,
      "grad_norm": 0.09934531897306442,
      "learning_rate": 4.8839904071619526e-05,
      "loss": 0.1814,
      "num_input_tokens_seen": 7389792,
      "step": 3895
    },
    {
      "epoch": 0.6362672322375398,
      "grad_norm": 0.6064696907997131,
      "learning_rate": 4.8836946302856955e-05,
      "loss": 0.0516,
      "num_input_tokens_seen": 7400720,
      "step": 3900
    },
    {
      "epoch": 0.6370829594583571,
      "grad_norm": 0.5957384705543518,
      "learning_rate": 4.8833984858137715e-05,
      "loss": 0.1984,
      "num_input_tokens_seen": 7409856,
      "step": 3905
    },
    {
      "epoch": 0.6378986866791745,
      "grad_norm": 0.43513137102127075,
      "learning_rate": 4.8831019737918494e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 7420368,
      "step": 3910
    },
    {
      "epoch": 0.6387144138999918,
      "grad_norm": 0.29126256704330444,
      "learning_rate": 4.882805094265655e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 7429408,
      "step": 3915
    },
    {
      "epoch": 0.6395301411208092,
      "grad_norm": 2.0995757579803467,
      "learning_rate": 4.8825078472809706e-05,
      "loss": 0.1869,
      "num_input_tokens_seen": 7438640,
      "step": 3920
    },
    {
      "epoch": 0.6403458683416265,
      "grad_norm": 0.2930694818496704,
      "learning_rate": 4.882210232883635e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 7448768,
      "step": 3925
    },
    {
      "epoch": 0.6411615955624439,
      "grad_norm": 1.2018523216247559,
      "learning_rate": 4.881912251119546e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 7458896,
      "step": 3930
    },
    {
      "epoch": 0.6419773227832613,
      "grad_norm": 1.5628619194030762,
      "learning_rate": 4.881613902034654e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 7468528,
      "step": 3935
    },
    {
      "epoch": 0.6427930500040786,
      "grad_norm": 0.3254159390926361,
      "learning_rate": 4.88131518567497e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 7478240,
      "step": 3940
    },
    {
      "epoch": 0.643608777224896,
      "grad_norm": 1.7367810010910034,
      "learning_rate": 4.881016102086558e-05,
      "loss": 0.1598,
      "num_input_tokens_seen": 7487792,
      "step": 3945
    },
    {
      "epoch": 0.6444245044457133,
      "grad_norm": 0.43338844180107117,
      "learning_rate": 4.8807166513155425e-05,
      "loss": 0.2153,
      "num_input_tokens_seen": 7498000,
      "step": 3950
    },
    {
      "epoch": 0.6452402316665307,
      "grad_norm": 0.5677065253257751,
      "learning_rate": 4.8804168334081004e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 7508144,
      "step": 3955
    },
    {
      "epoch": 0.6460559588873481,
      "grad_norm": 0.1863517016172409,
      "learning_rate": 4.880116648410468e-05,
      "loss": 0.2578,
      "num_input_tokens_seen": 7517984,
      "step": 3960
    },
    {
      "epoch": 0.6468716861081655,
      "grad_norm": 0.22170871496200562,
      "learning_rate": 4.879816096368939e-05,
      "loss": 0.1146,
      "num_input_tokens_seen": 7527216,
      "step": 3965
    },
    {
      "epoch": 0.6476874133289828,
      "grad_norm": 0.38001763820648193,
      "learning_rate": 4.879515177329861e-05,
      "loss": 0.0691,
      "num_input_tokens_seen": 7535968,
      "step": 3970
    },
    {
      "epoch": 0.6485031405498002,
      "grad_norm": 0.47573089599609375,
      "learning_rate": 4.8792138913396394e-05,
      "loss": 0.0782,
      "num_input_tokens_seen": 7546064,
      "step": 3975
    },
    {
      "epoch": 0.6493188677706175,
      "grad_norm": 1.0914911031723022,
      "learning_rate": 4.8789122384447374e-05,
      "loss": 0.2034,
      "num_input_tokens_seen": 7555408,
      "step": 3980
    },
    {
      "epoch": 0.6501345949914349,
      "grad_norm": 0.024192171171307564,
      "learning_rate": 4.878610218691673e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 7564768,
      "step": 3985
    },
    {
      "epoch": 0.6509503222122522,
      "grad_norm": 0.1680450439453125,
      "learning_rate": 4.87830783212702e-05,
      "loss": 0.3221,
      "num_input_tokens_seen": 7574368,
      "step": 3990
    },
    {
      "epoch": 0.6517660494330696,
      "grad_norm": 0.09507381916046143,
      "learning_rate": 4.878005078797413e-05,
      "loss": 0.2899,
      "num_input_tokens_seen": 7583280,
      "step": 3995
    },
    {
      "epoch": 0.652581776653887,
      "grad_norm": 1.6238200664520264,
      "learning_rate": 4.877701958749539e-05,
      "loss": 0.2068,
      "num_input_tokens_seen": 7593024,
      "step": 4000
    },
    {
      "epoch": 0.652581776653887,
      "eval_loss": 0.16437526047229767,
      "eval_runtime": 133.2206,
      "eval_samples_per_second": 20.455,
      "eval_steps_per_second": 10.231,
      "num_input_tokens_seen": 7593024,
      "step": 4000
    },
    {
      "epoch": 0.6533975038747043,
      "grad_norm": 0.586816132068634,
      "learning_rate": 4.877398472030142e-05,
      "loss": 0.095,
      "num_input_tokens_seen": 7602256,
      "step": 4005
    },
    {
      "epoch": 0.6542132310955217,
      "grad_norm": 0.08943546563386917,
      "learning_rate": 4.877094618686024e-05,
      "loss": 0.0874,
      "num_input_tokens_seen": 7612336,
      "step": 4010
    },
    {
      "epoch": 0.655028958316339,
      "grad_norm": 2.1514248847961426,
      "learning_rate": 4.876790398764045e-05,
      "loss": 0.2021,
      "num_input_tokens_seen": 7621952,
      "step": 4015
    },
    {
      "epoch": 0.6558446855371564,
      "grad_norm": 0.528711199760437,
      "learning_rate": 4.8764858123111167e-05,
      "loss": 0.2013,
      "num_input_tokens_seen": 7631056,
      "step": 4020
    },
    {
      "epoch": 0.6566604127579737,
      "grad_norm": 1.5565980672836304,
      "learning_rate": 4.876180859374212e-05,
      "loss": 0.1664,
      "num_input_tokens_seen": 7640768,
      "step": 4025
    },
    {
      "epoch": 0.6574761399787911,
      "grad_norm": 0.8230895400047302,
      "learning_rate": 4.875875540000357e-05,
      "loss": 0.3095,
      "num_input_tokens_seen": 7651248,
      "step": 4030
    },
    {
      "epoch": 0.6582918671996084,
      "grad_norm": 1.193546175956726,
      "learning_rate": 4.8755698542366376e-05,
      "loss": 0.2169,
      "num_input_tokens_seen": 7661616,
      "step": 4035
    },
    {
      "epoch": 0.6591075944204258,
      "grad_norm": 1.885005235671997,
      "learning_rate": 4.875263802130193e-05,
      "loss": 0.1987,
      "num_input_tokens_seen": 7672656,
      "step": 4040
    },
    {
      "epoch": 0.6599233216412431,
      "grad_norm": 0.3082989454269409,
      "learning_rate": 4.8749573837282207e-05,
      "loss": 0.2323,
      "num_input_tokens_seen": 7681728,
      "step": 4045
    },
    {
      "epoch": 0.6607390488620605,
      "grad_norm": 0.21549297869205475,
      "learning_rate": 4.874650599077974e-05,
      "loss": 0.1191,
      "num_input_tokens_seen": 7690672,
      "step": 4050
    },
    {
      "epoch": 0.6615547760828779,
      "grad_norm": 0.5694040656089783,
      "learning_rate": 4.874343448226764e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 7700544,
      "step": 4055
    },
    {
      "epoch": 0.6623705033036953,
      "grad_norm": 0.4630415737628937,
      "learning_rate": 4.874035931221955e-05,
      "loss": 0.1158,
      "num_input_tokens_seen": 7710512,
      "step": 4060
    },
    {
      "epoch": 0.6631862305245126,
      "grad_norm": 0.2511799931526184,
      "learning_rate": 4.8737280481109724e-05,
      "loss": 0.0761,
      "num_input_tokens_seen": 7719744,
      "step": 4065
    },
    {
      "epoch": 0.66400195774533,
      "grad_norm": 1.7346254587173462,
      "learning_rate": 4.873419798941294e-05,
      "loss": 0.0999,
      "num_input_tokens_seen": 7729888,
      "step": 4070
    },
    {
      "epoch": 0.6648176849661473,
      "grad_norm": 0.7451698184013367,
      "learning_rate": 4.873111183760458e-05,
      "loss": 0.0578,
      "num_input_tokens_seen": 7740992,
      "step": 4075
    },
    {
      "epoch": 0.6656334121869647,
      "grad_norm": 0.4431256353855133,
      "learning_rate": 4.8728022026160537e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 7749232,
      "step": 4080
    },
    {
      "epoch": 0.666449139407782,
      "grad_norm": 1.1514657735824585,
      "learning_rate": 4.872492855555732e-05,
      "loss": 0.2176,
      "num_input_tokens_seen": 7758672,
      "step": 4085
    },
    {
      "epoch": 0.6672648666285994,
      "grad_norm": 0.257472962141037,
      "learning_rate": 4.8721831426271956e-05,
      "loss": 0.1333,
      "num_input_tokens_seen": 7766896,
      "step": 4090
    },
    {
      "epoch": 0.6680805938494168,
      "grad_norm": 0.3210359513759613,
      "learning_rate": 4.87187306387821e-05,
      "loss": 0.0646,
      "num_input_tokens_seen": 7777072,
      "step": 4095
    },
    {
      "epoch": 0.6688963210702341,
      "grad_norm": 0.04513811320066452,
      "learning_rate": 4.87156261935659e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 7787296,
      "step": 4100
    },
    {
      "epoch": 0.6697120482910515,
      "grad_norm": 1.5296412706375122,
      "learning_rate": 4.871251809110211e-05,
      "loss": 0.2262,
      "num_input_tokens_seen": 7797072,
      "step": 4105
    },
    {
      "epoch": 0.6705277755118688,
      "grad_norm": 2.561166286468506,
      "learning_rate": 4.8709406331870044e-05,
      "loss": 0.3528,
      "num_input_tokens_seen": 7806944,
      "step": 4110
    },
    {
      "epoch": 0.6713435027326862,
      "grad_norm": 0.10428949445486069,
      "learning_rate": 4.8706290916349574e-05,
      "loss": 0.2233,
      "num_input_tokens_seen": 7815328,
      "step": 4115
    },
    {
      "epoch": 0.6721592299535035,
      "grad_norm": 1.2535370588302612,
      "learning_rate": 4.8703171845021134e-05,
      "loss": 0.1692,
      "num_input_tokens_seen": 7824976,
      "step": 4120
    },
    {
      "epoch": 0.6729749571743209,
      "grad_norm": 1.2083402872085571,
      "learning_rate": 4.870004911836572e-05,
      "loss": 0.1932,
      "num_input_tokens_seen": 7834672,
      "step": 4125
    },
    {
      "epoch": 0.6737906843951382,
      "grad_norm": 0.10395753383636475,
      "learning_rate": 4.869692273686489e-05,
      "loss": 0.1481,
      "num_input_tokens_seen": 7844304,
      "step": 4130
    },
    {
      "epoch": 0.6746064116159556,
      "grad_norm": 0.10840736329555511,
      "learning_rate": 4.869379270100079e-05,
      "loss": 0.1292,
      "num_input_tokens_seen": 7853968,
      "step": 4135
    },
    {
      "epoch": 0.6754221388367729,
      "grad_norm": 2.044423818588257,
      "learning_rate": 4.86906590112561e-05,
      "loss": 0.2021,
      "num_input_tokens_seen": 7862928,
      "step": 4140
    },
    {
      "epoch": 0.6762378660575903,
      "grad_norm": 0.6965160965919495,
      "learning_rate": 4.8687521668114064e-05,
      "loss": 0.1691,
      "num_input_tokens_seen": 7871840,
      "step": 4145
    },
    {
      "epoch": 0.6770535932784078,
      "grad_norm": 2.6188526153564453,
      "learning_rate": 4.868438067205853e-05,
      "loss": 0.2612,
      "num_input_tokens_seen": 7880912,
      "step": 4150
    },
    {
      "epoch": 0.6778693204992251,
      "grad_norm": 0.7764772176742554,
      "learning_rate": 4.8681236023573844e-05,
      "loss": 0.213,
      "num_input_tokens_seen": 7890480,
      "step": 4155
    },
    {
      "epoch": 0.6786850477200425,
      "grad_norm": 0.29666271805763245,
      "learning_rate": 4.867808772314497e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 7900304,
      "step": 4160
    },
    {
      "epoch": 0.6795007749408598,
      "grad_norm": 0.7796967029571533,
      "learning_rate": 4.867493577125741e-05,
      "loss": 0.1346,
      "num_input_tokens_seen": 7909440,
      "step": 4165
    },
    {
      "epoch": 0.6803165021616772,
      "grad_norm": 0.09411506354808807,
      "learning_rate": 4.867178016839725e-05,
      "loss": 0.2287,
      "num_input_tokens_seen": 7919248,
      "step": 4170
    },
    {
      "epoch": 0.6811322293824945,
      "grad_norm": 1.91325843334198,
      "learning_rate": 4.8668620915051094e-05,
      "loss": 0.2279,
      "num_input_tokens_seen": 7928592,
      "step": 4175
    },
    {
      "epoch": 0.6819479566033119,
      "grad_norm": 1.105081558227539,
      "learning_rate": 4.866545801170616e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 7939040,
      "step": 4180
    },
    {
      "epoch": 0.6827636838241292,
      "grad_norm": 0.417776495218277,
      "learning_rate": 4.86622914588502e-05,
      "loss": 0.058,
      "num_input_tokens_seen": 7949008,
      "step": 4185
    },
    {
      "epoch": 0.6835794110449466,
      "grad_norm": 0.21675479412078857,
      "learning_rate": 4.865912125697154e-05,
      "loss": 0.1966,
      "num_input_tokens_seen": 7958528,
      "step": 4190
    },
    {
      "epoch": 0.6843951382657639,
      "grad_norm": 0.25338712334632874,
      "learning_rate": 4.865594740655907e-05,
      "loss": 0.3022,
      "num_input_tokens_seen": 7967424,
      "step": 4195
    },
    {
      "epoch": 0.6852108654865813,
      "grad_norm": 0.8459044694900513,
      "learning_rate": 4.865276990810222e-05,
      "loss": 0.0542,
      "num_input_tokens_seen": 7977072,
      "step": 4200
    },
    {
      "epoch": 0.6852108654865813,
      "eval_loss": 0.1623077690601349,
      "eval_runtime": 133.2828,
      "eval_samples_per_second": 20.445,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 7977072,
      "step": 4200
    },
    {
      "epoch": 0.6860265927073986,
      "grad_norm": 2.2078206539154053,
      "learning_rate": 4.8649588762091016e-05,
      "loss": 0.2976,
      "num_input_tokens_seen": 7987088,
      "step": 4205
    },
    {
      "epoch": 0.686842319928216,
      "grad_norm": 1.990769863128662,
      "learning_rate": 4.8646403969016016e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 7996672,
      "step": 4210
    },
    {
      "epoch": 0.6876580471490333,
      "grad_norm": 0.5172216296195984,
      "learning_rate": 4.864321552936838e-05,
      "loss": 0.0949,
      "num_input_tokens_seen": 8006704,
      "step": 4215
    },
    {
      "epoch": 0.6884737743698507,
      "grad_norm": 0.046215906739234924,
      "learning_rate": 4.864002344363978e-05,
      "loss": 0.106,
      "num_input_tokens_seen": 8016544,
      "step": 4220
    },
    {
      "epoch": 0.689289501590668,
      "grad_norm": 0.11669854074716568,
      "learning_rate": 4.863682771232248e-05,
      "loss": 0.095,
      "num_input_tokens_seen": 8026016,
      "step": 4225
    },
    {
      "epoch": 0.6901052288114854,
      "grad_norm": 0.0944276824593544,
      "learning_rate": 4.8633628335909324e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 8037184,
      "step": 4230
    },
    {
      "epoch": 0.6909209560323027,
      "grad_norm": 0.21959491074085236,
      "learning_rate": 4.8630425314893676e-05,
      "loss": 0.0934,
      "num_input_tokens_seen": 8046336,
      "step": 4235
    },
    {
      "epoch": 0.6917366832531202,
      "grad_norm": 0.12993931770324707,
      "learning_rate": 4.862721864976948e-05,
      "loss": 0.0704,
      "num_input_tokens_seen": 8054896,
      "step": 4240
    },
    {
      "epoch": 0.6925524104739376,
      "grad_norm": 0.044984370470047,
      "learning_rate": 4.862400834103125e-05,
      "loss": 0.0399,
      "num_input_tokens_seen": 8063632,
      "step": 4245
    },
    {
      "epoch": 0.6933681376947549,
      "grad_norm": 0.8113365173339844,
      "learning_rate": 4.862079438917406e-05,
      "loss": 0.1923,
      "num_input_tokens_seen": 8073072,
      "step": 4250
    },
    {
      "epoch": 0.6941838649155723,
      "grad_norm": 0.268297016620636,
      "learning_rate": 4.8617576794693536e-05,
      "loss": 0.104,
      "num_input_tokens_seen": 8081264,
      "step": 4255
    },
    {
      "epoch": 0.6949995921363896,
      "grad_norm": 0.26318204402923584,
      "learning_rate": 4.8614355558085875e-05,
      "loss": 0.0648,
      "num_input_tokens_seen": 8090368,
      "step": 4260
    },
    {
      "epoch": 0.695815319357207,
      "grad_norm": 0.7408422827720642,
      "learning_rate": 4.861113067984783e-05,
      "loss": 0.2659,
      "num_input_tokens_seen": 8100080,
      "step": 4265
    },
    {
      "epoch": 0.6966310465780243,
      "grad_norm": 0.32677707076072693,
      "learning_rate": 4.860790216047671e-05,
      "loss": 0.276,
      "num_input_tokens_seen": 8109056,
      "step": 4270
    },
    {
      "epoch": 0.6974467737988417,
      "grad_norm": 1.275887131690979,
      "learning_rate": 4.860467000047041e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 8117616,
      "step": 4275
    },
    {
      "epoch": 0.698262501019659,
      "grad_norm": 0.3908368945121765,
      "learning_rate": 4.860143420032737e-05,
      "loss": 0.0582,
      "num_input_tokens_seen": 8127408,
      "step": 4280
    },
    {
      "epoch": 0.6990782282404764,
      "grad_norm": 0.5804547071456909,
      "learning_rate": 4.859819476054657e-05,
      "loss": 0.2068,
      "num_input_tokens_seen": 8136624,
      "step": 4285
    },
    {
      "epoch": 0.6998939554612937,
      "grad_norm": 2.27117919921875,
      "learning_rate": 4.859495168162758e-05,
      "loss": 0.2472,
      "num_input_tokens_seen": 8145904,
      "step": 4290
    },
    {
      "epoch": 0.7007096826821111,
      "grad_norm": 0.045439526438713074,
      "learning_rate": 4.859170496407054e-05,
      "loss": 0.3012,
      "num_input_tokens_seen": 8154064,
      "step": 4295
    },
    {
      "epoch": 0.7015254099029284,
      "grad_norm": 0.6502560377120972,
      "learning_rate": 4.8588454608376114e-05,
      "loss": 0.0661,
      "num_input_tokens_seen": 8162384,
      "step": 4300
    },
    {
      "epoch": 0.7023411371237458,
      "grad_norm": 0.1775895208120346,
      "learning_rate": 4.8585200615045555e-05,
      "loss": 0.3503,
      "num_input_tokens_seen": 8171456,
      "step": 4305
    },
    {
      "epoch": 0.7031568643445631,
      "grad_norm": 0.7334839105606079,
      "learning_rate": 4.8581942984580674e-05,
      "loss": 0.1953,
      "num_input_tokens_seen": 8180880,
      "step": 4310
    },
    {
      "epoch": 0.7039725915653805,
      "grad_norm": 1.3030078411102295,
      "learning_rate": 4.857868171748384e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 8189952,
      "step": 4315
    },
    {
      "epoch": 0.7047883187861979,
      "grad_norm": 0.046325236558914185,
      "learning_rate": 4.8575416814257976e-05,
      "loss": 0.0369,
      "num_input_tokens_seen": 8199968,
      "step": 4320
    },
    {
      "epoch": 0.7056040460070152,
      "grad_norm": 0.11104127019643784,
      "learning_rate": 4.857214827540657e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 8209808,
      "step": 4325
    },
    {
      "epoch": 0.7064197732278326,
      "grad_norm": 0.27715054154396057,
      "learning_rate": 4.856887610143367e-05,
      "loss": 0.228,
      "num_input_tokens_seen": 8220496,
      "step": 4330
    },
    {
      "epoch": 0.70723550044865,
      "grad_norm": 0.9219970703125,
      "learning_rate": 4.8565600292843896e-05,
      "loss": 0.0794,
      "num_input_tokens_seen": 8229600,
      "step": 4335
    },
    {
      "epoch": 0.7080512276694674,
      "grad_norm": 0.7138078808784485,
      "learning_rate": 4.856232085014241e-05,
      "loss": 0.1961,
      "num_input_tokens_seen": 8240032,
      "step": 4340
    },
    {
      "epoch": 0.7088669548902847,
      "grad_norm": 0.2511295974254608,
      "learning_rate": 4.855903777383495e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 8250624,
      "step": 4345
    },
    {
      "epoch": 0.7096826821111021,
      "grad_norm": 0.4318373501300812,
      "learning_rate": 4.85557510644278e-05,
      "loss": 0.1604,
      "num_input_tokens_seen": 8260576,
      "step": 4350
    },
    {
      "epoch": 0.7104984093319194,
      "grad_norm": 0.6676339507102966,
      "learning_rate": 4.855246072242782e-05,
      "loss": 0.0749,
      "num_input_tokens_seen": 8270528,
      "step": 4355
    },
    {
      "epoch": 0.7113141365527368,
      "grad_norm": 0.4278769791126251,
      "learning_rate": 4.8549166748342414e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 8279280,
      "step": 4360
    },
    {
      "epoch": 0.7121298637735541,
      "grad_norm": 0.2745625376701355,
      "learning_rate": 4.8545869142679556e-05,
      "loss": 0.1839,
      "num_input_tokens_seen": 8287600,
      "step": 4365
    },
    {
      "epoch": 0.7129455909943715,
      "grad_norm": 0.4992622137069702,
      "learning_rate": 4.8542567905947776e-05,
      "loss": 0.1095,
      "num_input_tokens_seen": 8297424,
      "step": 4370
    },
    {
      "epoch": 0.7137613182151888,
      "grad_norm": 0.7502542734146118,
      "learning_rate": 4.853926303865618e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 8306512,
      "step": 4375
    },
    {
      "epoch": 0.7145770454360062,
      "grad_norm": 0.9016392827033997,
      "learning_rate": 4.853595454131441e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 8315152,
      "step": 4380
    },
    {
      "epoch": 0.7153927726568236,
      "grad_norm": 1.0554429292678833,
      "learning_rate": 4.8532642414432674e-05,
      "loss": 0.1215,
      "num_input_tokens_seen": 8324624,
      "step": 4385
    },
    {
      "epoch": 0.7162084998776409,
      "grad_norm": 0.2979889214038849,
      "learning_rate": 4.8529326658521754e-05,
      "loss": 0.0905,
      "num_input_tokens_seen": 8334528,
      "step": 4390
    },
    {
      "epoch": 0.7170242270984583,
      "grad_norm": 1.7082575559616089,
      "learning_rate": 4.8526007274092965e-05,
      "loss": 0.2646,
      "num_input_tokens_seen": 8343680,
      "step": 4395
    },
    {
      "epoch": 0.7178399543192756,
      "grad_norm": 1.8589950799942017,
      "learning_rate": 4.852268426165822e-05,
      "loss": 0.2057,
      "num_input_tokens_seen": 8353296,
      "step": 4400
    },
    {
      "epoch": 0.7178399543192756,
      "eval_loss": 0.16488176584243774,
      "eval_runtime": 133.1817,
      "eval_samples_per_second": 20.461,
      "eval_steps_per_second": 10.234,
      "num_input_tokens_seen": 8353296,
      "step": 4400
    },
    {
      "epoch": 0.718655681540093,
      "grad_norm": 2.764061689376831,
      "learning_rate": 4.851935762172995e-05,
      "loss": 0.1239,
      "num_input_tokens_seen": 8362112,
      "step": 4405
    },
    {
      "epoch": 0.7194714087609103,
      "grad_norm": 1.7433537244796753,
      "learning_rate": 4.8516027354821175e-05,
      "loss": 0.3637,
      "num_input_tokens_seen": 8372016,
      "step": 4410
    },
    {
      "epoch": 0.7202871359817277,
      "grad_norm": 0.4651854336261749,
      "learning_rate": 4.851269346144546e-05,
      "loss": 0.273,
      "num_input_tokens_seen": 8380784,
      "step": 4415
    },
    {
      "epoch": 0.721102863202545,
      "grad_norm": 2.5116047859191895,
      "learning_rate": 4.850935594211693e-05,
      "loss": 0.2441,
      "num_input_tokens_seen": 8389792,
      "step": 4420
    },
    {
      "epoch": 0.7219185904233625,
      "grad_norm": 0.3743191063404083,
      "learning_rate": 4.850601479735029e-05,
      "loss": 0.058,
      "num_input_tokens_seen": 8399072,
      "step": 4425
    },
    {
      "epoch": 0.7227343176441798,
      "grad_norm": 0.15206047892570496,
      "learning_rate": 4.850267002766076e-05,
      "loss": 0.0606,
      "num_input_tokens_seen": 8407824,
      "step": 4430
    },
    {
      "epoch": 0.7235500448649972,
      "grad_norm": 0.4369896948337555,
      "learning_rate": 4.849932163356417e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 8417648,
      "step": 4435
    },
    {
      "epoch": 0.7243657720858145,
      "grad_norm": 0.6203990578651428,
      "learning_rate": 4.8495969615576864e-05,
      "loss": 0.2246,
      "num_input_tokens_seen": 8427488,
      "step": 4440
    },
    {
      "epoch": 0.7251814993066319,
      "grad_norm": 0.056157782673835754,
      "learning_rate": 4.849261397421577e-05,
      "loss": 0.1019,
      "num_input_tokens_seen": 8437072,
      "step": 4445
    },
    {
      "epoch": 0.7259972265274492,
      "grad_norm": 0.6553892493247986,
      "learning_rate": 4.848925470999839e-05,
      "loss": 0.1301,
      "num_input_tokens_seen": 8446176,
      "step": 4450
    },
    {
      "epoch": 0.7268129537482666,
      "grad_norm": 1.2416876554489136,
      "learning_rate": 4.848589182344273e-05,
      "loss": 0.2348,
      "num_input_tokens_seen": 8455072,
      "step": 4455
    },
    {
      "epoch": 0.727628680969084,
      "grad_norm": 1.6935995817184448,
      "learning_rate": 4.848252531506742e-05,
      "loss": 0.2119,
      "num_input_tokens_seen": 8464224,
      "step": 4460
    },
    {
      "epoch": 0.7284444081899013,
      "grad_norm": 0.23933634161949158,
      "learning_rate": 4.847915518539161e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 8474992,
      "step": 4465
    },
    {
      "epoch": 0.7292601354107187,
      "grad_norm": 1.709952473640442,
      "learning_rate": 4.847578143493501e-05,
      "loss": 0.1181,
      "num_input_tokens_seen": 8485680,
      "step": 4470
    },
    {
      "epoch": 0.730075862631536,
      "grad_norm": 1.4522161483764648,
      "learning_rate": 4.847240406421789e-05,
      "loss": 0.2697,
      "num_input_tokens_seen": 8495184,
      "step": 4475
    },
    {
      "epoch": 0.7308915898523534,
      "grad_norm": 1.910836100578308,
      "learning_rate": 4.84690230737611e-05,
      "loss": 0.2892,
      "num_input_tokens_seen": 8505216,
      "step": 4480
    },
    {
      "epoch": 0.7317073170731707,
      "grad_norm": 0.2670787274837494,
      "learning_rate": 4.846563846408602e-05,
      "loss": 0.0722,
      "num_input_tokens_seen": 8514704,
      "step": 4485
    },
    {
      "epoch": 0.7325230442939881,
      "grad_norm": 1.7570483684539795,
      "learning_rate": 4.84622502357146e-05,
      "loss": 0.457,
      "num_input_tokens_seen": 8524320,
      "step": 4490
    },
    {
      "epoch": 0.7333387715148054,
      "grad_norm": 0.51678067445755,
      "learning_rate": 4.8458858389169345e-05,
      "loss": 0.0777,
      "num_input_tokens_seen": 8533744,
      "step": 4495
    },
    {
      "epoch": 0.7341544987356228,
      "grad_norm": 1.1287490129470825,
      "learning_rate": 4.8455462924973334e-05,
      "loss": 0.2124,
      "num_input_tokens_seen": 8542192,
      "step": 4500
    },
    {
      "epoch": 0.7349702259564401,
      "grad_norm": 1.5955466032028198,
      "learning_rate": 4.845206384365018e-05,
      "loss": 0.2204,
      "num_input_tokens_seen": 8551568,
      "step": 4505
    },
    {
      "epoch": 0.7357859531772575,
      "grad_norm": 0.2662765383720398,
      "learning_rate": 4.844866114572405e-05,
      "loss": 0.1709,
      "num_input_tokens_seen": 8559984,
      "step": 4510
    },
    {
      "epoch": 0.7366016803980748,
      "grad_norm": 1.6485036611557007,
      "learning_rate": 4.8445254831719706e-05,
      "loss": 0.2942,
      "num_input_tokens_seen": 8569504,
      "step": 4515
    },
    {
      "epoch": 0.7374174076188923,
      "grad_norm": 0.24039682745933533,
      "learning_rate": 4.8441844902162434e-05,
      "loss": 0.0871,
      "num_input_tokens_seen": 8577824,
      "step": 4520
    },
    {
      "epoch": 0.7382331348397096,
      "grad_norm": 0.5262110829353333,
      "learning_rate": 4.843843135757809e-05,
      "loss": 0.0636,
      "num_input_tokens_seen": 8588448,
      "step": 4525
    },
    {
      "epoch": 0.739048862060527,
      "grad_norm": 0.4354724586009979,
      "learning_rate": 4.843501419849308e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 8598384,
      "step": 4530
    },
    {
      "epoch": 0.7398645892813444,
      "grad_norm": 1.0522600412368774,
      "learning_rate": 4.8431593425434386e-05,
      "loss": 0.114,
      "num_input_tokens_seen": 8608864,
      "step": 4535
    },
    {
      "epoch": 0.7406803165021617,
      "grad_norm": 0.40594321489334106,
      "learning_rate": 4.8428169038929526e-05,
      "loss": 0.1038,
      "num_input_tokens_seen": 8618704,
      "step": 4540
    },
    {
      "epoch": 0.7414960437229791,
      "grad_norm": 2.105203151702881,
      "learning_rate": 4.8424741039506575e-05,
      "loss": 0.1424,
      "num_input_tokens_seen": 8627360,
      "step": 4545
    },
    {
      "epoch": 0.7423117709437964,
      "grad_norm": 0.48537248373031616,
      "learning_rate": 4.842130942769419e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 8637728,
      "step": 4550
    },
    {
      "epoch": 0.7431274981646138,
      "grad_norm": 0.6437379121780396,
      "learning_rate": 4.841787420402156e-05,
      "loss": 0.2754,
      "num_input_tokens_seen": 8646704,
      "step": 4555
    },
    {
      "epoch": 0.7439432253854311,
      "grad_norm": 0.34507814049720764,
      "learning_rate": 4.841443536901844e-05,
      "loss": 0.0958,
      "num_input_tokens_seen": 8656832,
      "step": 4560
    },
    {
      "epoch": 0.7447589526062485,
      "grad_norm": 0.4304993748664856,
      "learning_rate": 4.841099292321514e-05,
      "loss": 0.0989,
      "num_input_tokens_seen": 8665280,
      "step": 4565
    },
    {
      "epoch": 0.7455746798270658,
      "grad_norm": 1.419045329093933,
      "learning_rate": 4.8407546867142525e-05,
      "loss": 0.2203,
      "num_input_tokens_seen": 8674112,
      "step": 4570
    },
    {
      "epoch": 0.7463904070478832,
      "grad_norm": 1.5268312692642212,
      "learning_rate": 4.840409720133203e-05,
      "loss": 0.1961,
      "num_input_tokens_seen": 8683792,
      "step": 4575
    },
    {
      "epoch": 0.7472061342687005,
      "grad_norm": 0.4903491735458374,
      "learning_rate": 4.8400643926315634e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 8693280,
      "step": 4580
    },
    {
      "epoch": 0.7480218614895179,
      "grad_norm": 0.0776124969124794,
      "learning_rate": 4.839718704262587e-05,
      "loss": 0.1843,
      "num_input_tokens_seen": 8702752,
      "step": 4585
    },
    {
      "epoch": 0.7488375887103352,
      "grad_norm": 0.3611227869987488,
      "learning_rate": 4.839372655079585e-05,
      "loss": 0.0553,
      "num_input_tokens_seen": 8711728,
      "step": 4590
    },
    {
      "epoch": 0.7496533159311526,
      "grad_norm": 1.2200939655303955,
      "learning_rate": 4.83902624513592e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 8723024,
      "step": 4595
    },
    {
      "epoch": 0.7504690431519699,
      "grad_norm": 1.2349631786346436,
      "learning_rate": 4.838679474485014e-05,
      "loss": 0.1223,
      "num_input_tokens_seen": 8733232,
      "step": 4600
    },
    {
      "epoch": 0.7504690431519699,
      "eval_loss": 0.16062884032726288,
      "eval_runtime": 133.1564,
      "eval_samples_per_second": 20.465,
      "eval_steps_per_second": 10.236,
      "num_input_tokens_seen": 8733232,
      "step": 4600
    },
    {
      "epoch": 0.7512847703727873,
      "grad_norm": 1.382806420326233,
      "learning_rate": 4.838332343180343e-05,
      "loss": 0.1936,
      "num_input_tokens_seen": 8743008,
      "step": 4605
    },
    {
      "epoch": 0.7521004975936046,
      "grad_norm": 0.3487876057624817,
      "learning_rate": 4.83798485127544e-05,
      "loss": 0.0529,
      "num_input_tokens_seen": 8753232,
      "step": 4610
    },
    {
      "epoch": 0.7529162248144221,
      "grad_norm": 0.6141017079353333,
      "learning_rate": 4.837636998823892e-05,
      "loss": 0.0399,
      "num_input_tokens_seen": 8762240,
      "step": 4615
    },
    {
      "epoch": 0.7537319520352395,
      "grad_norm": 0.15347932279109955,
      "learning_rate": 4.8372887858793414e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 8770992,
      "step": 4620
    },
    {
      "epoch": 0.7545476792560568,
      "grad_norm": 1.2201181650161743,
      "learning_rate": 4.836940212495489e-05,
      "loss": 0.2917,
      "num_input_tokens_seen": 8780432,
      "step": 4625
    },
    {
      "epoch": 0.7553634064768742,
      "grad_norm": 0.8749929070472717,
      "learning_rate": 4.836591278726087e-05,
      "loss": 0.157,
      "num_input_tokens_seen": 8788800,
      "step": 4630
    },
    {
      "epoch": 0.7561791336976915,
      "grad_norm": 1.7049325704574585,
      "learning_rate": 4.836241984624947e-05,
      "loss": 0.1494,
      "num_input_tokens_seen": 8797840,
      "step": 4635
    },
    {
      "epoch": 0.7569948609185089,
      "grad_norm": 0.7847076058387756,
      "learning_rate": 4.8358923302459336e-05,
      "loss": 0.1372,
      "num_input_tokens_seen": 8807552,
      "step": 4640
    },
    {
      "epoch": 0.7578105881393262,
      "grad_norm": 0.26704367995262146,
      "learning_rate": 4.835542315642968e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 8817088,
      "step": 4645
    },
    {
      "epoch": 0.7586263153601436,
      "grad_norm": 1.2354627847671509,
      "learning_rate": 4.8351919408700274e-05,
      "loss": 0.2182,
      "num_input_tokens_seen": 8827328,
      "step": 4650
    },
    {
      "epoch": 0.7594420425809609,
      "grad_norm": 0.3296438157558441,
      "learning_rate": 4.834841205981144e-05,
      "loss": 0.217,
      "num_input_tokens_seen": 8836560,
      "step": 4655
    },
    {
      "epoch": 0.7602577698017783,
      "grad_norm": 3.027596950531006,
      "learning_rate": 4.8344901110304054e-05,
      "loss": 0.3298,
      "num_input_tokens_seen": 8845856,
      "step": 4660
    },
    {
      "epoch": 0.7610734970225956,
      "grad_norm": 0.14554128050804138,
      "learning_rate": 4.8341386560719534e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 8855840,
      "step": 4665
    },
    {
      "epoch": 0.761889224243413,
      "grad_norm": 0.6268901824951172,
      "learning_rate": 4.833786841159989e-05,
      "loss": 0.0585,
      "num_input_tokens_seen": 8866160,
      "step": 4670
    },
    {
      "epoch": 0.7627049514642303,
      "grad_norm": 1.612763524055481,
      "learning_rate": 4.833434666348765e-05,
      "loss": 0.283,
      "num_input_tokens_seen": 8875456,
      "step": 4675
    },
    {
      "epoch": 0.7635206786850477,
      "grad_norm": 0.6317954063415527,
      "learning_rate": 4.833082131692592e-05,
      "loss": 0.2131,
      "num_input_tokens_seen": 8884544,
      "step": 4680
    },
    {
      "epoch": 0.764336405905865,
      "grad_norm": 1.957848310470581,
      "learning_rate": 4.832729237245835e-05,
      "loss": 0.2016,
      "num_input_tokens_seen": 8893648,
      "step": 4685
    },
    {
      "epoch": 0.7651521331266824,
      "grad_norm": 0.9668093323707581,
      "learning_rate": 4.8323759830629145e-05,
      "loss": 0.1928,
      "num_input_tokens_seen": 8902144,
      "step": 4690
    },
    {
      "epoch": 0.7659678603474998,
      "grad_norm": 0.6503303050994873,
      "learning_rate": 4.8320223691983066e-05,
      "loss": 0.2073,
      "num_input_tokens_seen": 8912096,
      "step": 4695
    },
    {
      "epoch": 0.7667835875683171,
      "grad_norm": 0.047638315707445145,
      "learning_rate": 4.831668395706544e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 8922928,
      "step": 4700
    },
    {
      "epoch": 0.7675993147891346,
      "grad_norm": 0.16456271708011627,
      "learning_rate": 4.8313140626422125e-05,
      "loss": 0.0658,
      "num_input_tokens_seen": 8932576,
      "step": 4705
    },
    {
      "epoch": 0.7684150420099519,
      "grad_norm": 0.13372062146663666,
      "learning_rate": 4.830959370059956e-05,
      "loss": 0.0965,
      "num_input_tokens_seen": 8942048,
      "step": 4710
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.3164895474910736,
      "learning_rate": 4.830604318014472e-05,
      "loss": 0.0576,
      "num_input_tokens_seen": 8951296,
      "step": 4715
    },
    {
      "epoch": 0.7700464964515866,
      "grad_norm": 0.2912452220916748,
      "learning_rate": 4.830248906560514e-05,
      "loss": 0.0919,
      "num_input_tokens_seen": 8959968,
      "step": 4720
    },
    {
      "epoch": 0.770862223672404,
      "grad_norm": 0.5721297264099121,
      "learning_rate": 4.829893135752891e-05,
      "loss": 0.1467,
      "num_input_tokens_seen": 8969424,
      "step": 4725
    },
    {
      "epoch": 0.7716779508932213,
      "grad_norm": 0.2681201100349426,
      "learning_rate": 4.829537005646466e-05,
      "loss": 0.1562,
      "num_input_tokens_seen": 8979248,
      "step": 4730
    },
    {
      "epoch": 0.7724936781140387,
      "grad_norm": 0.434024840593338,
      "learning_rate": 4.8291805162961615e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 8989760,
      "step": 4735
    },
    {
      "epoch": 0.773309405334856,
      "grad_norm": 0.9243844747543335,
      "learning_rate": 4.82882366775695e-05,
      "loss": 0.1452,
      "num_input_tokens_seen": 8997920,
      "step": 4740
    },
    {
      "epoch": 0.7741251325556734,
      "grad_norm": 2.102956533432007,
      "learning_rate": 4.828466460083864e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 9007488,
      "step": 4745
    },
    {
      "epoch": 0.7749408597764907,
      "grad_norm": 0.1778687983751297,
      "learning_rate": 4.8281088933319877e-05,
      "loss": 0.0783,
      "num_input_tokens_seen": 9016576,
      "step": 4750
    },
    {
      "epoch": 0.7757565869973081,
      "grad_norm": 0.5692369341850281,
      "learning_rate": 4.827750967556464e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 9026656,
      "step": 4755
    },
    {
      "epoch": 0.7765723142181254,
      "grad_norm": 0.17029385268688202,
      "learning_rate": 4.827392682812488e-05,
      "loss": 0.0993,
      "num_input_tokens_seen": 9035872,
      "step": 4760
    },
    {
      "epoch": 0.7773880414389428,
      "grad_norm": 0.16910982131958008,
      "learning_rate": 4.827034039155312e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 9046240,
      "step": 4765
    },
    {
      "epoch": 0.7782037686597602,
      "grad_norm": 0.2726154923439026,
      "learning_rate": 4.8266750366402445e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 9056128,
      "step": 4770
    },
    {
      "epoch": 0.7790194958805775,
      "grad_norm": 2.138503074645996,
      "learning_rate": 4.8263156753226476e-05,
      "loss": 0.2812,
      "num_input_tokens_seen": 9064368,
      "step": 4775
    },
    {
      "epoch": 0.7798352231013949,
      "grad_norm": 1.19856595993042,
      "learning_rate": 4.8259559552579394e-05,
      "loss": 0.2262,
      "num_input_tokens_seen": 9074048,
      "step": 4780
    },
    {
      "epoch": 0.7806509503222122,
      "grad_norm": 0.7976983189582825,
      "learning_rate": 4.825595876501593e-05,
      "loss": 0.1593,
      "num_input_tokens_seen": 9084176,
      "step": 4785
    },
    {
      "epoch": 0.7814666775430296,
      "grad_norm": 1.153951644897461,
      "learning_rate": 4.825235439109137e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 9094016,
      "step": 4790
    },
    {
      "epoch": 0.7822824047638469,
      "grad_norm": 1.403702974319458,
      "learning_rate": 4.824874643136156e-05,
      "loss": 0.1216,
      "num_input_tokens_seen": 9103680,
      "step": 4795
    },
    {
      "epoch": 0.7830981319846644,
      "grad_norm": 1.8241143226623535,
      "learning_rate": 4.824513488638288e-05,
      "loss": 0.2777,
      "num_input_tokens_seen": 9113632,
      "step": 4800
    },
    {
      "epoch": 0.7830981319846644,
      "eval_loss": 0.15930184721946716,
      "eval_runtime": 133.2329,
      "eval_samples_per_second": 20.453,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 9113632,
      "step": 4800
    },
    {
      "epoch": 0.7839138592054817,
      "grad_norm": 1.1145846843719482,
      "learning_rate": 4.8241519756712293e-05,
      "loss": 0.3214,
      "num_input_tokens_seen": 9122192,
      "step": 4805
    },
    {
      "epoch": 0.7847295864262991,
      "grad_norm": 0.743476927280426,
      "learning_rate": 4.8237901042907285e-05,
      "loss": 0.1839,
      "num_input_tokens_seen": 9132512,
      "step": 4810
    },
    {
      "epoch": 0.7855453136471164,
      "grad_norm": 0.6610101461410522,
      "learning_rate": 4.823427874552591e-05,
      "loss": 0.0502,
      "num_input_tokens_seen": 9141264,
      "step": 4815
    },
    {
      "epoch": 0.7863610408679338,
      "grad_norm": 0.6163780093193054,
      "learning_rate": 4.823065286512677e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 9150160,
      "step": 4820
    },
    {
      "epoch": 0.7871767680887511,
      "grad_norm": 0.6705014109611511,
      "learning_rate": 4.8227023402269025e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 9160096,
      "step": 4825
    },
    {
      "epoch": 0.7879924953095685,
      "grad_norm": 0.5862709879875183,
      "learning_rate": 4.822339035751239e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 9168864,
      "step": 4830
    },
    {
      "epoch": 0.7888082225303858,
      "grad_norm": 0.09982135146856308,
      "learning_rate": 4.8219753731417104e-05,
      "loss": 0.0673,
      "num_input_tokens_seen": 9178608,
      "step": 4835
    },
    {
      "epoch": 0.7896239497512032,
      "grad_norm": 1.3063241243362427,
      "learning_rate": 4.821611352454401e-05,
      "loss": 0.1993,
      "num_input_tokens_seen": 9187824,
      "step": 4840
    },
    {
      "epoch": 0.7904396769720206,
      "grad_norm": 2.3094232082366943,
      "learning_rate": 4.8212469737454444e-05,
      "loss": 0.2564,
      "num_input_tokens_seen": 9196464,
      "step": 4845
    },
    {
      "epoch": 0.7912554041928379,
      "grad_norm": 1.0153855085372925,
      "learning_rate": 4.820882237071035e-05,
      "loss": 0.2913,
      "num_input_tokens_seen": 9206544,
      "step": 4850
    },
    {
      "epoch": 0.7920711314136553,
      "grad_norm": 0.34004220366477966,
      "learning_rate": 4.820517142487417e-05,
      "loss": 0.1107,
      "num_input_tokens_seen": 9214288,
      "step": 4855
    },
    {
      "epoch": 0.7928868586344726,
      "grad_norm": 0.12369544804096222,
      "learning_rate": 4.8201516900508956e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 9222496,
      "step": 4860
    },
    {
      "epoch": 0.79370258585529,
      "grad_norm": 1.0425060987472534,
      "learning_rate": 4.819785879817827e-05,
      "loss": 0.2527,
      "num_input_tokens_seen": 9232480,
      "step": 4865
    },
    {
      "epoch": 0.7945183130761073,
      "grad_norm": 0.7983195781707764,
      "learning_rate": 4.8194197118446226e-05,
      "loss": 0.0582,
      "num_input_tokens_seen": 9241408,
      "step": 4870
    },
    {
      "epoch": 0.7953340402969247,
      "grad_norm": 1.1822229623794556,
      "learning_rate": 4.819053186187752e-05,
      "loss": 0.1232,
      "num_input_tokens_seen": 9250880,
      "step": 4875
    },
    {
      "epoch": 0.796149767517742,
      "grad_norm": 0.6311345100402832,
      "learning_rate": 4.818686302903736e-05,
      "loss": 0.2107,
      "num_input_tokens_seen": 9259808,
      "step": 4880
    },
    {
      "epoch": 0.7969654947385594,
      "grad_norm": 0.2976803183555603,
      "learning_rate": 4.818319062049154e-05,
      "loss": 0.0718,
      "num_input_tokens_seen": 9269552,
      "step": 4885
    },
    {
      "epoch": 0.7977812219593767,
      "grad_norm": 0.9641924500465393,
      "learning_rate": 4.817951463680639e-05,
      "loss": 0.2485,
      "num_input_tokens_seen": 9279104,
      "step": 4890
    },
    {
      "epoch": 0.7985969491801942,
      "grad_norm": 0.06866971403360367,
      "learning_rate": 4.817583507854879e-05,
      "loss": 0.2175,
      "num_input_tokens_seen": 9288944,
      "step": 4895
    },
    {
      "epoch": 0.7994126764010115,
      "grad_norm": 1.3213670253753662,
      "learning_rate": 4.817215194628617e-05,
      "loss": 0.127,
      "num_input_tokens_seen": 9297504,
      "step": 4900
    },
    {
      "epoch": 0.8002284036218289,
      "grad_norm": 0.5379509329795837,
      "learning_rate": 4.816846524058653e-05,
      "loss": 0.1209,
      "num_input_tokens_seen": 9307856,
      "step": 4905
    },
    {
      "epoch": 0.8010441308426463,
      "grad_norm": 1.186208963394165,
      "learning_rate": 4.816477496201839e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 9316432,
      "step": 4910
    },
    {
      "epoch": 0.8018598580634636,
      "grad_norm": 1.4451849460601807,
      "learning_rate": 4.8161081111150845e-05,
      "loss": 0.1794,
      "num_input_tokens_seen": 9326016,
      "step": 4915
    },
    {
      "epoch": 0.802675585284281,
      "grad_norm": 0.6029987931251526,
      "learning_rate": 4.815738368855354e-05,
      "loss": 0.0828,
      "num_input_tokens_seen": 9333776,
      "step": 4920
    },
    {
      "epoch": 0.8034913125050983,
      "grad_norm": 0.2790224552154541,
      "learning_rate": 4.815368269479664e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 9342640,
      "step": 4925
    },
    {
      "epoch": 0.8043070397259157,
      "grad_norm": 0.32002437114715576,
      "learning_rate": 4.814997813045092e-05,
      "loss": 0.1195,
      "num_input_tokens_seen": 9352112,
      "step": 4930
    },
    {
      "epoch": 0.805122766946733,
      "grad_norm": 1.941563606262207,
      "learning_rate": 4.814626999608764e-05,
      "loss": 0.291,
      "num_input_tokens_seen": 9361088,
      "step": 4935
    },
    {
      "epoch": 0.8059384941675504,
      "grad_norm": 1.1033622026443481,
      "learning_rate": 4.814255829227865e-05,
      "loss": 0.1868,
      "num_input_tokens_seen": 9370768,
      "step": 4940
    },
    {
      "epoch": 0.8067542213883677,
      "grad_norm": 1.2346739768981934,
      "learning_rate": 4.813884301959635e-05,
      "loss": 0.2154,
      "num_input_tokens_seen": 9381104,
      "step": 4945
    },
    {
      "epoch": 0.8075699486091851,
      "grad_norm": 0.34602928161621094,
      "learning_rate": 4.813512417861368e-05,
      "loss": 0.075,
      "num_input_tokens_seen": 9391568,
      "step": 4950
    },
    {
      "epoch": 0.8083856758300024,
      "grad_norm": 0.09541311860084534,
      "learning_rate": 4.813140176990411e-05,
      "loss": 0.1055,
      "num_input_tokens_seen": 9401568,
      "step": 4955
    },
    {
      "epoch": 0.8092014030508198,
      "grad_norm": 0.30739277601242065,
      "learning_rate": 4.8127675794041714e-05,
      "loss": 0.0805,
      "num_input_tokens_seen": 9409248,
      "step": 4960
    },
    {
      "epoch": 0.8100171302716371,
      "grad_norm": 0.9042321443557739,
      "learning_rate": 4.812394625160107e-05,
      "loss": 0.1302,
      "num_input_tokens_seen": 9418944,
      "step": 4965
    },
    {
      "epoch": 0.8108328574924545,
      "grad_norm": 0.11287104338407516,
      "learning_rate": 4.812021314315732e-05,
      "loss": 0.1933,
      "num_input_tokens_seen": 9429168,
      "step": 4970
    },
    {
      "epoch": 0.8116485847132718,
      "grad_norm": 0.8007352948188782,
      "learning_rate": 4.811647646928616e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 9440128,
      "step": 4975
    },
    {
      "epoch": 0.8124643119340892,
      "grad_norm": 1.1883207559585571,
      "learning_rate": 4.8112736230563814e-05,
      "loss": 0.1025,
      "num_input_tokens_seen": 9449712,
      "step": 4980
    },
    {
      "epoch": 0.8132800391549067,
      "grad_norm": 1.6142244338989258,
      "learning_rate": 4.81089924275671e-05,
      "loss": 0.184,
      "num_input_tokens_seen": 9459376,
      "step": 4985
    },
    {
      "epoch": 0.814095766375724,
      "grad_norm": 1.1496920585632324,
      "learning_rate": 4.810524506087335e-05,
      "loss": 0.193,
      "num_input_tokens_seen": 9469520,
      "step": 4990
    },
    {
      "epoch": 0.8149114935965414,
      "grad_norm": 1.3937302827835083,
      "learning_rate": 4.810149413106044e-05,
      "loss": 0.1394,
      "num_input_tokens_seen": 9478032,
      "step": 4995
    },
    {
      "epoch": 0.8157272208173587,
      "grad_norm": 0.7058119177818298,
      "learning_rate": 4.809773963870684e-05,
      "loss": 0.0492,
      "num_input_tokens_seen": 9487952,
      "step": 5000
    },
    {
      "epoch": 0.8157272208173587,
      "eval_loss": 0.16420899331569672,
      "eval_runtime": 133.3082,
      "eval_samples_per_second": 20.441,
      "eval_steps_per_second": 10.224,
      "num_input_tokens_seen": 9487952,
      "step": 5000
    },
    {
      "epoch": 0.8165429480381761,
      "grad_norm": 0.3949891924858093,
      "learning_rate": 4.809398158439151e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 9497568,
      "step": 5005
    },
    {
      "epoch": 0.8173586752589934,
      "grad_norm": 0.5863240957260132,
      "learning_rate": 4.8090219968694005e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 9507776,
      "step": 5010
    },
    {
      "epoch": 0.8181744024798108,
      "grad_norm": 0.08193662017583847,
      "learning_rate": 4.808645479219442e-05,
      "loss": 0.1888,
      "num_input_tokens_seen": 9516320,
      "step": 5015
    },
    {
      "epoch": 0.8189901297006281,
      "grad_norm": 0.06045374274253845,
      "learning_rate": 4.8082686055473375e-05,
      "loss": 0.0407,
      "num_input_tokens_seen": 9526640,
      "step": 5020
    },
    {
      "epoch": 0.8198058569214455,
      "grad_norm": 0.5293352007865906,
      "learning_rate": 4.8078913759112066e-05,
      "loss": 0.0663,
      "num_input_tokens_seen": 9535952,
      "step": 5025
    },
    {
      "epoch": 0.8206215841422628,
      "grad_norm": 0.3707755208015442,
      "learning_rate": 4.807513790369223e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 9545328,
      "step": 5030
    },
    {
      "epoch": 0.8214373113630802,
      "grad_norm": 1.7615561485290527,
      "learning_rate": 4.8071358489796145e-05,
      "loss": 0.2022,
      "num_input_tokens_seen": 9555520,
      "step": 5035
    },
    {
      "epoch": 0.8222530385838975,
      "grad_norm": 0.9478245377540588,
      "learning_rate": 4.806757551800665e-05,
      "loss": 0.2107,
      "num_input_tokens_seen": 9564752,
      "step": 5040
    },
    {
      "epoch": 0.8230687658047149,
      "grad_norm": 0.9166566729545593,
      "learning_rate": 4.806378898890713e-05,
      "loss": 0.1289,
      "num_input_tokens_seen": 9573680,
      "step": 5045
    },
    {
      "epoch": 0.8238844930255322,
      "grad_norm": 0.8851613998413086,
      "learning_rate": 4.80599989030815e-05,
      "loss": 0.1087,
      "num_input_tokens_seen": 9583072,
      "step": 5050
    },
    {
      "epoch": 0.8247002202463496,
      "grad_norm": 0.3896315097808838,
      "learning_rate": 4.805620526111426e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 9592816,
      "step": 5055
    },
    {
      "epoch": 0.8255159474671669,
      "grad_norm": 1.0497361421585083,
      "learning_rate": 4.805240806359042e-05,
      "loss": 0.2529,
      "num_input_tokens_seen": 9602000,
      "step": 5060
    },
    {
      "epoch": 0.8263316746879843,
      "grad_norm": 2.222187042236328,
      "learning_rate": 4.804860731109557e-05,
      "loss": 0.1965,
      "num_input_tokens_seen": 9610176,
      "step": 5065
    },
    {
      "epoch": 0.8271474019088016,
      "grad_norm": 1.1587998867034912,
      "learning_rate": 4.804480300421581e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 9618752,
      "step": 5070
    },
    {
      "epoch": 0.827963129129619,
      "grad_norm": 0.4542011320590973,
      "learning_rate": 4.804099514353784e-05,
      "loss": 0.1238,
      "num_input_tokens_seen": 9628128,
      "step": 5075
    },
    {
      "epoch": 0.8287788563504365,
      "grad_norm": 0.056272972375154495,
      "learning_rate": 4.8037183729648867e-05,
      "loss": 0.0865,
      "num_input_tokens_seen": 9636960,
      "step": 5080
    },
    {
      "epoch": 0.8295945835712538,
      "grad_norm": 1.6419249773025513,
      "learning_rate": 4.803336876313666e-05,
      "loss": 0.1365,
      "num_input_tokens_seen": 9646928,
      "step": 5085
    },
    {
      "epoch": 0.8304103107920712,
      "grad_norm": 0.830941915512085,
      "learning_rate": 4.802955024458953e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 9655216,
      "step": 5090
    },
    {
      "epoch": 0.8312260380128885,
      "grad_norm": 0.18344920873641968,
      "learning_rate": 4.802572817459634e-05,
      "loss": 0.2023,
      "num_input_tokens_seen": 9664544,
      "step": 5095
    },
    {
      "epoch": 0.8320417652337059,
      "grad_norm": 0.1557053178548813,
      "learning_rate": 4.802190255374651e-05,
      "loss": 0.1008,
      "num_input_tokens_seen": 9673856,
      "step": 5100
    },
    {
      "epoch": 0.8328574924545232,
      "grad_norm": 1.9543811082839966,
      "learning_rate": 4.801807338263e-05,
      "loss": 0.1959,
      "num_input_tokens_seen": 9683440,
      "step": 5105
    },
    {
      "epoch": 0.8336732196753406,
      "grad_norm": 0.7518640160560608,
      "learning_rate": 4.8014240661837306e-05,
      "loss": 0.0996,
      "num_input_tokens_seen": 9693392,
      "step": 5110
    },
    {
      "epoch": 0.8344889468961579,
      "grad_norm": 0.6097130179405212,
      "learning_rate": 4.80104043919595e-05,
      "loss": 0.1963,
      "num_input_tokens_seen": 9703104,
      "step": 5115
    },
    {
      "epoch": 0.8353046741169753,
      "grad_norm": 1.966873049736023,
      "learning_rate": 4.800656457358815e-05,
      "loss": 0.2155,
      "num_input_tokens_seen": 9711760,
      "step": 5120
    },
    {
      "epoch": 0.8361204013377926,
      "grad_norm": 0.6476613879203796,
      "learning_rate": 4.800272120731544e-05,
      "loss": 0.1071,
      "num_input_tokens_seen": 9721072,
      "step": 5125
    },
    {
      "epoch": 0.83693612855861,
      "grad_norm": 0.22035574913024902,
      "learning_rate": 4.799887429373404e-05,
      "loss": 0.2932,
      "num_input_tokens_seen": 9730672,
      "step": 5130
    },
    {
      "epoch": 0.8377518557794273,
      "grad_norm": 0.2853173017501831,
      "learning_rate": 4.79950238334372e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 9740208,
      "step": 5135
    },
    {
      "epoch": 0.8385675830002447,
      "grad_norm": 0.3176376521587372,
      "learning_rate": 4.799116982701872e-05,
      "loss": 0.0725,
      "num_input_tokens_seen": 9749616,
      "step": 5140
    },
    {
      "epoch": 0.839383310221062,
      "grad_norm": 1.241943597793579,
      "learning_rate": 4.7987312275072926e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 9757760,
      "step": 5145
    },
    {
      "epoch": 0.8401990374418794,
      "grad_norm": 0.1247173473238945,
      "learning_rate": 4.79834511781947e-05,
      "loss": 0.157,
      "num_input_tokens_seen": 9766016,
      "step": 5150
    },
    {
      "epoch": 0.8410147646626968,
      "grad_norm": 0.23184657096862793,
      "learning_rate": 4.797958653697947e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 9774112,
      "step": 5155
    },
    {
      "epoch": 0.8418304918835141,
      "grad_norm": 1.3132284879684448,
      "learning_rate": 4.7975718352023225e-05,
      "loss": 0.084,
      "num_input_tokens_seen": 9784512,
      "step": 5160
    },
    {
      "epoch": 0.8426462191043315,
      "grad_norm": 0.42378783226013184,
      "learning_rate": 4.7971846623922476e-05,
      "loss": 0.0948,
      "num_input_tokens_seen": 9794288,
      "step": 5165
    },
    {
      "epoch": 0.8434619463251488,
      "grad_norm": 0.37518128752708435,
      "learning_rate": 4.7967971353274294e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 9802848,
      "step": 5170
    },
    {
      "epoch": 0.8442776735459663,
      "grad_norm": 0.22992366552352905,
      "learning_rate": 4.79640925406763e-05,
      "loss": 0.1222,
      "num_input_tokens_seen": 9812720,
      "step": 5175
    },
    {
      "epoch": 0.8450934007667836,
      "grad_norm": 0.7130159139633179,
      "learning_rate": 4.796021018672664e-05,
      "loss": 0.2195,
      "num_input_tokens_seen": 9822928,
      "step": 5180
    },
    {
      "epoch": 0.845909127987601,
      "grad_norm": 0.913700520992279,
      "learning_rate": 4.795632429202405e-05,
      "loss": 0.3357,
      "num_input_tokens_seen": 9832176,
      "step": 5185
    },
    {
      "epoch": 0.8467248552084183,
      "grad_norm": 0.41719406843185425,
      "learning_rate": 4.795243485716775e-05,
      "loss": 0.1758,
      "num_input_tokens_seen": 9842352,
      "step": 5190
    },
    {
      "epoch": 0.8475405824292357,
      "grad_norm": 0.1331203430891037,
      "learning_rate": 4.794854188275757e-05,
      "loss": 0.1786,
      "num_input_tokens_seen": 9852448,
      "step": 5195
    },
    {
      "epoch": 0.848356309650053,
      "grad_norm": 0.763278067111969,
      "learning_rate": 4.794464536939384e-05,
      "loss": 0.1335,
      "num_input_tokens_seen": 9861104,
      "step": 5200
    },
    {
      "epoch": 0.848356309650053,
      "eval_loss": 0.16007471084594727,
      "eval_runtime": 133.1928,
      "eval_samples_per_second": 20.459,
      "eval_steps_per_second": 10.233,
      "num_input_tokens_seen": 9861104,
      "step": 5200
    },
    {
      "epoch": 0.8491720368708704,
      "grad_norm": 0.7138174176216125,
      "learning_rate": 4.794074531767745e-05,
      "loss": 0.0961,
      "num_input_tokens_seen": 9870608,
      "step": 5205
    },
    {
      "epoch": 0.8499877640916877,
      "grad_norm": 0.2969825565814972,
      "learning_rate": 4.7936841728209834e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 9878880,
      "step": 5210
    },
    {
      "epoch": 0.8508034913125051,
      "grad_norm": 0.06388988345861435,
      "learning_rate": 4.7932934601593e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 9887856,
      "step": 5215
    },
    {
      "epoch": 0.8516192185333225,
      "grad_norm": 0.4148397147655487,
      "learning_rate": 4.792902393842943e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 9898224,
      "step": 5220
    },
    {
      "epoch": 0.8524349457541398,
      "grad_norm": 1.3452162742614746,
      "learning_rate": 4.792510973932225e-05,
      "loss": 0.1084,
      "num_input_tokens_seen": 9906864,
      "step": 5225
    },
    {
      "epoch": 0.8532506729749572,
      "grad_norm": 0.3172058165073395,
      "learning_rate": 4.7921192004875036e-05,
      "loss": 0.2833,
      "num_input_tokens_seen": 9916544,
      "step": 5230
    },
    {
      "epoch": 0.8540664001957745,
      "grad_norm": 0.6083715558052063,
      "learning_rate": 4.791727073569198e-05,
      "loss": 0.2843,
      "num_input_tokens_seen": 9925456,
      "step": 5235
    },
    {
      "epoch": 0.8548821274165919,
      "grad_norm": 0.10611522942781448,
      "learning_rate": 4.7913345932377775e-05,
      "loss": 0.0575,
      "num_input_tokens_seen": 9934224,
      "step": 5240
    },
    {
      "epoch": 0.8556978546374092,
      "grad_norm": 0.03711768239736557,
      "learning_rate": 4.790941759553769e-05,
      "loss": 0.1162,
      "num_input_tokens_seen": 9943072,
      "step": 5245
    },
    {
      "epoch": 0.8565135818582266,
      "grad_norm": 1.5961629152297974,
      "learning_rate": 4.79054857257775e-05,
      "loss": 0.2455,
      "num_input_tokens_seen": 9953472,
      "step": 5250
    },
    {
      "epoch": 0.8573293090790439,
      "grad_norm": 0.18051421642303467,
      "learning_rate": 4.790155032370357e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 9961792,
      "step": 5255
    },
    {
      "epoch": 0.8581450362998613,
      "grad_norm": 1.362596869468689,
      "learning_rate": 4.789761138992278e-05,
      "loss": 0.1993,
      "num_input_tokens_seen": 9971040,
      "step": 5260
    },
    {
      "epoch": 0.8589607635206787,
      "grad_norm": 1.0200001001358032,
      "learning_rate": 4.7893668925042565e-05,
      "loss": 0.0676,
      "num_input_tokens_seen": 9980832,
      "step": 5265
    },
    {
      "epoch": 0.8597764907414961,
      "grad_norm": 0.17819049954414368,
      "learning_rate": 4.78897229296709e-05,
      "loss": 0.3332,
      "num_input_tokens_seen": 9990368,
      "step": 5270
    },
    {
      "epoch": 0.8605922179623134,
      "grad_norm": 0.10441245138645172,
      "learning_rate": 4.7885773404416315e-05,
      "loss": 0.1758,
      "num_input_tokens_seen": 9998544,
      "step": 5275
    },
    {
      "epoch": 0.8614079451831308,
      "grad_norm": 0.30139976739883423,
      "learning_rate": 4.788182034988786e-05,
      "loss": 0.0487,
      "num_input_tokens_seen": 10008640,
      "step": 5280
    },
    {
      "epoch": 0.8622236724039481,
      "grad_norm": 1.9587390422821045,
      "learning_rate": 4.787786376669516e-05,
      "loss": 0.229,
      "num_input_tokens_seen": 10018336,
      "step": 5285
    },
    {
      "epoch": 0.8630393996247655,
      "grad_norm": 0.4186311960220337,
      "learning_rate": 4.787390365544837e-05,
      "loss": 0.0691,
      "num_input_tokens_seen": 10028464,
      "step": 5290
    },
    {
      "epoch": 0.8638551268455829,
      "grad_norm": 0.05751468986272812,
      "learning_rate": 4.786994001675818e-05,
      "loss": 0.1992,
      "num_input_tokens_seen": 10038560,
      "step": 5295
    },
    {
      "epoch": 0.8646708540664002,
      "grad_norm": 0.5762830376625061,
      "learning_rate": 4.786597285123584e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 10046560,
      "step": 5300
    },
    {
      "epoch": 0.8654865812872176,
      "grad_norm": 0.6729532480239868,
      "learning_rate": 4.7862002159493135e-05,
      "loss": 0.2057,
      "num_input_tokens_seen": 10055696,
      "step": 5305
    },
    {
      "epoch": 0.8663023085080349,
      "grad_norm": 1.0750205516815186,
      "learning_rate": 4.785802794214239e-05,
      "loss": 0.1405,
      "num_input_tokens_seen": 10065136,
      "step": 5310
    },
    {
      "epoch": 0.8671180357288523,
      "grad_norm": 1.0948188304901123,
      "learning_rate": 4.7854050199796495e-05,
      "loss": 0.1698,
      "num_input_tokens_seen": 10075824,
      "step": 5315
    },
    {
      "epoch": 0.8679337629496696,
      "grad_norm": 0.3878529369831085,
      "learning_rate": 4.7850068933068845e-05,
      "loss": 0.1318,
      "num_input_tokens_seen": 10085712,
      "step": 5320
    },
    {
      "epoch": 0.868749490170487,
      "grad_norm": 0.8942821025848389,
      "learning_rate": 4.7846084142573425e-05,
      "loss": 0.0728,
      "num_input_tokens_seen": 10096272,
      "step": 5325
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 1.561762809753418,
      "learning_rate": 4.7842095828924725e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 10104864,
      "step": 5330
    },
    {
      "epoch": 0.8703809446121217,
      "grad_norm": 0.8265649676322937,
      "learning_rate": 4.783810399273779e-05,
      "loss": 0.0887,
      "num_input_tokens_seen": 10114656,
      "step": 5335
    },
    {
      "epoch": 0.871196671832939,
      "grad_norm": 1.5199567079544067,
      "learning_rate": 4.7834108634628226e-05,
      "loss": 0.2986,
      "num_input_tokens_seen": 10124288,
      "step": 5340
    },
    {
      "epoch": 0.8720123990537564,
      "grad_norm": 0.5957947373390198,
      "learning_rate": 4.783010975521216e-05,
      "loss": 0.1053,
      "num_input_tokens_seen": 10134224,
      "step": 5345
    },
    {
      "epoch": 0.8728281262745737,
      "grad_norm": 0.49811381101608276,
      "learning_rate": 4.782610735510626e-05,
      "loss": 0.1275,
      "num_input_tokens_seen": 10143840,
      "step": 5350
    },
    {
      "epoch": 0.8736438534953911,
      "grad_norm": 0.6986739039421082,
      "learning_rate": 4.782210143492776e-05,
      "loss": 0.2287,
      "num_input_tokens_seen": 10153296,
      "step": 5355
    },
    {
      "epoch": 0.8744595807162086,
      "grad_norm": 0.17363399267196655,
      "learning_rate": 4.781809199529442e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 10164160,
      "step": 5360
    },
    {
      "epoch": 0.8752753079370259,
      "grad_norm": 0.14009331166744232,
      "learning_rate": 4.781407903682454e-05,
      "loss": 0.0991,
      "num_input_tokens_seen": 10174336,
      "step": 5365
    },
    {
      "epoch": 0.8760910351578433,
      "grad_norm": 0.5723888278007507,
      "learning_rate": 4.781006256013698e-05,
      "loss": 0.0953,
      "num_input_tokens_seen": 10184640,
      "step": 5370
    },
    {
      "epoch": 0.8769067623786606,
      "grad_norm": 0.13459061086177826,
      "learning_rate": 4.7806042565851115e-05,
      "loss": 0.0921,
      "num_input_tokens_seen": 10194080,
      "step": 5375
    },
    {
      "epoch": 0.877722489599478,
      "grad_norm": 1.5484014749526978,
      "learning_rate": 4.7802019054586895e-05,
      "loss": 0.1874,
      "num_input_tokens_seen": 10203200,
      "step": 5380
    },
    {
      "epoch": 0.8785382168202953,
      "grad_norm": 0.3649432957172394,
      "learning_rate": 4.779799202696479e-05,
      "loss": 0.1593,
      "num_input_tokens_seen": 10211792,
      "step": 5385
    },
    {
      "epoch": 0.8793539440411127,
      "grad_norm": 0.7651509046554565,
      "learning_rate": 4.779396148360581e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 10221712,
      "step": 5390
    },
    {
      "epoch": 0.88016967126193,
      "grad_norm": 0.268216997385025,
      "learning_rate": 4.7789927425131517e-05,
      "loss": 0.081,
      "num_input_tokens_seen": 10230400,
      "step": 5395
    },
    {
      "epoch": 0.8809853984827474,
      "grad_norm": 0.23631741106510162,
      "learning_rate": 4.778588985216403e-05,
      "loss": 0.1896,
      "num_input_tokens_seen": 10239088,
      "step": 5400
    },
    {
      "epoch": 0.8809853984827474,
      "eval_loss": 0.1586613655090332,
      "eval_runtime": 133.2257,
      "eval_samples_per_second": 20.454,
      "eval_steps_per_second": 10.231,
      "num_input_tokens_seen": 10239088,
      "step": 5400
    },
    {
      "epoch": 0.8818011257035647,
      "grad_norm": 1.0001271963119507,
      "learning_rate": 4.778184876532598e-05,
      "loss": 0.177,
      "num_input_tokens_seen": 10249088,
      "step": 5405
    },
    {
      "epoch": 0.8826168529243821,
      "grad_norm": 0.4450065791606903,
      "learning_rate": 4.7777804165240556e-05,
      "loss": 0.0899,
      "num_input_tokens_seen": 10257088,
      "step": 5410
    },
    {
      "epoch": 0.8834325801451994,
      "grad_norm": 0.3195333480834961,
      "learning_rate": 4.7773756052531485e-05,
      "loss": 0.1953,
      "num_input_tokens_seen": 10266992,
      "step": 5415
    },
    {
      "epoch": 0.8842483073660168,
      "grad_norm": 0.4520282745361328,
      "learning_rate": 4.7769704427823035e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 10276000,
      "step": 5420
    },
    {
      "epoch": 0.8850640345868341,
      "grad_norm": 0.30608800053596497,
      "learning_rate": 4.776564929174003e-05,
      "loss": 0.0551,
      "num_input_tokens_seen": 10285024,
      "step": 5425
    },
    {
      "epoch": 0.8858797618076515,
      "grad_norm": 0.28260430693626404,
      "learning_rate": 4.7761590644907806e-05,
      "loss": 0.1389,
      "num_input_tokens_seen": 10295472,
      "step": 5430
    },
    {
      "epoch": 0.8866954890284688,
      "grad_norm": 0.9211229681968689,
      "learning_rate": 4.7757528487952263e-05,
      "loss": 0.2089,
      "num_input_tokens_seen": 10304896,
      "step": 5435
    },
    {
      "epoch": 0.8875112162492862,
      "grad_norm": 1.4005506038665771,
      "learning_rate": 4.7753462821499836e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 10314560,
      "step": 5440
    },
    {
      "epoch": 0.8883269434701035,
      "grad_norm": 0.20772601664066315,
      "learning_rate": 4.774939364617751e-05,
      "loss": 0.243,
      "num_input_tokens_seen": 10324064,
      "step": 5445
    },
    {
      "epoch": 0.889142670690921,
      "grad_norm": 1.4187809228897095,
      "learning_rate": 4.7745320962612795e-05,
      "loss": 0.2348,
      "num_input_tokens_seen": 10333968,
      "step": 5450
    },
    {
      "epoch": 0.8899583979117384,
      "grad_norm": 2.884138822555542,
      "learning_rate": 4.7741244771433756e-05,
      "loss": 0.3656,
      "num_input_tokens_seen": 10342832,
      "step": 5455
    },
    {
      "epoch": 0.8907741251325557,
      "grad_norm": 0.11634336411952972,
      "learning_rate": 4.7737165073268985e-05,
      "loss": 0.2865,
      "num_input_tokens_seen": 10352384,
      "step": 5460
    },
    {
      "epoch": 0.8915898523533731,
      "grad_norm": 0.22534959018230438,
      "learning_rate": 4.7733081868747626e-05,
      "loss": 0.1024,
      "num_input_tokens_seen": 10360656,
      "step": 5465
    },
    {
      "epoch": 0.8924055795741904,
      "grad_norm": 0.5906075835227966,
      "learning_rate": 4.772899515849936e-05,
      "loss": 0.2078,
      "num_input_tokens_seen": 10370016,
      "step": 5470
    },
    {
      "epoch": 0.8932213067950078,
      "grad_norm": 0.6199472546577454,
      "learning_rate": 4.7724904943154414e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 10380016,
      "step": 5475
    },
    {
      "epoch": 0.8940370340158251,
      "grad_norm": 0.08740100264549255,
      "learning_rate": 4.772081122334354e-05,
      "loss": 0.0944,
      "num_input_tokens_seen": 10389936,
      "step": 5480
    },
    {
      "epoch": 0.8948527612366425,
      "grad_norm": 0.8958070874214172,
      "learning_rate": 4.771671399969806e-05,
      "loss": 0.1301,
      "num_input_tokens_seen": 10399376,
      "step": 5485
    },
    {
      "epoch": 0.8956684884574598,
      "grad_norm": 0.4656934142112732,
      "learning_rate": 4.7712613272849794e-05,
      "loss": 0.3272,
      "num_input_tokens_seen": 10408576,
      "step": 5490
    },
    {
      "epoch": 0.8964842156782772,
      "grad_norm": 0.598686695098877,
      "learning_rate": 4.770850904343114e-05,
      "loss": 0.0981,
      "num_input_tokens_seen": 10419232,
      "step": 5495
    },
    {
      "epoch": 0.8972999428990945,
      "grad_norm": 0.13843095302581787,
      "learning_rate": 4.770440131207502e-05,
      "loss": 0.2101,
      "num_input_tokens_seen": 10429360,
      "step": 5500
    },
    {
      "epoch": 0.8981156701199119,
      "grad_norm": 0.4008427858352661,
      "learning_rate": 4.7700290079414896e-05,
      "loss": 0.1694,
      "num_input_tokens_seen": 10438912,
      "step": 5505
    },
    {
      "epoch": 0.8989313973407292,
      "grad_norm": 1.3470638990402222,
      "learning_rate": 4.769617534608477e-05,
      "loss": 0.179,
      "num_input_tokens_seen": 10448416,
      "step": 5510
    },
    {
      "epoch": 0.8997471245615466,
      "grad_norm": 0.23516157269477844,
      "learning_rate": 4.7692057112719193e-05,
      "loss": 0.0596,
      "num_input_tokens_seen": 10457792,
      "step": 5515
    },
    {
      "epoch": 0.900562851782364,
      "grad_norm": 0.4293440282344818,
      "learning_rate": 4.7687935379953234e-05,
      "loss": 0.1622,
      "num_input_tokens_seen": 10466752,
      "step": 5520
    },
    {
      "epoch": 0.9013785790031813,
      "grad_norm": 0.8440062999725342,
      "learning_rate": 4.7683810148422534e-05,
      "loss": 0.2886,
      "num_input_tokens_seen": 10475776,
      "step": 5525
    },
    {
      "epoch": 0.9021943062239987,
      "grad_norm": 0.7966614365577698,
      "learning_rate": 4.767968141876324e-05,
      "loss": 0.1054,
      "num_input_tokens_seen": 10485328,
      "step": 5530
    },
    {
      "epoch": 0.903010033444816,
      "grad_norm": 0.31128203868865967,
      "learning_rate": 4.767554919161207e-05,
      "loss": 0.0855,
      "num_input_tokens_seen": 10493792,
      "step": 5535
    },
    {
      "epoch": 0.9038257606656334,
      "grad_norm": 0.37373828887939453,
      "learning_rate": 4.767141346760624e-05,
      "loss": 0.0491,
      "num_input_tokens_seen": 10503712,
      "step": 5540
    },
    {
      "epoch": 0.9046414878864508,
      "grad_norm": 1.1986572742462158,
      "learning_rate": 4.766727424738356e-05,
      "loss": 0.3787,
      "num_input_tokens_seen": 10513392,
      "step": 5545
    },
    {
      "epoch": 0.9054572151072682,
      "grad_norm": 0.8355146050453186,
      "learning_rate": 4.7663131531582325e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 10523456,
      "step": 5550
    },
    {
      "epoch": 0.9062729423280855,
      "grad_norm": 2.0137805938720703,
      "learning_rate": 4.765898532084142e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 10533104,
      "step": 5555
    },
    {
      "epoch": 0.9070886695489029,
      "grad_norm": 0.9134929776191711,
      "learning_rate": 4.765483561580022e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 10543648,
      "step": 5560
    },
    {
      "epoch": 0.9079043967697202,
      "grad_norm": 0.22070784866809845,
      "learning_rate": 4.7650682417098666e-05,
      "loss": 0.0928,
      "num_input_tokens_seen": 10551888,
      "step": 5565
    },
    {
      "epoch": 0.9087201239905376,
      "grad_norm": 1.2114595174789429,
      "learning_rate": 4.7646525725377244e-05,
      "loss": 0.1959,
      "num_input_tokens_seen": 10561696,
      "step": 5570
    },
    {
      "epoch": 0.9095358512113549,
      "grad_norm": 0.21987158060073853,
      "learning_rate": 4.764236554127696e-05,
      "loss": 0.1551,
      "num_input_tokens_seen": 10572000,
      "step": 5575
    },
    {
      "epoch": 0.9103515784321723,
      "grad_norm": 0.20012637972831726,
      "learning_rate": 4.7638201865439356e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 10581920,
      "step": 5580
    },
    {
      "epoch": 0.9111673056529896,
      "grad_norm": 1.3446975946426392,
      "learning_rate": 4.7634034698506545e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 10590800,
      "step": 5585
    },
    {
      "epoch": 0.911983032873807,
      "grad_norm": 0.3007095158100128,
      "learning_rate": 4.762986404112115e-05,
      "loss": 0.0993,
      "num_input_tokens_seen": 10600272,
      "step": 5590
    },
    {
      "epoch": 0.9127987600946244,
      "grad_norm": 0.8557928800582886,
      "learning_rate": 4.762568989392633e-05,
      "loss": 0.2182,
      "num_input_tokens_seen": 10609808,
      "step": 5595
    },
    {
      "epoch": 0.9136144873154417,
      "grad_norm": 0.3245173990726471,
      "learning_rate": 4.76215122575658e-05,
      "loss": 0.2386,
      "num_input_tokens_seen": 10619840,
      "step": 5600
    },
    {
      "epoch": 0.9136144873154417,
      "eval_loss": 0.15691335499286652,
      "eval_runtime": 133.0175,
      "eval_samples_per_second": 20.486,
      "eval_steps_per_second": 10.247,
      "num_input_tokens_seen": 10619840,
      "step": 5600
    },
    {
      "epoch": 0.9144302145362591,
      "grad_norm": 0.2999814748764038,
      "learning_rate": 4.7617331132683795e-05,
      "loss": 0.2131,
      "num_input_tokens_seen": 10629344,
      "step": 5605
    },
    {
      "epoch": 0.9152459417570764,
      "grad_norm": 1.413383960723877,
      "learning_rate": 4.7613146519925105e-05,
      "loss": 0.2419,
      "num_input_tokens_seen": 10639136,
      "step": 5610
    },
    {
      "epoch": 0.9160616689778938,
      "grad_norm": 1.6526219844818115,
      "learning_rate": 4.7608958419935045e-05,
      "loss": 0.0944,
      "num_input_tokens_seen": 10647632,
      "step": 5615
    },
    {
      "epoch": 0.9168773961987111,
      "grad_norm": 0.2933381199836731,
      "learning_rate": 4.760476683335948e-05,
      "loss": 0.0718,
      "num_input_tokens_seen": 10656368,
      "step": 5620
    },
    {
      "epoch": 0.9176931234195285,
      "grad_norm": 0.9489834308624268,
      "learning_rate": 4.760057176084479e-05,
      "loss": 0.0787,
      "num_input_tokens_seen": 10665424,
      "step": 5625
    },
    {
      "epoch": 0.9185088506403458,
      "grad_norm": 1.3978222608566284,
      "learning_rate": 4.759637320303793e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 10675312,
      "step": 5630
    },
    {
      "epoch": 0.9193245778611632,
      "grad_norm": 0.9703343510627747,
      "learning_rate": 4.759217116058635e-05,
      "loss": 0.0864,
      "num_input_tokens_seen": 10684016,
      "step": 5635
    },
    {
      "epoch": 0.9201403050819806,
      "grad_norm": 0.5093905925750732,
      "learning_rate": 4.758796563413807e-05,
      "loss": 0.0917,
      "num_input_tokens_seen": 10693616,
      "step": 5640
    },
    {
      "epoch": 0.920956032302798,
      "grad_norm": 2.016280174255371,
      "learning_rate": 4.758375662434163e-05,
      "loss": 0.2589,
      "num_input_tokens_seen": 10702640,
      "step": 5645
    },
    {
      "epoch": 0.9217717595236153,
      "grad_norm": 4.363397121429443,
      "learning_rate": 4.7579544131846114e-05,
      "loss": 0.1833,
      "num_input_tokens_seen": 10711488,
      "step": 5650
    },
    {
      "epoch": 0.9225874867444327,
      "grad_norm": 0.23450565338134766,
      "learning_rate": 4.757532815730114e-05,
      "loss": 0.1665,
      "num_input_tokens_seen": 10720992,
      "step": 5655
    },
    {
      "epoch": 0.92340321396525,
      "grad_norm": 2.2160263061523438,
      "learning_rate": 4.7571108701356865e-05,
      "loss": 0.1717,
      "num_input_tokens_seen": 10729872,
      "step": 5660
    },
    {
      "epoch": 0.9242189411860674,
      "grad_norm": 1.5513887405395508,
      "learning_rate": 4.756688576466398e-05,
      "loss": 0.2806,
      "num_input_tokens_seen": 10739968,
      "step": 5665
    },
    {
      "epoch": 0.9250346684068848,
      "grad_norm": 0.3783125579357147,
      "learning_rate": 4.756265934787372e-05,
      "loss": 0.2194,
      "num_input_tokens_seen": 10748976,
      "step": 5670
    },
    {
      "epoch": 0.9258503956277021,
      "grad_norm": 0.24833565950393677,
      "learning_rate": 4.755842945163785e-05,
      "loss": 0.177,
      "num_input_tokens_seen": 10757968,
      "step": 5675
    },
    {
      "epoch": 0.9266661228485195,
      "grad_norm": 1.7818739414215088,
      "learning_rate": 4.755419607660867e-05,
      "loss": 0.2511,
      "num_input_tokens_seen": 10767888,
      "step": 5680
    },
    {
      "epoch": 0.9274818500693368,
      "grad_norm": 0.5473159551620483,
      "learning_rate": 4.7549959223439016e-05,
      "loss": 0.1848,
      "num_input_tokens_seen": 10777296,
      "step": 5685
    },
    {
      "epoch": 0.9282975772901542,
      "grad_norm": 1.6793205738067627,
      "learning_rate": 4.754571889278228e-05,
      "loss": 0.2017,
      "num_input_tokens_seen": 10787216,
      "step": 5690
    },
    {
      "epoch": 0.9291133045109715,
      "grad_norm": 0.29320451617240906,
      "learning_rate": 4.754147508529235e-05,
      "loss": 0.1234,
      "num_input_tokens_seen": 10797024,
      "step": 5695
    },
    {
      "epoch": 0.9299290317317889,
      "grad_norm": 2.0719733238220215,
      "learning_rate": 4.75372278016237e-05,
      "loss": 0.2257,
      "num_input_tokens_seen": 10806240,
      "step": 5700
    },
    {
      "epoch": 0.9307447589526062,
      "grad_norm": 0.159564808011055,
      "learning_rate": 4.753297704243129e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 10815680,
      "step": 5705
    },
    {
      "epoch": 0.9315604861734236,
      "grad_norm": 0.5849626064300537,
      "learning_rate": 4.752872280837066e-05,
      "loss": 0.0929,
      "num_input_tokens_seen": 10825664,
      "step": 5710
    },
    {
      "epoch": 0.9323762133942409,
      "grad_norm": 0.38415756821632385,
      "learning_rate": 4.752446510009786e-05,
      "loss": 0.0705,
      "num_input_tokens_seen": 10836224,
      "step": 5715
    },
    {
      "epoch": 0.9331919406150583,
      "grad_norm": 0.4245564341545105,
      "learning_rate": 4.7520203918269476e-05,
      "loss": 0.2193,
      "num_input_tokens_seen": 10844960,
      "step": 5720
    },
    {
      "epoch": 0.9340076678358756,
      "grad_norm": 0.7232838869094849,
      "learning_rate": 4.751593926354265e-05,
      "loss": 0.2391,
      "num_input_tokens_seen": 10854496,
      "step": 5725
    },
    {
      "epoch": 0.9348233950566931,
      "grad_norm": 0.3136577010154724,
      "learning_rate": 4.751167113657503e-05,
      "loss": 0.1176,
      "num_input_tokens_seen": 10863040,
      "step": 5730
    },
    {
      "epoch": 0.9356391222775104,
      "grad_norm": 0.7827233076095581,
      "learning_rate": 4.7507399538024834e-05,
      "loss": 0.1083,
      "num_input_tokens_seen": 10872576,
      "step": 5735
    },
    {
      "epoch": 0.9364548494983278,
      "grad_norm": 0.9011039137840271,
      "learning_rate": 4.750312446855077e-05,
      "loss": 0.1665,
      "num_input_tokens_seen": 10882032,
      "step": 5740
    },
    {
      "epoch": 0.9372705767191452,
      "grad_norm": 0.4380326569080353,
      "learning_rate": 4.749884592881212e-05,
      "loss": 0.2274,
      "num_input_tokens_seen": 10891728,
      "step": 5745
    },
    {
      "epoch": 0.9380863039399625,
      "grad_norm": 1.2037184238433838,
      "learning_rate": 4.74945639194687e-05,
      "loss": 0.2845,
      "num_input_tokens_seen": 10901760,
      "step": 5750
    },
    {
      "epoch": 0.9389020311607799,
      "grad_norm": 0.30257630348205566,
      "learning_rate": 4.749027844118083e-05,
      "loss": 0.0715,
      "num_input_tokens_seen": 10910736,
      "step": 5755
    },
    {
      "epoch": 0.9397177583815972,
      "grad_norm": 0.06105760484933853,
      "learning_rate": 4.7485989494609395e-05,
      "loss": 0.0493,
      "num_input_tokens_seen": 10921408,
      "step": 5760
    },
    {
      "epoch": 0.9405334856024146,
      "grad_norm": 0.17151518166065216,
      "learning_rate": 4.748169708041581e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 10931056,
      "step": 5765
    },
    {
      "epoch": 0.9413492128232319,
      "grad_norm": 0.17714154720306396,
      "learning_rate": 4.7477401199262004e-05,
      "loss": 0.2491,
      "num_input_tokens_seen": 10940736,
      "step": 5770
    },
    {
      "epoch": 0.9421649400440493,
      "grad_norm": 0.07650994509458542,
      "learning_rate": 4.747310185181048e-05,
      "loss": 0.3516,
      "num_input_tokens_seen": 10950320,
      "step": 5775
    },
    {
      "epoch": 0.9429806672648666,
      "grad_norm": 0.4255955219268799,
      "learning_rate": 4.746879903872422e-05,
      "loss": 0.0685,
      "num_input_tokens_seen": 10958592,
      "step": 5780
    },
    {
      "epoch": 0.943796394485684,
      "grad_norm": 0.27400192618370056,
      "learning_rate": 4.746449276066679e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 10967088,
      "step": 5785
    },
    {
      "epoch": 0.9446121217065013,
      "grad_norm": 0.2560301721096039,
      "learning_rate": 4.746018301830227e-05,
      "loss": 0.1989,
      "num_input_tokens_seen": 10976416,
      "step": 5790
    },
    {
      "epoch": 0.9454278489273187,
      "grad_norm": 0.595703125,
      "learning_rate": 4.7455869812295275e-05,
      "loss": 0.2958,
      "num_input_tokens_seen": 10985344,
      "step": 5795
    },
    {
      "epoch": 0.946243576148136,
      "grad_norm": 0.2032541036605835,
      "learning_rate": 4.7451553143310964e-05,
      "loss": 0.2734,
      "num_input_tokens_seen": 10994720,
      "step": 5800
    },
    {
      "epoch": 0.946243576148136,
      "eval_loss": 0.15637259185314178,
      "eval_runtime": 133.0624,
      "eval_samples_per_second": 20.479,
      "eval_steps_per_second": 10.243,
      "num_input_tokens_seen": 10994720,
      "step": 5800
    },
    {
      "epoch": 0.9470593033689534,
      "grad_norm": 0.2938968539237976,
      "learning_rate": 4.744723301201501e-05,
      "loss": 0.122,
      "num_input_tokens_seen": 11004464,
      "step": 5805
    },
    {
      "epoch": 0.9478750305897707,
      "grad_norm": 0.030944833531975746,
      "learning_rate": 4.744290941907364e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 11013968,
      "step": 5810
    },
    {
      "epoch": 0.9486907578105881,
      "grad_norm": 1.0575133562088013,
      "learning_rate": 4.7438582365153594e-05,
      "loss": 0.1007,
      "num_input_tokens_seen": 11023632,
      "step": 5815
    },
    {
      "epoch": 0.9495064850314054,
      "grad_norm": 3.3201329708099365,
      "learning_rate": 4.743425185092217e-05,
      "loss": 0.3201,
      "num_input_tokens_seen": 11033840,
      "step": 5820
    },
    {
      "epoch": 0.9503222122522229,
      "grad_norm": 0.2909562289714813,
      "learning_rate": 4.742991787704719e-05,
      "loss": 0.2334,
      "num_input_tokens_seen": 11044304,
      "step": 5825
    },
    {
      "epoch": 0.9511379394730403,
      "grad_norm": 0.25577494502067566,
      "learning_rate": 4.7425580444196994e-05,
      "loss": 0.0275,
      "num_input_tokens_seen": 11054320,
      "step": 5830
    },
    {
      "epoch": 0.9519536666938576,
      "grad_norm": 0.8433043956756592,
      "learning_rate": 4.742123955304048e-05,
      "loss": 0.3045,
      "num_input_tokens_seen": 11063728,
      "step": 5835
    },
    {
      "epoch": 0.952769393914675,
      "grad_norm": 1.2251555919647217,
      "learning_rate": 4.741689520424706e-05,
      "loss": 0.2182,
      "num_input_tokens_seen": 11072672,
      "step": 5840
    },
    {
      "epoch": 0.9535851211354923,
      "grad_norm": 0.11660721898078918,
      "learning_rate": 4.741254739848669e-05,
      "loss": 0.0741,
      "num_input_tokens_seen": 11082336,
      "step": 5845
    },
    {
      "epoch": 0.9544008483563097,
      "grad_norm": 1.0236564874649048,
      "learning_rate": 4.740819613642987e-05,
      "loss": 0.1078,
      "num_input_tokens_seen": 11091232,
      "step": 5850
    },
    {
      "epoch": 0.955216575577127,
      "grad_norm": 0.7976194620132446,
      "learning_rate": 4.74038414187476e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 11100272,
      "step": 5855
    },
    {
      "epoch": 0.9560323027979444,
      "grad_norm": 1.799959659576416,
      "learning_rate": 4.739948324611144e-05,
      "loss": 0.2072,
      "num_input_tokens_seen": 11110304,
      "step": 5860
    },
    {
      "epoch": 0.9568480300187617,
      "grad_norm": 0.395037442445755,
      "learning_rate": 4.7395121619193465e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 11119216,
      "step": 5865
    },
    {
      "epoch": 0.9576637572395791,
      "grad_norm": 0.2014250010251999,
      "learning_rate": 4.7390756538666313e-05,
      "loss": 0.0476,
      "num_input_tokens_seen": 11128752,
      "step": 5870
    },
    {
      "epoch": 0.9584794844603964,
      "grad_norm": 0.17500074207782745,
      "learning_rate": 4.738638800520311e-05,
      "loss": 0.1871,
      "num_input_tokens_seen": 11137664,
      "step": 5875
    },
    {
      "epoch": 0.9592952116812138,
      "grad_norm": 0.21255739033222198,
      "learning_rate": 4.738201601947757e-05,
      "loss": 0.1473,
      "num_input_tokens_seen": 11147200,
      "step": 5880
    },
    {
      "epoch": 0.9601109389020311,
      "grad_norm": 0.7499558925628662,
      "learning_rate": 4.7377640582163876e-05,
      "loss": 0.262,
      "num_input_tokens_seen": 11156864,
      "step": 5885
    },
    {
      "epoch": 0.9609266661228485,
      "grad_norm": 0.9261627793312073,
      "learning_rate": 4.7373261693936786e-05,
      "loss": 0.2236,
      "num_input_tokens_seen": 11165792,
      "step": 5890
    },
    {
      "epoch": 0.9617423933436658,
      "grad_norm": 0.051382437348365784,
      "learning_rate": 4.7368879355471595e-05,
      "loss": 0.1039,
      "num_input_tokens_seen": 11175040,
      "step": 5895
    },
    {
      "epoch": 0.9625581205644832,
      "grad_norm": 0.5109784007072449,
      "learning_rate": 4.736449356744409e-05,
      "loss": 0.2495,
      "num_input_tokens_seen": 11183744,
      "step": 5900
    },
    {
      "epoch": 0.9633738477853006,
      "grad_norm": 0.6701955199241638,
      "learning_rate": 4.736010433053064e-05,
      "loss": 0.0915,
      "num_input_tokens_seen": 11193232,
      "step": 5905
    },
    {
      "epoch": 0.9641895750061179,
      "grad_norm": 2.113588571548462,
      "learning_rate": 4.73557116454081e-05,
      "loss": 0.168,
      "num_input_tokens_seen": 11202512,
      "step": 5910
    },
    {
      "epoch": 0.9650053022269353,
      "grad_norm": 0.48498451709747314,
      "learning_rate": 4.735131551275389e-05,
      "loss": 0.3279,
      "num_input_tokens_seen": 11212624,
      "step": 5915
    },
    {
      "epoch": 0.9658210294477527,
      "grad_norm": 0.4217439293861389,
      "learning_rate": 4.734691593324594e-05,
      "loss": 0.1044,
      "num_input_tokens_seen": 11222512,
      "step": 5920
    },
    {
      "epoch": 0.9666367566685701,
      "grad_norm": 1.6198352575302124,
      "learning_rate": 4.734251290756272e-05,
      "loss": 0.2133,
      "num_input_tokens_seen": 11232144,
      "step": 5925
    },
    {
      "epoch": 0.9674524838893874,
      "grad_norm": 0.6514819264411926,
      "learning_rate": 4.7338106436383246e-05,
      "loss": 0.168,
      "num_input_tokens_seen": 11241168,
      "step": 5930
    },
    {
      "epoch": 0.9682682111102048,
      "grad_norm": 2.4229536056518555,
      "learning_rate": 4.733369652038703e-05,
      "loss": 0.0857,
      "num_input_tokens_seen": 11251552,
      "step": 5935
    },
    {
      "epoch": 0.9690839383310221,
      "grad_norm": 0.5682224631309509,
      "learning_rate": 4.7329283160254156e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 11262240,
      "step": 5940
    },
    {
      "epoch": 0.9698996655518395,
      "grad_norm": 0.8549497723579407,
      "learning_rate": 4.732486635666521e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 11272256,
      "step": 5945
    },
    {
      "epoch": 0.9707153927726568,
      "grad_norm": 0.7438622713088989,
      "learning_rate": 4.732044611030132e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 11281904,
      "step": 5950
    },
    {
      "epoch": 0.9715311199934742,
      "grad_norm": 1.0991896390914917,
      "learning_rate": 4.731602242184414e-05,
      "loss": 0.2055,
      "num_input_tokens_seen": 11291184,
      "step": 5955
    },
    {
      "epoch": 0.9723468472142915,
      "grad_norm": 0.7279292941093445,
      "learning_rate": 4.7311595291975864e-05,
      "loss": 0.189,
      "num_input_tokens_seen": 11301312,
      "step": 5960
    },
    {
      "epoch": 0.9731625744351089,
      "grad_norm": 0.09585081040859222,
      "learning_rate": 4.7307164721379216e-05,
      "loss": 0.0984,
      "num_input_tokens_seen": 11311840,
      "step": 5965
    },
    {
      "epoch": 0.9739783016559262,
      "grad_norm": 0.4323747456073761,
      "learning_rate": 4.730273071073743e-05,
      "loss": 0.1646,
      "num_input_tokens_seen": 11321264,
      "step": 5970
    },
    {
      "epoch": 0.9747940288767436,
      "grad_norm": 2.8953568935394287,
      "learning_rate": 4.729829326073429e-05,
      "loss": 0.3838,
      "num_input_tokens_seen": 11330096,
      "step": 5975
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 2.0525076389312744,
      "learning_rate": 4.7293852372054126e-05,
      "loss": 0.4418,
      "num_input_tokens_seen": 11340064,
      "step": 5980
    },
    {
      "epoch": 0.9764254833183783,
      "grad_norm": 1.8824132680892944,
      "learning_rate": 4.728940804538176e-05,
      "loss": 0.2526,
      "num_input_tokens_seen": 11348608,
      "step": 5985
    },
    {
      "epoch": 0.9772412105391957,
      "grad_norm": 0.566747784614563,
      "learning_rate": 4.7284960281402556e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 11358384,
      "step": 5990
    },
    {
      "epoch": 0.978056937760013,
      "grad_norm": 1.9073556661605835,
      "learning_rate": 4.728050908080244e-05,
      "loss": 0.2618,
      "num_input_tokens_seen": 11368608,
      "step": 5995
    },
    {
      "epoch": 0.9788726649808304,
      "grad_norm": 0.46308454871177673,
      "learning_rate": 4.727605444426782e-05,
      "loss": 0.1958,
      "num_input_tokens_seen": 11376976,
      "step": 6000
    },
    {
      "epoch": 0.9788726649808304,
      "eval_loss": 0.15852771699428558,
      "eval_runtime": 133.2632,
      "eval_samples_per_second": 20.448,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 11376976,
      "step": 6000
    },
    {
      "epoch": 0.9796883922016477,
      "grad_norm": 0.8648409247398376,
      "learning_rate": 4.727159637248567e-05,
      "loss": 0.0965,
      "num_input_tokens_seen": 11386688,
      "step": 6005
    },
    {
      "epoch": 0.9805041194224652,
      "grad_norm": 0.33311986923217773,
      "learning_rate": 4.7267134866143474e-05,
      "loss": 0.1699,
      "num_input_tokens_seen": 11395552,
      "step": 6010
    },
    {
      "epoch": 0.9813198466432825,
      "grad_norm": 0.10596083849668503,
      "learning_rate": 4.726266992592926e-05,
      "loss": 0.0363,
      "num_input_tokens_seen": 11403968,
      "step": 6015
    },
    {
      "epoch": 0.9821355738640999,
      "grad_norm": 0.1458829939365387,
      "learning_rate": 4.725820155253157e-05,
      "loss": 0.0499,
      "num_input_tokens_seen": 11413888,
      "step": 6020
    },
    {
      "epoch": 0.9829513010849172,
      "grad_norm": 0.15650710463523865,
      "learning_rate": 4.725372974663948e-05,
      "loss": 0.1451,
      "num_input_tokens_seen": 11423904,
      "step": 6025
    },
    {
      "epoch": 0.9837670283057346,
      "grad_norm": 1.366938591003418,
      "learning_rate": 4.724925450894262e-05,
      "loss": 0.2369,
      "num_input_tokens_seen": 11434176,
      "step": 6030
    },
    {
      "epoch": 0.984582755526552,
      "grad_norm": 2.392028331756592,
      "learning_rate": 4.72447758401311e-05,
      "loss": 0.3109,
      "num_input_tokens_seen": 11443792,
      "step": 6035
    },
    {
      "epoch": 0.9853984827473693,
      "grad_norm": 0.4548121988773346,
      "learning_rate": 4.7240293740895616e-05,
      "loss": 0.1398,
      "num_input_tokens_seen": 11453264,
      "step": 6040
    },
    {
      "epoch": 0.9862142099681867,
      "grad_norm": 0.4304293692111969,
      "learning_rate": 4.723580821192733e-05,
      "loss": 0.0906,
      "num_input_tokens_seen": 11462256,
      "step": 6045
    },
    {
      "epoch": 0.987029937189004,
      "grad_norm": 1.2856439352035522,
      "learning_rate": 4.7231319253917996e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 11470560,
      "step": 6050
    },
    {
      "epoch": 0.9878456644098214,
      "grad_norm": 0.14370989799499512,
      "learning_rate": 4.722682686755986e-05,
      "loss": 0.093,
      "num_input_tokens_seen": 11481088,
      "step": 6055
    },
    {
      "epoch": 0.9886613916306387,
      "grad_norm": 1.1885584592819214,
      "learning_rate": 4.722233105354569e-05,
      "loss": 0.3004,
      "num_input_tokens_seen": 11490800,
      "step": 6060
    },
    {
      "epoch": 0.9894771188514561,
      "grad_norm": 0.18330034613609314,
      "learning_rate": 4.7217831812568815e-05,
      "loss": 0.0969,
      "num_input_tokens_seen": 11500720,
      "step": 6065
    },
    {
      "epoch": 0.9902928460722734,
      "grad_norm": 0.09654026478528976,
      "learning_rate": 4.721332914532307e-05,
      "loss": 0.0659,
      "num_input_tokens_seen": 11510592,
      "step": 6070
    },
    {
      "epoch": 0.9911085732930908,
      "grad_norm": 0.5997007489204407,
      "learning_rate": 4.720882305250281e-05,
      "loss": 0.1771,
      "num_input_tokens_seen": 11520096,
      "step": 6075
    },
    {
      "epoch": 0.9919243005139081,
      "grad_norm": 0.8108166456222534,
      "learning_rate": 4.720431353480295e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 11529824,
      "step": 6080
    },
    {
      "epoch": 0.9927400277347255,
      "grad_norm": 1.3864598274230957,
      "learning_rate": 4.719980059291891e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 11537984,
      "step": 6085
    },
    {
      "epoch": 0.9935557549555428,
      "grad_norm": 0.3627852201461792,
      "learning_rate": 4.7195284227546634e-05,
      "loss": 0.1286,
      "num_input_tokens_seen": 11547232,
      "step": 6090
    },
    {
      "epoch": 0.9943714821763602,
      "grad_norm": 1.292478084564209,
      "learning_rate": 4.7190764439382604e-05,
      "loss": 0.2482,
      "num_input_tokens_seen": 11557840,
      "step": 6095
    },
    {
      "epoch": 0.9951872093971775,
      "grad_norm": 0.2780191898345947,
      "learning_rate": 4.7186241229123826e-05,
      "loss": 0.1694,
      "num_input_tokens_seen": 11566880,
      "step": 6100
    },
    {
      "epoch": 0.996002936617995,
      "grad_norm": 0.2546718120574951,
      "learning_rate": 4.718171459746785e-05,
      "loss": 0.2096,
      "num_input_tokens_seen": 11575424,
      "step": 6105
    },
    {
      "epoch": 0.9968186638388123,
      "grad_norm": 0.4323120713233948,
      "learning_rate": 4.717718454511273e-05,
      "loss": 0.0772,
      "num_input_tokens_seen": 11584480,
      "step": 6110
    },
    {
      "epoch": 0.9976343910596297,
      "grad_norm": 0.08625513315200806,
      "learning_rate": 4.7172651072757056e-05,
      "loss": 0.1125,
      "num_input_tokens_seen": 11594112,
      "step": 6115
    },
    {
      "epoch": 0.998450118280447,
      "grad_norm": 0.31197208166122437,
      "learning_rate": 4.7168114181099945e-05,
      "loss": 0.209,
      "num_input_tokens_seen": 11604128,
      "step": 6120
    },
    {
      "epoch": 0.9992658455012644,
      "grad_norm": 0.6904513239860535,
      "learning_rate": 4.716357387084105e-05,
      "loss": 0.201,
      "num_input_tokens_seen": 11614416,
      "step": 6125
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.08742290735244751,
      "learning_rate": 4.715903014268054e-05,
      "loss": 0.1568,
      "num_input_tokens_seen": 11623536,
      "step": 6130
    },
    {
      "epoch": 1.0008157272208174,
      "grad_norm": 0.8763659596443176,
      "learning_rate": 4.715448299731911e-05,
      "loss": 0.1179,
      "num_input_tokens_seen": 11633472,
      "step": 6135
    },
    {
      "epoch": 1.0016314544416347,
      "grad_norm": 0.5487121343612671,
      "learning_rate": 4.7149932435457986e-05,
      "loss": 0.1002,
      "num_input_tokens_seen": 11643408,
      "step": 6140
    },
    {
      "epoch": 1.002447181662452,
      "grad_norm": 0.4367242157459259,
      "learning_rate": 4.714537845779894e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 11652832,
      "step": 6145
    },
    {
      "epoch": 1.0032629088832694,
      "grad_norm": 0.28045526146888733,
      "learning_rate": 4.714082106504423e-05,
      "loss": 0.1849,
      "num_input_tokens_seen": 11663344,
      "step": 6150
    },
    {
      "epoch": 1.0040786361040868,
      "grad_norm": 0.5211780667304993,
      "learning_rate": 4.713626025789667e-05,
      "loss": 0.1077,
      "num_input_tokens_seen": 11672720,
      "step": 6155
    },
    {
      "epoch": 1.0048943633249041,
      "grad_norm": 0.05184370279312134,
      "learning_rate": 4.7131696037059606e-05,
      "loss": 0.0484,
      "num_input_tokens_seen": 11681664,
      "step": 6160
    },
    {
      "epoch": 1.0057100905457215,
      "grad_norm": 0.19494542479515076,
      "learning_rate": 4.712712840323689e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 11691664,
      "step": 6165
    },
    {
      "epoch": 1.0065258177665388,
      "grad_norm": 0.44210073351860046,
      "learning_rate": 4.71225573571329e-05,
      "loss": 0.1367,
      "num_input_tokens_seen": 11701792,
      "step": 6170
    },
    {
      "epoch": 1.0073415449873562,
      "grad_norm": 0.2346869558095932,
      "learning_rate": 4.711798289945256e-05,
      "loss": 0.1919,
      "num_input_tokens_seen": 11711296,
      "step": 6175
    },
    {
      "epoch": 1.0081572722081735,
      "grad_norm": 0.8011819124221802,
      "learning_rate": 4.71134050309013e-05,
      "loss": 0.1789,
      "num_input_tokens_seen": 11721168,
      "step": 6180
    },
    {
      "epoch": 1.0089729994289909,
      "grad_norm": 0.6971073150634766,
      "learning_rate": 4.710882375218509e-05,
      "loss": 0.088,
      "num_input_tokens_seen": 11731344,
      "step": 6185
    },
    {
      "epoch": 1.0097887266498082,
      "grad_norm": 0.12228992581367493,
      "learning_rate": 4.7104239064010424e-05,
      "loss": 0.1135,
      "num_input_tokens_seen": 11740896,
      "step": 6190
    },
    {
      "epoch": 1.0106044538706256,
      "grad_norm": 1.6460298299789429,
      "learning_rate": 4.709965096708432e-05,
      "loss": 0.3237,
      "num_input_tokens_seen": 11748640,
      "step": 6195
    },
    {
      "epoch": 1.011420181091443,
      "grad_norm": 0.7989368438720703,
      "learning_rate": 4.709505946211431e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 11758656,
      "step": 6200
    },
    {
      "epoch": 1.011420181091443,
      "eval_loss": 0.15703138709068298,
      "eval_runtime": 132.9182,
      "eval_samples_per_second": 20.501,
      "eval_steps_per_second": 10.254,
      "num_input_tokens_seen": 11758656,
      "step": 6200
    },
    {
      "epoch": 1.0122359083122603,
      "grad_norm": 0.5020419955253601,
      "learning_rate": 4.709046454980846e-05,
      "loss": 0.1024,
      "num_input_tokens_seen": 11769376,
      "step": 6205
    },
    {
      "epoch": 1.0130516355330776,
      "grad_norm": 0.4701400697231293,
      "learning_rate": 4.708586623087538e-05,
      "loss": 0.1034,
      "num_input_tokens_seen": 11778768,
      "step": 6210
    },
    {
      "epoch": 1.013867362753895,
      "grad_norm": 0.33598053455352783,
      "learning_rate": 4.708126450602418e-05,
      "loss": 0.0822,
      "num_input_tokens_seen": 11789360,
      "step": 6215
    },
    {
      "epoch": 1.0146830899747123,
      "grad_norm": 0.20227575302124023,
      "learning_rate": 4.7076659375964495e-05,
      "loss": 0.0923,
      "num_input_tokens_seen": 11799376,
      "step": 6220
    },
    {
      "epoch": 1.01549881719553,
      "grad_norm": 1.5454515218734741,
      "learning_rate": 4.707205084140651e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 11807328,
      "step": 6225
    },
    {
      "epoch": 1.0163145444163473,
      "grad_norm": 0.39750906825065613,
      "learning_rate": 4.7067438903060904e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 11817120,
      "step": 6230
    },
    {
      "epoch": 1.0171302716371646,
      "grad_norm": 1.6657257080078125,
      "learning_rate": 4.70628235616389e-05,
      "loss": 0.1636,
      "num_input_tokens_seen": 11826128,
      "step": 6235
    },
    {
      "epoch": 1.017945998857982,
      "grad_norm": 0.6505307555198669,
      "learning_rate": 4.7058204817852256e-05,
      "loss": 0.1723,
      "num_input_tokens_seen": 11835424,
      "step": 6240
    },
    {
      "epoch": 1.0187617260787993,
      "grad_norm": 0.8056198954582214,
      "learning_rate": 4.705358267241322e-05,
      "loss": 0.0954,
      "num_input_tokens_seen": 11845568,
      "step": 6245
    },
    {
      "epoch": 1.0195774532996167,
      "grad_norm": 0.5658764839172363,
      "learning_rate": 4.704895712603459e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 11855552,
      "step": 6250
    },
    {
      "epoch": 1.020393180520434,
      "grad_norm": 0.5010653138160706,
      "learning_rate": 4.704432817942969e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 11864880,
      "step": 6255
    },
    {
      "epoch": 1.0212089077412514,
      "grad_norm": 0.37684163451194763,
      "learning_rate": 4.703969583331236e-05,
      "loss": 0.1777,
      "num_input_tokens_seen": 11874592,
      "step": 6260
    },
    {
      "epoch": 1.0220246349620687,
      "grad_norm": 0.5179290771484375,
      "learning_rate": 4.7035060088396965e-05,
      "loss": 0.0661,
      "num_input_tokens_seen": 11883792,
      "step": 6265
    },
    {
      "epoch": 1.022840362182886,
      "grad_norm": 0.29456886649131775,
      "learning_rate": 4.703042094539839e-05,
      "loss": 0.2914,
      "num_input_tokens_seen": 11892032,
      "step": 6270
    },
    {
      "epoch": 1.0236560894037035,
      "grad_norm": 0.18667669594287872,
      "learning_rate": 4.702577840503206e-05,
      "loss": 0.0469,
      "num_input_tokens_seen": 11901264,
      "step": 6275
    },
    {
      "epoch": 1.0244718166245208,
      "grad_norm": 0.1587585061788559,
      "learning_rate": 4.70211324680139e-05,
      "loss": 0.1798,
      "num_input_tokens_seen": 11910784,
      "step": 6280
    },
    {
      "epoch": 1.0252875438453382,
      "grad_norm": 0.9214814901351929,
      "learning_rate": 4.7016483135060386e-05,
      "loss": 0.1161,
      "num_input_tokens_seen": 11921504,
      "step": 6285
    },
    {
      "epoch": 1.0261032710661555,
      "grad_norm": 0.600765585899353,
      "learning_rate": 4.701183040688849e-05,
      "loss": 0.048,
      "num_input_tokens_seen": 11930768,
      "step": 6290
    },
    {
      "epoch": 1.0269189982869729,
      "grad_norm": 0.33678290247917175,
      "learning_rate": 4.700717428421573e-05,
      "loss": 0.048,
      "num_input_tokens_seen": 11941072,
      "step": 6295
    },
    {
      "epoch": 1.0277347255077902,
      "grad_norm": 1.0478671789169312,
      "learning_rate": 4.700251476776014e-05,
      "loss": 0.2077,
      "num_input_tokens_seen": 11950144,
      "step": 6300
    },
    {
      "epoch": 1.0285504527286076,
      "grad_norm": 1.3147079944610596,
      "learning_rate": 4.699785185824026e-05,
      "loss": 0.1937,
      "num_input_tokens_seen": 11959536,
      "step": 6305
    },
    {
      "epoch": 1.029366179949425,
      "grad_norm": 1.227594256401062,
      "learning_rate": 4.699318555637519e-05,
      "loss": 0.0907,
      "num_input_tokens_seen": 11968624,
      "step": 6310
    },
    {
      "epoch": 1.0301819071702423,
      "grad_norm": 0.14377592504024506,
      "learning_rate": 4.6988515862884525e-05,
      "loss": 0.2538,
      "num_input_tokens_seen": 11979248,
      "step": 6315
    },
    {
      "epoch": 1.0309976343910596,
      "grad_norm": 1.0071113109588623,
      "learning_rate": 4.698384277848838e-05,
      "loss": 0.1665,
      "num_input_tokens_seen": 11988912,
      "step": 6320
    },
    {
      "epoch": 1.031813361611877,
      "grad_norm": 0.5628362894058228,
      "learning_rate": 4.6979166303907425e-05,
      "loss": 0.206,
      "num_input_tokens_seen": 11998656,
      "step": 6325
    },
    {
      "epoch": 1.0326290888326943,
      "grad_norm": 1.17864990234375,
      "learning_rate": 4.697448643986281e-05,
      "loss": 0.0697,
      "num_input_tokens_seen": 12008816,
      "step": 6330
    },
    {
      "epoch": 1.0334448160535117,
      "grad_norm": 0.3375724256038666,
      "learning_rate": 4.696980318707624e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 12017968,
      "step": 6335
    },
    {
      "epoch": 1.034260543274329,
      "grad_norm": 0.24709728360176086,
      "learning_rate": 4.6965116546269924e-05,
      "loss": 0.1203,
      "num_input_tokens_seen": 12027456,
      "step": 6340
    },
    {
      "epoch": 1.0350762704951464,
      "grad_norm": 0.5686892867088318,
      "learning_rate": 4.6960426518166615e-05,
      "loss": 0.1104,
      "num_input_tokens_seen": 12037424,
      "step": 6345
    },
    {
      "epoch": 1.0358919977159637,
      "grad_norm": 0.7837275266647339,
      "learning_rate": 4.6955733103489556e-05,
      "loss": 0.0536,
      "num_input_tokens_seen": 12048048,
      "step": 6350
    },
    {
      "epoch": 1.036707724936781,
      "grad_norm": 0.1480250060558319,
      "learning_rate": 4.695103630296255e-05,
      "loss": 0.0647,
      "num_input_tokens_seen": 12058096,
      "step": 6355
    },
    {
      "epoch": 1.0375234521575984,
      "grad_norm": 2.7747058868408203,
      "learning_rate": 4.694633611730988e-05,
      "loss": 0.2821,
      "num_input_tokens_seen": 12067360,
      "step": 6360
    },
    {
      "epoch": 1.0383391793784158,
      "grad_norm": 1.217763066291809,
      "learning_rate": 4.694163254725639e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 12077680,
      "step": 6365
    },
    {
      "epoch": 1.0391549065992332,
      "grad_norm": 0.14074435830116272,
      "learning_rate": 4.693692559352743e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 12087552,
      "step": 6370
    },
    {
      "epoch": 1.0399706338200505,
      "grad_norm": 1.4889978170394897,
      "learning_rate": 4.693221525684886e-05,
      "loss": 0.0856,
      "num_input_tokens_seen": 12096432,
      "step": 6375
    },
    {
      "epoch": 1.0407863610408679,
      "grad_norm": 1.05909264087677,
      "learning_rate": 4.6927501537947084e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 12105600,
      "step": 6380
    },
    {
      "epoch": 1.0416020882616852,
      "grad_norm": 0.2163044661283493,
      "learning_rate": 4.692278443754901e-05,
      "loss": 0.1944,
      "num_input_tokens_seen": 12114992,
      "step": 6385
    },
    {
      "epoch": 1.0424178154825026,
      "grad_norm": 0.7352786660194397,
      "learning_rate": 4.691806395638208e-05,
      "loss": 0.1792,
      "num_input_tokens_seen": 12124496,
      "step": 6390
    },
    {
      "epoch": 1.04323354270332,
      "grad_norm": 0.5110735893249512,
      "learning_rate": 4.6913340095174255e-05,
      "loss": 0.2171,
      "num_input_tokens_seen": 12135424,
      "step": 6395
    },
    {
      "epoch": 1.0440492699241373,
      "grad_norm": 0.7959671020507812,
      "learning_rate": 4.690861285465399e-05,
      "loss": 0.2747,
      "num_input_tokens_seen": 12144016,
      "step": 6400
    },
    {
      "epoch": 1.0440492699241373,
      "eval_loss": 0.1572587937116623,
      "eval_runtime": 133.1846,
      "eval_samples_per_second": 20.46,
      "eval_steps_per_second": 10.234,
      "num_input_tokens_seen": 12144016,
      "step": 6400
    },
    {
      "epoch": 1.0448649971449546,
      "grad_norm": 1.8579497337341309,
      "learning_rate": 4.690388223555031e-05,
      "loss": 0.1455,
      "num_input_tokens_seen": 12153872,
      "step": 6405
    },
    {
      "epoch": 1.0456807243657722,
      "grad_norm": 0.05929253622889519,
      "learning_rate": 4.689914823859273e-05,
      "loss": 0.0815,
      "num_input_tokens_seen": 12164064,
      "step": 6410
    },
    {
      "epoch": 1.0464964515865895,
      "grad_norm": 0.777371883392334,
      "learning_rate": 4.689441086451129e-05,
      "loss": 0.0608,
      "num_input_tokens_seen": 12172144,
      "step": 6415
    },
    {
      "epoch": 1.047312178807407,
      "grad_norm": 0.42126280069351196,
      "learning_rate": 4.688967011403655e-05,
      "loss": 0.3606,
      "num_input_tokens_seen": 12182048,
      "step": 6420
    },
    {
      "epoch": 1.0481279060282243,
      "grad_norm": 0.29917168617248535,
      "learning_rate": 4.68849259878996e-05,
      "loss": 0.0331,
      "num_input_tokens_seen": 12192208,
      "step": 6425
    },
    {
      "epoch": 1.0489436332490416,
      "grad_norm": 1.942789077758789,
      "learning_rate": 4.6880178486832036e-05,
      "loss": 0.2126,
      "num_input_tokens_seen": 12201472,
      "step": 6430
    },
    {
      "epoch": 1.049759360469859,
      "grad_norm": 0.06164529547095299,
      "learning_rate": 4.687542761156598e-05,
      "loss": 0.0383,
      "num_input_tokens_seen": 12211136,
      "step": 6435
    },
    {
      "epoch": 1.0505750876906763,
      "grad_norm": 0.49921664595603943,
      "learning_rate": 4.6870673362834096e-05,
      "loss": 0.0755,
      "num_input_tokens_seen": 12219776,
      "step": 6440
    },
    {
      "epoch": 1.0513908149114937,
      "grad_norm": 0.3749704658985138,
      "learning_rate": 4.6865915741369526e-05,
      "loss": 0.0783,
      "num_input_tokens_seen": 12230384,
      "step": 6445
    },
    {
      "epoch": 1.052206542132311,
      "grad_norm": 1.0969359874725342,
      "learning_rate": 4.686115474790597e-05,
      "loss": 0.0859,
      "num_input_tokens_seen": 12239984,
      "step": 6450
    },
    {
      "epoch": 1.0530222693531284,
      "grad_norm": 0.557282567024231,
      "learning_rate": 4.685639038317762e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 12248336,
      "step": 6455
    },
    {
      "epoch": 1.0538379965739457,
      "grad_norm": 0.06710069626569748,
      "learning_rate": 4.685162264791921e-05,
      "loss": 0.0374,
      "num_input_tokens_seen": 12259312,
      "step": 6460
    },
    {
      "epoch": 1.054653723794763,
      "grad_norm": 0.6169060468673706,
      "learning_rate": 4.684685154286599e-05,
      "loss": 0.1162,
      "num_input_tokens_seen": 12268288,
      "step": 6465
    },
    {
      "epoch": 1.0554694510155804,
      "grad_norm": 1.1837488412857056,
      "learning_rate": 4.684207706875371e-05,
      "loss": 0.0901,
      "num_input_tokens_seen": 12277824,
      "step": 6470
    },
    {
      "epoch": 1.0562851782363978,
      "grad_norm": 0.553504467010498,
      "learning_rate": 4.683729922631866e-05,
      "loss": 0.0876,
      "num_input_tokens_seen": 12287520,
      "step": 6475
    },
    {
      "epoch": 1.0571009054572151,
      "grad_norm": 1.171808123588562,
      "learning_rate": 4.683251801629765e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 12297216,
      "step": 6480
    },
    {
      "epoch": 1.0579166326780325,
      "grad_norm": 0.8211436867713928,
      "learning_rate": 4.6827733439428e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 12307264,
      "step": 6485
    },
    {
      "epoch": 1.0587323598988498,
      "grad_norm": 0.08440938591957092,
      "learning_rate": 4.682294549644754e-05,
      "loss": 0.1666,
      "num_input_tokens_seen": 12317552,
      "step": 6490
    },
    {
      "epoch": 1.0595480871196672,
      "grad_norm": 0.0646049901843071,
      "learning_rate": 4.681815418809464e-05,
      "loss": 0.007,
      "num_input_tokens_seen": 12326496,
      "step": 6495
    },
    {
      "epoch": 1.0603638143404845,
      "grad_norm": 1.0287950038909912,
      "learning_rate": 4.681335951510819e-05,
      "loss": 0.0444,
      "num_input_tokens_seen": 12334848,
      "step": 6500
    },
    {
      "epoch": 1.061179541561302,
      "grad_norm": 2.07326602935791,
      "learning_rate": 4.6808561478227576e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 12345120,
      "step": 6505
    },
    {
      "epoch": 1.0619952687821193,
      "grad_norm": 0.7087347507476807,
      "learning_rate": 4.680376007819271e-05,
      "loss": 0.09,
      "num_input_tokens_seen": 12355264,
      "step": 6510
    },
    {
      "epoch": 1.0628109960029366,
      "grad_norm": 0.34535062313079834,
      "learning_rate": 4.679895531574405e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 12365408,
      "step": 6515
    },
    {
      "epoch": 1.063626723223754,
      "grad_norm": 1.866843819618225,
      "learning_rate": 4.679414719162253e-05,
      "loss": 0.2415,
      "num_input_tokens_seen": 12374368,
      "step": 6520
    },
    {
      "epoch": 1.0644424504445713,
      "grad_norm": 1.9818313121795654,
      "learning_rate": 4.6789335706569635e-05,
      "loss": 0.1502,
      "num_input_tokens_seen": 12383664,
      "step": 6525
    },
    {
      "epoch": 1.0652581776653887,
      "grad_norm": 0.6227132678031921,
      "learning_rate": 4.678452086132734e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 12393952,
      "step": 6530
    },
    {
      "epoch": 1.066073904886206,
      "grad_norm": 0.2108432948589325,
      "learning_rate": 4.677970265663818e-05,
      "loss": 0.1942,
      "num_input_tokens_seen": 12404464,
      "step": 6535
    },
    {
      "epoch": 1.0668896321070234,
      "grad_norm": 0.11784226447343826,
      "learning_rate": 4.677488109324517e-05,
      "loss": 0.0513,
      "num_input_tokens_seen": 12412880,
      "step": 6540
    },
    {
      "epoch": 1.0677053593278407,
      "grad_norm": 0.763379693031311,
      "learning_rate": 4.6770056171891846e-05,
      "loss": 0.0893,
      "num_input_tokens_seen": 12423600,
      "step": 6545
    },
    {
      "epoch": 1.068521086548658,
      "grad_norm": 0.13191822171211243,
      "learning_rate": 4.6765227893322286e-05,
      "loss": 0.0626,
      "num_input_tokens_seen": 12433296,
      "step": 6550
    },
    {
      "epoch": 1.0693368137694754,
      "grad_norm": 0.1632230579853058,
      "learning_rate": 4.676039625828107e-05,
      "loss": 0.3097,
      "num_input_tokens_seen": 12443104,
      "step": 6555
    },
    {
      "epoch": 1.0701525409902928,
      "grad_norm": 0.24727113544940948,
      "learning_rate": 4.675556126751328e-05,
      "loss": 0.1726,
      "num_input_tokens_seen": 12452800,
      "step": 6560
    },
    {
      "epoch": 1.0709682682111101,
      "grad_norm": 0.0550929419696331,
      "learning_rate": 4.6750722921764556e-05,
      "loss": 0.0356,
      "num_input_tokens_seen": 12463280,
      "step": 6565
    },
    {
      "epoch": 1.0717839954319275,
      "grad_norm": 0.19642959535121918,
      "learning_rate": 4.674588122178102e-05,
      "loss": 0.1798,
      "num_input_tokens_seen": 12472992,
      "step": 6570
    },
    {
      "epoch": 1.0725997226527448,
      "grad_norm": 0.6916045546531677,
      "learning_rate": 4.674103616830931e-05,
      "loss": 0.1933,
      "num_input_tokens_seen": 12483488,
      "step": 6575
    },
    {
      "epoch": 1.0734154498735622,
      "grad_norm": 0.34768596291542053,
      "learning_rate": 4.673618776209663e-05,
      "loss": 0.1615,
      "num_input_tokens_seen": 12493728,
      "step": 6580
    },
    {
      "epoch": 1.0742311770943795,
      "grad_norm": 1.2183055877685547,
      "learning_rate": 4.673133600389063e-05,
      "loss": 0.1735,
      "num_input_tokens_seen": 12502880,
      "step": 6585
    },
    {
      "epoch": 1.075046904315197,
      "grad_norm": 0.13018326461315155,
      "learning_rate": 4.672648089443953e-05,
      "loss": 0.0359,
      "num_input_tokens_seen": 12512160,
      "step": 6590
    },
    {
      "epoch": 1.0758626315360145,
      "grad_norm": 0.11285679042339325,
      "learning_rate": 4.672162243449204e-05,
      "loss": 0.1187,
      "num_input_tokens_seen": 12522432,
      "step": 6595
    },
    {
      "epoch": 1.0766783587568316,
      "grad_norm": 1.2818528413772583,
      "learning_rate": 4.67167606247974e-05,
      "loss": 0.1405,
      "num_input_tokens_seen": 12531776,
      "step": 6600
    },
    {
      "epoch": 1.0766783587568316,
      "eval_loss": 0.16131167113780975,
      "eval_runtime": 133.0846,
      "eval_samples_per_second": 20.476,
      "eval_steps_per_second": 10.242,
      "num_input_tokens_seen": 12531776,
      "step": 6600
    },
    {
      "epoch": 1.0774940859776492,
      "grad_norm": 1.402472734451294,
      "learning_rate": 4.671189546610536e-05,
      "loss": 0.0957,
      "num_input_tokens_seen": 12541728,
      "step": 6605
    },
    {
      "epoch": 1.0783098131984665,
      "grad_norm": 0.5086261034011841,
      "learning_rate": 4.67070269591662e-05,
      "loss": 0.0977,
      "num_input_tokens_seen": 12551008,
      "step": 6610
    },
    {
      "epoch": 1.0791255404192839,
      "grad_norm": 0.7135043144226074,
      "learning_rate": 4.670215510473068e-05,
      "loss": 0.0642,
      "num_input_tokens_seen": 12560784,
      "step": 6615
    },
    {
      "epoch": 1.0799412676401012,
      "grad_norm": 2.0689468383789062,
      "learning_rate": 4.669727990355013e-05,
      "loss": 0.0595,
      "num_input_tokens_seen": 12570064,
      "step": 6620
    },
    {
      "epoch": 1.0807569948609186,
      "grad_norm": 0.18358337879180908,
      "learning_rate": 4.669240135637635e-05,
      "loss": 0.2057,
      "num_input_tokens_seen": 12578400,
      "step": 6625
    },
    {
      "epoch": 1.081572722081736,
      "grad_norm": 0.6882326006889343,
      "learning_rate": 4.6687519463961675e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 12588784,
      "step": 6630
    },
    {
      "epoch": 1.0823884493025533,
      "grad_norm": 0.1923181116580963,
      "learning_rate": 4.668263422705896e-05,
      "loss": 0.0865,
      "num_input_tokens_seen": 12597680,
      "step": 6635
    },
    {
      "epoch": 1.0832041765233706,
      "grad_norm": 0.5121501088142395,
      "learning_rate": 4.667774564642156e-05,
      "loss": 0.3029,
      "num_input_tokens_seen": 12605952,
      "step": 6640
    },
    {
      "epoch": 1.084019903744188,
      "grad_norm": 0.21121592819690704,
      "learning_rate": 4.6672853722803365e-05,
      "loss": 0.0692,
      "num_input_tokens_seen": 12615984,
      "step": 6645
    },
    {
      "epoch": 1.0848356309650053,
      "grad_norm": 0.3963451385498047,
      "learning_rate": 4.666795845695877e-05,
      "loss": 0.1651,
      "num_input_tokens_seen": 12625408,
      "step": 6650
    },
    {
      "epoch": 1.0856513581858227,
      "grad_norm": 0.20872648060321808,
      "learning_rate": 4.666305984964269e-05,
      "loss": 0.12,
      "num_input_tokens_seen": 12634480,
      "step": 6655
    },
    {
      "epoch": 1.08646708540664,
      "grad_norm": 2.2857744693756104,
      "learning_rate": 4.6658157901610535e-05,
      "loss": 0.353,
      "num_input_tokens_seen": 12645120,
      "step": 6660
    },
    {
      "epoch": 1.0872828126274574,
      "grad_norm": 1.01844322681427,
      "learning_rate": 4.665325261361826e-05,
      "loss": 0.2437,
      "num_input_tokens_seen": 12655856,
      "step": 6665
    },
    {
      "epoch": 1.0880985398482748,
      "grad_norm": 0.5759185552597046,
      "learning_rate": 4.664834398642232e-05,
      "loss": 0.1779,
      "num_input_tokens_seen": 12665072,
      "step": 6670
    },
    {
      "epoch": 1.0889142670690921,
      "grad_norm": 0.24786219000816345,
      "learning_rate": 4.6643432020779686e-05,
      "loss": 0.1166,
      "num_input_tokens_seen": 12674656,
      "step": 6675
    },
    {
      "epoch": 1.0897299942899095,
      "grad_norm": 0.49337872862815857,
      "learning_rate": 4.663851671744786e-05,
      "loss": 0.1956,
      "num_input_tokens_seen": 12684464,
      "step": 6680
    },
    {
      "epoch": 1.0905457215107268,
      "grad_norm": 0.09585364907979965,
      "learning_rate": 4.6633598077184815e-05,
      "loss": 0.0526,
      "num_input_tokens_seen": 12693232,
      "step": 6685
    },
    {
      "epoch": 1.0913614487315442,
      "grad_norm": 2.151754379272461,
      "learning_rate": 4.662867610074908e-05,
      "loss": 0.1717,
      "num_input_tokens_seen": 12702976,
      "step": 6690
    },
    {
      "epoch": 1.0921771759523615,
      "grad_norm": 0.9544690251350403,
      "learning_rate": 4.6623750788899696e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 12711552,
      "step": 6695
    },
    {
      "epoch": 1.0929929031731789,
      "grad_norm": 0.8411957025527954,
      "learning_rate": 4.6618822142396195e-05,
      "loss": 0.2664,
      "num_input_tokens_seen": 12721472,
      "step": 6700
    },
    {
      "epoch": 1.0938086303939962,
      "grad_norm": 0.9287692904472351,
      "learning_rate": 4.661389016199864e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 12731664,
      "step": 6705
    },
    {
      "epoch": 1.0946243576148136,
      "grad_norm": 0.4490640461444855,
      "learning_rate": 4.660895484846761e-05,
      "loss": 0.1465,
      "num_input_tokens_seen": 12741584,
      "step": 6710
    },
    {
      "epoch": 1.095440084835631,
      "grad_norm": 0.8545411825180054,
      "learning_rate": 4.660401620256418e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 12750832,
      "step": 6715
    },
    {
      "epoch": 1.0962558120564483,
      "grad_norm": 0.6386632919311523,
      "learning_rate": 4.659907422504997e-05,
      "loss": 0.0876,
      "num_input_tokens_seen": 12760096,
      "step": 6720
    },
    {
      "epoch": 1.0970715392772656,
      "grad_norm": 0.3286909759044647,
      "learning_rate": 4.6594128916687074e-05,
      "loss": 0.0524,
      "num_input_tokens_seen": 12768912,
      "step": 6725
    },
    {
      "epoch": 1.097887266498083,
      "grad_norm": 0.7000786662101746,
      "learning_rate": 4.658918027823813e-05,
      "loss": 0.0528,
      "num_input_tokens_seen": 12777824,
      "step": 6730
    },
    {
      "epoch": 1.0987029937189003,
      "grad_norm": 0.36363014578819275,
      "learning_rate": 4.658422831046628e-05,
      "loss": 0.0977,
      "num_input_tokens_seen": 12786400,
      "step": 6735
    },
    {
      "epoch": 1.0995187209397177,
      "grad_norm": 0.08765624463558197,
      "learning_rate": 4.657927301413518e-05,
      "loss": 0.0642,
      "num_input_tokens_seen": 12796752,
      "step": 6740
    },
    {
      "epoch": 1.100334448160535,
      "grad_norm": 0.9510843753814697,
      "learning_rate": 4.657431439000901e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 12804160,
      "step": 6745
    },
    {
      "epoch": 1.1011501753813524,
      "grad_norm": 1.0912609100341797,
      "learning_rate": 4.656935243885243e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 12813152,
      "step": 6750
    },
    {
      "epoch": 1.1019659026021698,
      "grad_norm": 0.8184593915939331,
      "learning_rate": 4.656438716143066e-05,
      "loss": 0.0664,
      "num_input_tokens_seen": 12823552,
      "step": 6755
    },
    {
      "epoch": 1.102781629822987,
      "grad_norm": 0.6331333518028259,
      "learning_rate": 4.6559418558509384e-05,
      "loss": 0.0925,
      "num_input_tokens_seen": 12833056,
      "step": 6760
    },
    {
      "epoch": 1.1035973570438045,
      "grad_norm": 0.10526573657989502,
      "learning_rate": 4.6554446630854833e-05,
      "loss": 0.2066,
      "num_input_tokens_seen": 12841008,
      "step": 6765
    },
    {
      "epoch": 1.1044130842646218,
      "grad_norm": 2.018831253051758,
      "learning_rate": 4.654947137923374e-05,
      "loss": 0.2146,
      "num_input_tokens_seen": 12849920,
      "step": 6770
    },
    {
      "epoch": 1.1052288114854392,
      "grad_norm": 1.7221311330795288,
      "learning_rate": 4.654449280441335e-05,
      "loss": 0.0908,
      "num_input_tokens_seen": 12858192,
      "step": 6775
    },
    {
      "epoch": 1.1060445387062567,
      "grad_norm": 0.36942100524902344,
      "learning_rate": 4.653951090716143e-05,
      "loss": 0.1194,
      "num_input_tokens_seen": 12866720,
      "step": 6780
    },
    {
      "epoch": 1.1068602659270739,
      "grad_norm": 2.4701039791107178,
      "learning_rate": 4.653452568824625e-05,
      "loss": 0.3463,
      "num_input_tokens_seen": 12876064,
      "step": 6785
    },
    {
      "epoch": 1.1076759931478914,
      "grad_norm": 0.04732106626033783,
      "learning_rate": 4.6529537148436585e-05,
      "loss": 0.076,
      "num_input_tokens_seen": 12886032,
      "step": 6790
    },
    {
      "epoch": 1.1084917203687088,
      "grad_norm": 0.8774205446243286,
      "learning_rate": 4.6524545288501734e-05,
      "loss": 0.2372,
      "num_input_tokens_seen": 12895248,
      "step": 6795
    },
    {
      "epoch": 1.1093074475895262,
      "grad_norm": 0.7548792362213135,
      "learning_rate": 4.6519550109211506e-05,
      "loss": 0.2157,
      "num_input_tokens_seen": 12905136,
      "step": 6800
    },
    {
      "epoch": 1.1093074475895262,
      "eval_loss": 0.1580633670091629,
      "eval_runtime": 133.1276,
      "eval_samples_per_second": 20.469,
      "eval_steps_per_second": 10.238,
      "num_input_tokens_seen": 12905136,
      "step": 6800
    },
    {
      "epoch": 1.1101231748103435,
      "grad_norm": 0.0499277338385582,
      "learning_rate": 4.651455161133622e-05,
      "loss": 0.0813,
      "num_input_tokens_seen": 12913632,
      "step": 6805
    },
    {
      "epoch": 1.1109389020311609,
      "grad_norm": 0.15537236630916595,
      "learning_rate": 4.6509549795646704e-05,
      "loss": 0.0607,
      "num_input_tokens_seen": 12924320,
      "step": 6810
    },
    {
      "epoch": 1.1117546292519782,
      "grad_norm": 0.5966999530792236,
      "learning_rate": 4.6504544662914306e-05,
      "loss": 0.0866,
      "num_input_tokens_seen": 12933872,
      "step": 6815
    },
    {
      "epoch": 1.1125703564727956,
      "grad_norm": 1.3461285829544067,
      "learning_rate": 4.6499536213910876e-05,
      "loss": 0.2697,
      "num_input_tokens_seen": 12943104,
      "step": 6820
    },
    {
      "epoch": 1.113386083693613,
      "grad_norm": 0.5574921369552612,
      "learning_rate": 4.6494524449408786e-05,
      "loss": 0.2469,
      "num_input_tokens_seen": 12951888,
      "step": 6825
    },
    {
      "epoch": 1.1142018109144303,
      "grad_norm": 0.0942043587565422,
      "learning_rate": 4.6489509370180903e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 12960992,
      "step": 6830
    },
    {
      "epoch": 1.1150175381352476,
      "grad_norm": 0.8254799842834473,
      "learning_rate": 4.648449097700063e-05,
      "loss": 0.0984,
      "num_input_tokens_seen": 12968944,
      "step": 6835
    },
    {
      "epoch": 1.115833265356065,
      "grad_norm": 0.03599895164370537,
      "learning_rate": 4.647946927064185e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 12978304,
      "step": 6840
    },
    {
      "epoch": 1.1166489925768823,
      "grad_norm": 1.6734503507614136,
      "learning_rate": 4.647444425187898e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 12987280,
      "step": 6845
    },
    {
      "epoch": 1.1174647197976997,
      "grad_norm": 0.5390193462371826,
      "learning_rate": 4.646941592148695e-05,
      "loss": 0.2988,
      "num_input_tokens_seen": 12997312,
      "step": 6850
    },
    {
      "epoch": 1.118280447018517,
      "grad_norm": 0.3488016128540039,
      "learning_rate": 4.646438428024117e-05,
      "loss": 0.0456,
      "num_input_tokens_seen": 13006320,
      "step": 6855
    },
    {
      "epoch": 1.1190961742393344,
      "grad_norm": 0.39596712589263916,
      "learning_rate": 4.64593493289176e-05,
      "loss": 0.1987,
      "num_input_tokens_seen": 13015968,
      "step": 6860
    },
    {
      "epoch": 1.1199119014601517,
      "grad_norm": 0.5476969480514526,
      "learning_rate": 4.64543110682927e-05,
      "loss": 0.0588,
      "num_input_tokens_seen": 13025728,
      "step": 6865
    },
    {
      "epoch": 1.120727628680969,
      "grad_norm": 1.3948476314544678,
      "learning_rate": 4.644926949914341e-05,
      "loss": 0.229,
      "num_input_tokens_seen": 13036032,
      "step": 6870
    },
    {
      "epoch": 1.1215433559017864,
      "grad_norm": 2.1028831005096436,
      "learning_rate": 4.644422462224722e-05,
      "loss": 0.0969,
      "num_input_tokens_seen": 13045088,
      "step": 6875
    },
    {
      "epoch": 1.1223590831226038,
      "grad_norm": 0.502511739730835,
      "learning_rate": 4.643917643838211e-05,
      "loss": 0.0867,
      "num_input_tokens_seen": 13053456,
      "step": 6880
    },
    {
      "epoch": 1.1231748103434211,
      "grad_norm": 0.040949173271656036,
      "learning_rate": 4.6434124948326564e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 13062800,
      "step": 6885
    },
    {
      "epoch": 1.1239905375642385,
      "grad_norm": 1.016668677330017,
      "learning_rate": 4.6429070152859594e-05,
      "loss": 0.326,
      "num_input_tokens_seen": 13071328,
      "step": 6890
    },
    {
      "epoch": 1.1248062647850559,
      "grad_norm": 0.13752195239067078,
      "learning_rate": 4.6424012052760714e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 13081376,
      "step": 6895
    },
    {
      "epoch": 1.1256219920058732,
      "grad_norm": 0.7543440461158752,
      "learning_rate": 4.6418950648809945e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 13091408,
      "step": 6900
    },
    {
      "epoch": 1.1264377192266906,
      "grad_norm": 0.893277108669281,
      "learning_rate": 4.641388594178782e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 13101296,
      "step": 6905
    },
    {
      "epoch": 1.127253446447508,
      "grad_norm": 1.5033403635025024,
      "learning_rate": 4.640881793247538e-05,
      "loss": 0.1288,
      "num_input_tokens_seen": 13111664,
      "step": 6910
    },
    {
      "epoch": 1.1280691736683253,
      "grad_norm": 0.8267878890037537,
      "learning_rate": 4.6403746621654173e-05,
      "loss": 0.1787,
      "num_input_tokens_seen": 13121328,
      "step": 6915
    },
    {
      "epoch": 1.1288849008891426,
      "grad_norm": 1.1307928562164307,
      "learning_rate": 4.639867201010626e-05,
      "loss": 0.2198,
      "num_input_tokens_seen": 13129776,
      "step": 6920
    },
    {
      "epoch": 1.12970062810996,
      "grad_norm": 1.7915825843811035,
      "learning_rate": 4.6393594098614204e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 13138880,
      "step": 6925
    },
    {
      "epoch": 1.1305163553307773,
      "grad_norm": 0.08696417510509491,
      "learning_rate": 4.63885128879611e-05,
      "loss": 0.0578,
      "num_input_tokens_seen": 13147584,
      "step": 6930
    },
    {
      "epoch": 1.1313320825515947,
      "grad_norm": 0.13960354030132294,
      "learning_rate": 4.638342837893052e-05,
      "loss": 0.3246,
      "num_input_tokens_seen": 13157600,
      "step": 6935
    },
    {
      "epoch": 1.132147809772412,
      "grad_norm": 1.8813673257827759,
      "learning_rate": 4.6378340572306565e-05,
      "loss": 0.303,
      "num_input_tokens_seen": 13166032,
      "step": 6940
    },
    {
      "epoch": 1.1329635369932294,
      "grad_norm": 1.2023513317108154,
      "learning_rate": 4.6373249468873833e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 13174336,
      "step": 6945
    },
    {
      "epoch": 1.1337792642140467,
      "grad_norm": 1.2668007612228394,
      "learning_rate": 4.636815506941744e-05,
      "loss": 0.0839,
      "num_input_tokens_seen": 13184112,
      "step": 6950
    },
    {
      "epoch": 1.134594991434864,
      "grad_norm": 0.42393600940704346,
      "learning_rate": 4.6363057374723004e-05,
      "loss": 0.0874,
      "num_input_tokens_seen": 13193616,
      "step": 6955
    },
    {
      "epoch": 1.1354107186556814,
      "grad_norm": 0.23226363956928253,
      "learning_rate": 4.635795638557666e-05,
      "loss": 0.1494,
      "num_input_tokens_seen": 13203584,
      "step": 6960
    },
    {
      "epoch": 1.136226445876499,
      "grad_norm": 0.6411020755767822,
      "learning_rate": 4.635285210276504e-05,
      "loss": 0.1543,
      "num_input_tokens_seen": 13213680,
      "step": 6965
    },
    {
      "epoch": 1.1370421730973161,
      "grad_norm": 0.683789849281311,
      "learning_rate": 4.6347744527075295e-05,
      "loss": 0.1119,
      "num_input_tokens_seen": 13222368,
      "step": 6970
    },
    {
      "epoch": 1.1378579003181337,
      "grad_norm": 0.6513491868972778,
      "learning_rate": 4.634263365929506e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 13230112,
      "step": 6975
    },
    {
      "epoch": 1.1386736275389508,
      "grad_norm": 0.3134056329727173,
      "learning_rate": 4.6337519500212515e-05,
      "loss": 0.0871,
      "num_input_tokens_seen": 13239520,
      "step": 6980
    },
    {
      "epoch": 1.1394893547597684,
      "grad_norm": 1.1106996536254883,
      "learning_rate": 4.633240205061632e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 13249280,
      "step": 6985
    },
    {
      "epoch": 1.1403050819805858,
      "grad_norm": 0.9604779481887817,
      "learning_rate": 4.632728131129565e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 13258640,
      "step": 6990
    },
    {
      "epoch": 1.1411208092014031,
      "grad_norm": 0.7356439828872681,
      "learning_rate": 4.632215728304018e-05,
      "loss": 0.1739,
      "num_input_tokens_seen": 13268064,
      "step": 6995
    },
    {
      "epoch": 1.1419365364222205,
      "grad_norm": 0.10049532353878021,
      "learning_rate": 4.63170299666401e-05,
      "loss": 0.1939,
      "num_input_tokens_seen": 13278096,
      "step": 7000
    },
    {
      "epoch": 1.1419365364222205,
      "eval_loss": 0.15593278408050537,
      "eval_runtime": 133.2646,
      "eval_samples_per_second": 20.448,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 13278096,
      "step": 7000
    },
    {
      "epoch": 1.1427522636430378,
      "grad_norm": 1.3479995727539062,
      "learning_rate": 4.631189936288612e-05,
      "loss": 0.0942,
      "num_input_tokens_seen": 13287296,
      "step": 7005
    },
    {
      "epoch": 1.1435679908638552,
      "grad_norm": 0.20722174644470215,
      "learning_rate": 4.630676547256944e-05,
      "loss": 0.1041,
      "num_input_tokens_seen": 13296256,
      "step": 7010
    },
    {
      "epoch": 1.1443837180846725,
      "grad_norm": 0.5679796934127808,
      "learning_rate": 4.630162829648176e-05,
      "loss": 0.0828,
      "num_input_tokens_seen": 13305360,
      "step": 7015
    },
    {
      "epoch": 1.14519944530549,
      "grad_norm": 0.95994633436203,
      "learning_rate": 4.629648783541531e-05,
      "loss": 0.1695,
      "num_input_tokens_seen": 13315344,
      "step": 7020
    },
    {
      "epoch": 1.1460151725263072,
      "grad_norm": 0.7302584648132324,
      "learning_rate": 4.6291344090162804e-05,
      "loss": 0.0991,
      "num_input_tokens_seen": 13324240,
      "step": 7025
    },
    {
      "epoch": 1.1468308997471246,
      "grad_norm": 0.5945215225219727,
      "learning_rate": 4.628619706151748e-05,
      "loss": 0.2156,
      "num_input_tokens_seen": 13333696,
      "step": 7030
    },
    {
      "epoch": 1.147646626967942,
      "grad_norm": 0.46647679805755615,
      "learning_rate": 4.628104675027306e-05,
      "loss": 0.0711,
      "num_input_tokens_seen": 13344400,
      "step": 7035
    },
    {
      "epoch": 1.1484623541887593,
      "grad_norm": 0.8931382298469543,
      "learning_rate": 4.6275893157223805e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 13353760,
      "step": 7040
    },
    {
      "epoch": 1.1492780814095767,
      "grad_norm": 0.28624239563941956,
      "learning_rate": 4.627073628316445e-05,
      "loss": 0.1197,
      "num_input_tokens_seen": 13362368,
      "step": 7045
    },
    {
      "epoch": 1.150093808630394,
      "grad_norm": 0.9045050740242004,
      "learning_rate": 4.626557612889026e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 13371120,
      "step": 7050
    },
    {
      "epoch": 1.1509095358512114,
      "grad_norm": 0.6935630440711975,
      "learning_rate": 4.626041269519699e-05,
      "loss": 0.1094,
      "num_input_tokens_seen": 13380880,
      "step": 7055
    },
    {
      "epoch": 1.1517252630720287,
      "grad_norm": 1.179116129875183,
      "learning_rate": 4.6255245982880905e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 13389568,
      "step": 7060
    },
    {
      "epoch": 1.152540990292846,
      "grad_norm": 1.6875971555709839,
      "learning_rate": 4.625007599273879e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 13400592,
      "step": 7065
    },
    {
      "epoch": 1.1533567175136634,
      "grad_norm": 0.018898380920290947,
      "learning_rate": 4.6244902725567895e-05,
      "loss": 0.1117,
      "num_input_tokens_seen": 13409488,
      "step": 7070
    },
    {
      "epoch": 1.1541724447344808,
      "grad_norm": 1.7361962795257568,
      "learning_rate": 4.6239726182166024e-05,
      "loss": 0.0809,
      "num_input_tokens_seen": 13418176,
      "step": 7075
    },
    {
      "epoch": 1.1549881719552981,
      "grad_norm": 0.5767305493354797,
      "learning_rate": 4.623454636333147e-05,
      "loss": 0.0801,
      "num_input_tokens_seen": 13427120,
      "step": 7080
    },
    {
      "epoch": 1.1558038991761155,
      "grad_norm": 0.2444661259651184,
      "learning_rate": 4.622936326986301e-05,
      "loss": 0.0719,
      "num_input_tokens_seen": 13436784,
      "step": 7085
    },
    {
      "epoch": 1.1566196263969328,
      "grad_norm": 0.060789886862039566,
      "learning_rate": 4.6224176902559946e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 13446944,
      "step": 7090
    },
    {
      "epoch": 1.1574353536177502,
      "grad_norm": 1.2274229526519775,
      "learning_rate": 4.621898726222209e-05,
      "loss": 0.2404,
      "num_input_tokens_seen": 13456400,
      "step": 7095
    },
    {
      "epoch": 1.1582510808385675,
      "grad_norm": 0.028212135657668114,
      "learning_rate": 4.6213794349649744e-05,
      "loss": 0.032,
      "num_input_tokens_seen": 13464512,
      "step": 7100
    },
    {
      "epoch": 1.159066808059385,
      "grad_norm": 0.24338066577911377,
      "learning_rate": 4.6208598165643715e-05,
      "loss": 0.1091,
      "num_input_tokens_seen": 13473712,
      "step": 7105
    },
    {
      "epoch": 1.1598825352802022,
      "grad_norm": 0.09357479214668274,
      "learning_rate": 4.620339871100533e-05,
      "loss": 0.031,
      "num_input_tokens_seen": 13481984,
      "step": 7110
    },
    {
      "epoch": 1.1606982625010196,
      "grad_norm": 0.27188441157341003,
      "learning_rate": 4.6198195986536394e-05,
      "loss": 0.014,
      "num_input_tokens_seen": 13490992,
      "step": 7115
    },
    {
      "epoch": 1.161513989721837,
      "grad_norm": 0.4848259389400482,
      "learning_rate": 4.619298999303926e-05,
      "loss": 0.0906,
      "num_input_tokens_seen": 13500416,
      "step": 7120
    },
    {
      "epoch": 1.1623297169426543,
      "grad_norm": 0.36028987169265747,
      "learning_rate": 4.618778073131673e-05,
      "loss": 0.1844,
      "num_input_tokens_seen": 13509920,
      "step": 7125
    },
    {
      "epoch": 1.1631454441634717,
      "grad_norm": 0.5226706862449646,
      "learning_rate": 4.618256820217215e-05,
      "loss": 0.0517,
      "num_input_tokens_seen": 13520176,
      "step": 7130
    },
    {
      "epoch": 1.163961171384289,
      "grad_norm": 0.07614932954311371,
      "learning_rate": 4.617735240640936e-05,
      "loss": 0.0868,
      "num_input_tokens_seen": 13530496,
      "step": 7135
    },
    {
      "epoch": 1.1647768986051064,
      "grad_norm": 0.6566006541252136,
      "learning_rate": 4.6172133344832705e-05,
      "loss": 0.0958,
      "num_input_tokens_seen": 13539536,
      "step": 7140
    },
    {
      "epoch": 1.1655926258259237,
      "grad_norm": 0.9671867489814758,
      "learning_rate": 4.6166911018247004e-05,
      "loss": 0.2155,
      "num_input_tokens_seen": 13548976,
      "step": 7145
    },
    {
      "epoch": 1.1664083530467413,
      "grad_norm": 0.4146480858325958,
      "learning_rate": 4.616168542745764e-05,
      "loss": 0.0991,
      "num_input_tokens_seen": 13559184,
      "step": 7150
    },
    {
      "epoch": 1.1672240802675584,
      "grad_norm": 1.854737401008606,
      "learning_rate": 4.6156456573270446e-05,
      "loss": 0.2353,
      "num_input_tokens_seen": 13567072,
      "step": 7155
    },
    {
      "epoch": 1.168039807488376,
      "grad_norm": 0.09628637135028839,
      "learning_rate": 4.615122445649177e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 13575536,
      "step": 7160
    },
    {
      "epoch": 1.1688555347091931,
      "grad_norm": 2.233520984649658,
      "learning_rate": 4.6145989077928486e-05,
      "loss": 0.1298,
      "num_input_tokens_seen": 13585056,
      "step": 7165
    },
    {
      "epoch": 1.1696712619300107,
      "grad_norm": 0.12174728512763977,
      "learning_rate": 4.6140750438387953e-05,
      "loss": 0.0576,
      "num_input_tokens_seen": 13594432,
      "step": 7170
    },
    {
      "epoch": 1.170486989150828,
      "grad_norm": 0.13238859176635742,
      "learning_rate": 4.613550853867803e-05,
      "loss": 0.1683,
      "num_input_tokens_seen": 13603904,
      "step": 7175
    },
    {
      "epoch": 1.1713027163716454,
      "grad_norm": 0.237400084733963,
      "learning_rate": 4.613026337960708e-05,
      "loss": 0.173,
      "num_input_tokens_seen": 13614000,
      "step": 7180
    },
    {
      "epoch": 1.1721184435924628,
      "grad_norm": 1.337221384048462,
      "learning_rate": 4.612501496198398e-05,
      "loss": 0.0969,
      "num_input_tokens_seen": 13622560,
      "step": 7185
    },
    {
      "epoch": 1.17293417081328,
      "grad_norm": 2.1345083713531494,
      "learning_rate": 4.61197632866181e-05,
      "loss": 0.313,
      "num_input_tokens_seen": 13632560,
      "step": 7190
    },
    {
      "epoch": 1.1737498980340975,
      "grad_norm": 0.3824680745601654,
      "learning_rate": 4.611450835431931e-05,
      "loss": 0.0186,
      "num_input_tokens_seen": 13641664,
      "step": 7195
    },
    {
      "epoch": 1.1745656252549148,
      "grad_norm": 0.5926434397697449,
      "learning_rate": 4.6109250165898e-05,
      "loss": 0.2038,
      "num_input_tokens_seen": 13651520,
      "step": 7200
    },
    {
      "epoch": 1.1745656252549148,
      "eval_loss": 0.1633707880973816,
      "eval_runtime": 133.1473,
      "eval_samples_per_second": 20.466,
      "eval_steps_per_second": 10.237,
      "num_input_tokens_seen": 13651520,
      "step": 7200
    },
    {
      "epoch": 1.1753813524757322,
      "grad_norm": 0.38543272018432617,
      "learning_rate": 4.610398872216503e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 13661328,
      "step": 7205
    },
    {
      "epoch": 1.1761970796965495,
      "grad_norm": 0.07557795196771622,
      "learning_rate": 4.6098724023931796e-05,
      "loss": 0.1418,
      "num_input_tokens_seen": 13671472,
      "step": 7210
    },
    {
      "epoch": 1.1770128069173669,
      "grad_norm": 0.08296660333871841,
      "learning_rate": 4.609345607201017e-05,
      "loss": 0.2059,
      "num_input_tokens_seen": 13679856,
      "step": 7215
    },
    {
      "epoch": 1.1778285341381842,
      "grad_norm": 0.3842545747756958,
      "learning_rate": 4.608818486721254e-05,
      "loss": 0.0839,
      "num_input_tokens_seen": 13689088,
      "step": 7220
    },
    {
      "epoch": 1.1786442613590016,
      "grad_norm": 1.2289690971374512,
      "learning_rate": 4.608291041035179e-05,
      "loss": 0.2341,
      "num_input_tokens_seen": 13698320,
      "step": 7225
    },
    {
      "epoch": 1.179459988579819,
      "grad_norm": 0.46621131896972656,
      "learning_rate": 4.607763270224132e-05,
      "loss": 0.057,
      "num_input_tokens_seen": 13707680,
      "step": 7230
    },
    {
      "epoch": 1.1802757158006363,
      "grad_norm": 0.8411877155303955,
      "learning_rate": 4.6072351743695e-05,
      "loss": 0.1089,
      "num_input_tokens_seen": 13716784,
      "step": 7235
    },
    {
      "epoch": 1.1810914430214536,
      "grad_norm": 0.42788830399513245,
      "learning_rate": 4.606706753552723e-05,
      "loss": 0.1767,
      "num_input_tokens_seen": 13727296,
      "step": 7240
    },
    {
      "epoch": 1.181907170242271,
      "grad_norm": 0.44552212953567505,
      "learning_rate": 4.6061780078552906e-05,
      "loss": 0.0732,
      "num_input_tokens_seen": 13737136,
      "step": 7245
    },
    {
      "epoch": 1.1827228974630883,
      "grad_norm": 0.41177570819854736,
      "learning_rate": 4.605648937358742e-05,
      "loss": 0.059,
      "num_input_tokens_seen": 13747088,
      "step": 7250
    },
    {
      "epoch": 1.1835386246839057,
      "grad_norm": 1.2295007705688477,
      "learning_rate": 4.605119542144665e-05,
      "loss": 0.1648,
      "num_input_tokens_seen": 13756608,
      "step": 7255
    },
    {
      "epoch": 1.184354351904723,
      "grad_norm": 0.9968419075012207,
      "learning_rate": 4.604589822294701e-05,
      "loss": 0.2168,
      "num_input_tokens_seen": 13764720,
      "step": 7260
    },
    {
      "epoch": 1.1851700791255404,
      "grad_norm": 0.41950246691703796,
      "learning_rate": 4.604059777890537e-05,
      "loss": 0.1136,
      "num_input_tokens_seen": 13774400,
      "step": 7265
    },
    {
      "epoch": 1.1859858063463578,
      "grad_norm": 1.2471587657928467,
      "learning_rate": 4.6035294090139145e-05,
      "loss": 0.2548,
      "num_input_tokens_seen": 13783824,
      "step": 7270
    },
    {
      "epoch": 1.186801533567175,
      "grad_norm": 1.237795352935791,
      "learning_rate": 4.6029987157466226e-05,
      "loss": 0.2405,
      "num_input_tokens_seen": 13793984,
      "step": 7275
    },
    {
      "epoch": 1.1876172607879925,
      "grad_norm": 0.4577502906322479,
      "learning_rate": 4.602467698170502e-05,
      "loss": 0.2961,
      "num_input_tokens_seen": 13805360,
      "step": 7280
    },
    {
      "epoch": 1.1884329880088098,
      "grad_norm": 0.15491031110286713,
      "learning_rate": 4.601936356367439e-05,
      "loss": 0.271,
      "num_input_tokens_seen": 13815008,
      "step": 7285
    },
    {
      "epoch": 1.1892487152296272,
      "grad_norm": 2.369436025619507,
      "learning_rate": 4.601404690419377e-05,
      "loss": 0.1974,
      "num_input_tokens_seen": 13825008,
      "step": 7290
    },
    {
      "epoch": 1.1900644424504445,
      "grad_norm": 0.7513375878334045,
      "learning_rate": 4.600872700408303e-05,
      "loss": 0.1777,
      "num_input_tokens_seen": 13833664,
      "step": 7295
    },
    {
      "epoch": 1.1908801696712619,
      "grad_norm": 1.85165536403656,
      "learning_rate": 4.600340386416258e-05,
      "loss": 0.2572,
      "num_input_tokens_seen": 13843312,
      "step": 7300
    },
    {
      "epoch": 1.1916958968920792,
      "grad_norm": 0.7860260605812073,
      "learning_rate": 4.5998077485253296e-05,
      "loss": 0.246,
      "num_input_tokens_seen": 13851408,
      "step": 7305
    },
    {
      "epoch": 1.1925116241128966,
      "grad_norm": 1.5950515270233154,
      "learning_rate": 4.59927478681766e-05,
      "loss": 0.3451,
      "num_input_tokens_seen": 13860928,
      "step": 7310
    },
    {
      "epoch": 1.193327351333714,
      "grad_norm": 0.20750626921653748,
      "learning_rate": 4.5987415013754366e-05,
      "loss": 0.0989,
      "num_input_tokens_seen": 13871168,
      "step": 7315
    },
    {
      "epoch": 1.1941430785545313,
      "grad_norm": 2.562544107437134,
      "learning_rate": 4.598207892280899e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 13882448,
      "step": 7320
    },
    {
      "epoch": 1.1949588057753486,
      "grad_norm": 0.30416417121887207,
      "learning_rate": 4.597673959616337e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 13891344,
      "step": 7325
    },
    {
      "epoch": 1.195774532996166,
      "grad_norm": 0.34077897667884827,
      "learning_rate": 4.597139703464089e-05,
      "loss": 0.0657,
      "num_input_tokens_seen": 13900736,
      "step": 7330
    },
    {
      "epoch": 1.1965902602169836,
      "grad_norm": 0.08822479099035263,
      "learning_rate": 4.596605123906545e-05,
      "loss": 0.3245,
      "num_input_tokens_seen": 13910960,
      "step": 7335
    },
    {
      "epoch": 1.1974059874378007,
      "grad_norm": 0.06774789094924927,
      "learning_rate": 4.596070221026143e-05,
      "loss": 0.0905,
      "num_input_tokens_seen": 13920304,
      "step": 7340
    },
    {
      "epoch": 1.1982217146586183,
      "grad_norm": 0.4950636327266693,
      "learning_rate": 4.595534994905372e-05,
      "loss": 0.096,
      "num_input_tokens_seen": 13930288,
      "step": 7345
    },
    {
      "epoch": 1.1990374418794354,
      "grad_norm": 0.13294747471809387,
      "learning_rate": 4.594999445626771e-05,
      "loss": 0.0232,
      "num_input_tokens_seen": 13939872,
      "step": 7350
    },
    {
      "epoch": 1.199853169100253,
      "grad_norm": 1.3869633674621582,
      "learning_rate": 4.5944635732729276e-05,
      "loss": 0.2366,
      "num_input_tokens_seen": 13948784,
      "step": 7355
    },
    {
      "epoch": 1.2006688963210703,
      "grad_norm": 0.6721004843711853,
      "learning_rate": 4.5939273779264804e-05,
      "loss": 0.2055,
      "num_input_tokens_seen": 13957840,
      "step": 7360
    },
    {
      "epoch": 1.2014846235418877,
      "grad_norm": 1.801819920539856,
      "learning_rate": 4.593390859670118e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 13967392,
      "step": 7365
    },
    {
      "epoch": 1.202300350762705,
      "grad_norm": 2.4319937229156494,
      "learning_rate": 4.5928540185865776e-05,
      "loss": 0.2683,
      "num_input_tokens_seen": 13976144,
      "step": 7370
    },
    {
      "epoch": 1.2031160779835224,
      "grad_norm": 1.1854373216629028,
      "learning_rate": 4.592316854758648e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 13985952,
      "step": 7375
    },
    {
      "epoch": 1.2039318052043397,
      "grad_norm": 1.4805951118469238,
      "learning_rate": 4.5917793682691646e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 13995520,
      "step": 7380
    },
    {
      "epoch": 1.204747532425157,
      "grad_norm": 0.10336259007453918,
      "learning_rate": 4.5912415592010164e-05,
      "loss": 0.0778,
      "num_input_tokens_seen": 14004496,
      "step": 7385
    },
    {
      "epoch": 1.2055632596459744,
      "grad_norm": 2.014683485031128,
      "learning_rate": 4.5907034276371386e-05,
      "loss": 0.3091,
      "num_input_tokens_seen": 14014368,
      "step": 7390
    },
    {
      "epoch": 1.2063789868667918,
      "grad_norm": 0.4054104685783386,
      "learning_rate": 4.5901649736605196e-05,
      "loss": 0.113,
      "num_input_tokens_seen": 14024464,
      "step": 7395
    },
    {
      "epoch": 1.2071947140876091,
      "grad_norm": 0.4783731997013092,
      "learning_rate": 4.589626197354195e-05,
      "loss": 0.0898,
      "num_input_tokens_seen": 14034784,
      "step": 7400
    },
    {
      "epoch": 1.2071947140876091,
      "eval_loss": 0.15798258781433105,
      "eval_runtime": 133.2196,
      "eval_samples_per_second": 20.455,
      "eval_steps_per_second": 10.231,
      "num_input_tokens_seen": 14034784,
      "step": 7400
    },
    {
      "epoch": 1.2080104413084265,
      "grad_norm": 1.19206964969635,
      "learning_rate": 4.5890870988012504e-05,
      "loss": 0.1013,
      "num_input_tokens_seen": 14043504,
      "step": 7405
    },
    {
      "epoch": 1.2088261685292438,
      "grad_norm": 1.072680115699768,
      "learning_rate": 4.5885476780848226e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 14053552,
      "step": 7410
    },
    {
      "epoch": 1.2096418957500612,
      "grad_norm": 0.4505373239517212,
      "learning_rate": 4.5880079352880964e-05,
      "loss": 0.1227,
      "num_input_tokens_seen": 14063360,
      "step": 7415
    },
    {
      "epoch": 1.2104576229708786,
      "grad_norm": 0.3921409249305725,
      "learning_rate": 4.5874678704943065e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 14073088,
      "step": 7420
    },
    {
      "epoch": 1.211273350191696,
      "grad_norm": 2.258009433746338,
      "learning_rate": 4.5869274837867394e-05,
      "loss": 0.1908,
      "num_input_tokens_seen": 14083216,
      "step": 7425
    },
    {
      "epoch": 1.2120890774125133,
      "grad_norm": 1.392492413520813,
      "learning_rate": 4.5863867752487275e-05,
      "loss": 0.2928,
      "num_input_tokens_seen": 14092448,
      "step": 7430
    },
    {
      "epoch": 1.2129048046333306,
      "grad_norm": 2.1462178230285645,
      "learning_rate": 4.5858457449636554e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 14102896,
      "step": 7435
    },
    {
      "epoch": 1.213720531854148,
      "grad_norm": 0.7163873314857483,
      "learning_rate": 4.5853043930149574e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 14111936,
      "step": 7440
    },
    {
      "epoch": 1.2145362590749653,
      "grad_norm": 0.3341209888458252,
      "learning_rate": 4.584762719486117e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 14122048,
      "step": 7445
    },
    {
      "epoch": 1.2153519862957827,
      "grad_norm": 1.6279914379119873,
      "learning_rate": 4.584220724460665e-05,
      "loss": 0.2451,
      "num_input_tokens_seen": 14132288,
      "step": 7450
    },
    {
      "epoch": 1.2161677135166,
      "grad_norm": 0.3485069274902344,
      "learning_rate": 4.5836784080221865e-05,
      "loss": 0.1279,
      "num_input_tokens_seen": 14141728,
      "step": 7455
    },
    {
      "epoch": 1.2169834407374174,
      "grad_norm": 0.44629594683647156,
      "learning_rate": 4.583135770254312e-05,
      "loss": 0.1094,
      "num_input_tokens_seen": 14151664,
      "step": 7460
    },
    {
      "epoch": 1.2177991679582347,
      "grad_norm": 0.9573984742164612,
      "learning_rate": 4.5825928112407236e-05,
      "loss": 0.2494,
      "num_input_tokens_seen": 14161008,
      "step": 7465
    },
    {
      "epoch": 1.218614895179052,
      "grad_norm": 1.6928529739379883,
      "learning_rate": 4.582049531065152e-05,
      "loss": 0.2942,
      "num_input_tokens_seen": 14170256,
      "step": 7470
    },
    {
      "epoch": 1.2194306223998694,
      "grad_norm": 0.5023638010025024,
      "learning_rate": 4.5815059298113783e-05,
      "loss": 0.2446,
      "num_input_tokens_seen": 14179136,
      "step": 7475
    },
    {
      "epoch": 1.2202463496206868,
      "grad_norm": 0.6636266708374023,
      "learning_rate": 4.580962007563232e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 14188224,
      "step": 7480
    },
    {
      "epoch": 1.2210620768415041,
      "grad_norm": 0.12146032601594925,
      "learning_rate": 4.5804177644045935e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 14197440,
      "step": 7485
    },
    {
      "epoch": 1.2218778040623215,
      "grad_norm": 0.06703417003154755,
      "learning_rate": 4.579873200419391e-05,
      "loss": 0.1924,
      "num_input_tokens_seen": 14206736,
      "step": 7490
    },
    {
      "epoch": 1.2226935312831388,
      "grad_norm": 0.9529284238815308,
      "learning_rate": 4.5793283156916046e-05,
      "loss": 0.2033,
      "num_input_tokens_seen": 14215200,
      "step": 7495
    },
    {
      "epoch": 1.2235092585039562,
      "grad_norm": 0.2079615741968155,
      "learning_rate": 4.578783110305261e-05,
      "loss": 0.204,
      "num_input_tokens_seen": 14224992,
      "step": 7500
    },
    {
      "epoch": 1.2243249857247736,
      "grad_norm": 0.6667476892471313,
      "learning_rate": 4.578237584344438e-05,
      "loss": 0.0922,
      "num_input_tokens_seen": 14235520,
      "step": 7505
    },
    {
      "epoch": 1.225140712945591,
      "grad_norm": 0.08030374348163605,
      "learning_rate": 4.577691737893263e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 14244880,
      "step": 7510
    },
    {
      "epoch": 1.2259564401664083,
      "grad_norm": 0.5367437601089478,
      "learning_rate": 4.577145571035912e-05,
      "loss": 0.1099,
      "num_input_tokens_seen": 14254352,
      "step": 7515
    },
    {
      "epoch": 1.2267721673872258,
      "grad_norm": 0.12834766507148743,
      "learning_rate": 4.576599083856611e-05,
      "loss": 0.1945,
      "num_input_tokens_seen": 14263600,
      "step": 7520
    },
    {
      "epoch": 1.227587894608043,
      "grad_norm": 0.794694185256958,
      "learning_rate": 4.576052276439635e-05,
      "loss": 0.1087,
      "num_input_tokens_seen": 14272400,
      "step": 7525
    },
    {
      "epoch": 1.2284036218288605,
      "grad_norm": 0.31698909401893616,
      "learning_rate": 4.575505148869308e-05,
      "loss": 0.2017,
      "num_input_tokens_seen": 14281872,
      "step": 7530
    },
    {
      "epoch": 1.2292193490496777,
      "grad_norm": 1.330129623413086,
      "learning_rate": 4.574957701230006e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 14290656,
      "step": 7535
    },
    {
      "epoch": 1.2300350762704952,
      "grad_norm": 0.41996774077415466,
      "learning_rate": 4.57440993360615e-05,
      "loss": 0.0785,
      "num_input_tokens_seen": 14300048,
      "step": 7540
    },
    {
      "epoch": 1.2308508034913126,
      "grad_norm": 0.21461711823940277,
      "learning_rate": 4.5738618460822134e-05,
      "loss": 0.1159,
      "num_input_tokens_seen": 14309456,
      "step": 7545
    },
    {
      "epoch": 1.23166653071213,
      "grad_norm": 1.5688306093215942,
      "learning_rate": 4.573313438742719e-05,
      "loss": 0.1668,
      "num_input_tokens_seen": 14318816,
      "step": 7550
    },
    {
      "epoch": 1.2324822579329473,
      "grad_norm": 1.4762049913406372,
      "learning_rate": 4.5727647116722374e-05,
      "loss": 0.1366,
      "num_input_tokens_seen": 14328752,
      "step": 7555
    },
    {
      "epoch": 1.2332979851537647,
      "grad_norm": 1.471588134765625,
      "learning_rate": 4.5722156649553884e-05,
      "loss": 0.193,
      "num_input_tokens_seen": 14338336,
      "step": 7560
    },
    {
      "epoch": 1.234113712374582,
      "grad_norm": 0.890152633190155,
      "learning_rate": 4.571666298676843e-05,
      "loss": 0.2201,
      "num_input_tokens_seen": 14348464,
      "step": 7565
    },
    {
      "epoch": 1.2349294395953994,
      "grad_norm": 0.4643767774105072,
      "learning_rate": 4.571116612921321e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 14357392,
      "step": 7570
    },
    {
      "epoch": 1.2357451668162167,
      "grad_norm": 0.12028701603412628,
      "learning_rate": 4.57056660777359e-05,
      "loss": 0.0359,
      "num_input_tokens_seen": 14368112,
      "step": 7575
    },
    {
      "epoch": 1.236560894037034,
      "grad_norm": 0.3906267285346985,
      "learning_rate": 4.5700162833184666e-05,
      "loss": 0.0593,
      "num_input_tokens_seen": 14377360,
      "step": 7580
    },
    {
      "epoch": 1.2373766212578514,
      "grad_norm": 0.512406051158905,
      "learning_rate": 4.5694656396408195e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 14386528,
      "step": 7585
    },
    {
      "epoch": 1.2381923484786688,
      "grad_norm": 1.8639320135116577,
      "learning_rate": 4.5689146768255646e-05,
      "loss": 0.2269,
      "num_input_tokens_seen": 14396800,
      "step": 7590
    },
    {
      "epoch": 1.2390080756994861,
      "grad_norm": 1.712605595588684,
      "learning_rate": 4.568363394957667e-05,
      "loss": 0.2664,
      "num_input_tokens_seen": 14406016,
      "step": 7595
    },
    {
      "epoch": 1.2398238029203035,
      "grad_norm": 1.7610973119735718,
      "learning_rate": 4.567811794122141e-05,
      "loss": 0.2486,
      "num_input_tokens_seen": 14415120,
      "step": 7600
    },
    {
      "epoch": 1.2398238029203035,
      "eval_loss": 0.15297992527484894,
      "eval_runtime": 133.3013,
      "eval_samples_per_second": 20.442,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 14415120,
      "step": 7600
    },
    {
      "epoch": 1.2406395301411208,
      "grad_norm": 1.5016933679580688,
      "learning_rate": 4.56725987440405e-05,
      "loss": 0.1567,
      "num_input_tokens_seen": 14424384,
      "step": 7605
    },
    {
      "epoch": 1.2414552573619382,
      "grad_norm": 0.5919935703277588,
      "learning_rate": 4.566707635888508e-05,
      "loss": 0.1706,
      "num_input_tokens_seen": 14433456,
      "step": 7610
    },
    {
      "epoch": 1.2422709845827555,
      "grad_norm": 1.3191584348678589,
      "learning_rate": 4.566155078660677e-05,
      "loss": 0.1888,
      "num_input_tokens_seen": 14441536,
      "step": 7615
    },
    {
      "epoch": 1.2430867118035729,
      "grad_norm": 0.3669265806674957,
      "learning_rate": 4.565602202805768e-05,
      "loss": 0.0829,
      "num_input_tokens_seen": 14451152,
      "step": 7620
    },
    {
      "epoch": 1.2439024390243902,
      "grad_norm": 0.396983802318573,
      "learning_rate": 4.56504900840904e-05,
      "loss": 0.1173,
      "num_input_tokens_seen": 14460112,
      "step": 7625
    },
    {
      "epoch": 1.2447181662452076,
      "grad_norm": 0.4038791358470917,
      "learning_rate": 4.564495495555805e-05,
      "loss": 0.0782,
      "num_input_tokens_seen": 14468928,
      "step": 7630
    },
    {
      "epoch": 1.245533893466025,
      "grad_norm": 0.23420225083827972,
      "learning_rate": 4.5639416643314204e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 14478736,
      "step": 7635
    },
    {
      "epoch": 1.2463496206868423,
      "grad_norm": 0.42850324511528015,
      "learning_rate": 4.5633875148212946e-05,
      "loss": 0.0534,
      "num_input_tokens_seen": 14488944,
      "step": 7640
    },
    {
      "epoch": 1.2471653479076596,
      "grad_norm": 2.716935634613037,
      "learning_rate": 4.562833047110883e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 14498928,
      "step": 7645
    },
    {
      "epoch": 1.247981075128477,
      "grad_norm": 1.0866584777832031,
      "learning_rate": 4.5622782612856923e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 14509232,
      "step": 7650
    },
    {
      "epoch": 1.2487968023492944,
      "grad_norm": 0.8575872182846069,
      "learning_rate": 4.561723157431278e-05,
      "loss": 0.1216,
      "num_input_tokens_seen": 14517984,
      "step": 7655
    },
    {
      "epoch": 1.2496125295701117,
      "grad_norm": 1.3950001001358032,
      "learning_rate": 4.5611677356332435e-05,
      "loss": 0.2841,
      "num_input_tokens_seen": 14527168,
      "step": 7660
    },
    {
      "epoch": 1.250428256790929,
      "grad_norm": 1.2596056461334229,
      "learning_rate": 4.560611995977242e-05,
      "loss": 0.2643,
      "num_input_tokens_seen": 14536528,
      "step": 7665
    },
    {
      "epoch": 1.2512439840117464,
      "grad_norm": 1.5097134113311768,
      "learning_rate": 4.560055938548975e-05,
      "loss": 0.2699,
      "num_input_tokens_seen": 14547456,
      "step": 7670
    },
    {
      "epoch": 1.2520597112325638,
      "grad_norm": 0.2644523084163666,
      "learning_rate": 4.5594995634341944e-05,
      "loss": 0.1995,
      "num_input_tokens_seen": 14556256,
      "step": 7675
    },
    {
      "epoch": 1.2528754384533811,
      "grad_norm": 0.6655407547950745,
      "learning_rate": 4.5589428707187e-05,
      "loss": 0.0313,
      "num_input_tokens_seen": 14566160,
      "step": 7680
    },
    {
      "epoch": 1.2536911656741985,
      "grad_norm": 0.4726583957672119,
      "learning_rate": 4.55838586048834e-05,
      "loss": 0.0921,
      "num_input_tokens_seen": 14575328,
      "step": 7685
    },
    {
      "epoch": 1.2545068928950158,
      "grad_norm": 0.7194830179214478,
      "learning_rate": 4.557828532829013e-05,
      "loss": 0.2383,
      "num_input_tokens_seen": 14586128,
      "step": 7690
    },
    {
      "epoch": 1.2553226201158334,
      "grad_norm": 0.832267165184021,
      "learning_rate": 4.557270887826667e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 14595552,
      "step": 7695
    },
    {
      "epoch": 1.2561383473366505,
      "grad_norm": 2.13584566116333,
      "learning_rate": 4.556712925567296e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 14605328,
      "step": 7700
    },
    {
      "epoch": 1.256954074557468,
      "grad_norm": 0.1929861307144165,
      "learning_rate": 4.5561546461369454e-05,
      "loss": 0.1021,
      "num_input_tokens_seen": 14615216,
      "step": 7705
    },
    {
      "epoch": 1.2577698017782852,
      "grad_norm": 0.08393897116184235,
      "learning_rate": 4.55559604962171e-05,
      "loss": 0.1145,
      "num_input_tokens_seen": 14623888,
      "step": 7710
    },
    {
      "epoch": 1.2585855289991028,
      "grad_norm": 0.2816116213798523,
      "learning_rate": 4.55503713610773e-05,
      "loss": 0.1263,
      "num_input_tokens_seen": 14632256,
      "step": 7715
    },
    {
      "epoch": 1.25940125621992,
      "grad_norm": 0.8714724779129028,
      "learning_rate": 4.5544779056812e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 14642672,
      "step": 7720
    },
    {
      "epoch": 1.2602169834407375,
      "grad_norm": 0.9867217540740967,
      "learning_rate": 4.553918358428358e-05,
      "loss": 0.1666,
      "num_input_tokens_seen": 14651536,
      "step": 7725
    },
    {
      "epoch": 1.2610327106615546,
      "grad_norm": 0.30517593026161194,
      "learning_rate": 4.553358494435494e-05,
      "loss": 0.0983,
      "num_input_tokens_seen": 14660912,
      "step": 7730
    },
    {
      "epoch": 1.2618484378823722,
      "grad_norm": 1.3997306823730469,
      "learning_rate": 4.5527983137889464e-05,
      "loss": 0.2685,
      "num_input_tokens_seen": 14670864,
      "step": 7735
    },
    {
      "epoch": 1.2626641651031896,
      "grad_norm": 0.9592278599739075,
      "learning_rate": 4.5522378165751015e-05,
      "loss": 0.0742,
      "num_input_tokens_seen": 14680144,
      "step": 7740
    },
    {
      "epoch": 1.263479892324007,
      "grad_norm": 0.07577800750732422,
      "learning_rate": 4.5516770028803954e-05,
      "loss": 0.1503,
      "num_input_tokens_seen": 14690544,
      "step": 7745
    },
    {
      "epoch": 1.2642956195448243,
      "grad_norm": 1.0535699129104614,
      "learning_rate": 4.5511158727913116e-05,
      "loss": 0.2158,
      "num_input_tokens_seen": 14698480,
      "step": 7750
    },
    {
      "epoch": 1.2651113467656416,
      "grad_norm": 0.5400977730751038,
      "learning_rate": 4.5505544263943856e-05,
      "loss": 0.3181,
      "num_input_tokens_seen": 14708272,
      "step": 7755
    },
    {
      "epoch": 1.265927073986459,
      "grad_norm": 0.09701835364103317,
      "learning_rate": 4.549992663776197e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 14717584,
      "step": 7760
    },
    {
      "epoch": 1.2667428012072763,
      "grad_norm": 1.3537999391555786,
      "learning_rate": 4.5494305850233786e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 14727248,
      "step": 7765
    },
    {
      "epoch": 1.2675585284280937,
      "grad_norm": 1.0374995470046997,
      "learning_rate": 4.5488681902226094e-05,
      "loss": 0.1532,
      "num_input_tokens_seen": 14736752,
      "step": 7770
    },
    {
      "epoch": 1.268374255648911,
      "grad_norm": 0.5626136660575867,
      "learning_rate": 4.5483054794606174e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 14748256,
      "step": 7775
    },
    {
      "epoch": 1.2691899828697284,
      "grad_norm": 0.25482186675071716,
      "learning_rate": 4.547742452824179e-05,
      "loss": 0.092,
      "num_input_tokens_seen": 14757280,
      "step": 7780
    },
    {
      "epoch": 1.2700057100905457,
      "grad_norm": 0.08093944191932678,
      "learning_rate": 4.5471791104001215e-05,
      "loss": 0.187,
      "num_input_tokens_seen": 14766992,
      "step": 7785
    },
    {
      "epoch": 1.270821437311363,
      "grad_norm": 2.231778144836426,
      "learning_rate": 4.546615452275319e-05,
      "loss": 0.1974,
      "num_input_tokens_seen": 14776592,
      "step": 7790
    },
    {
      "epoch": 1.2716371645321805,
      "grad_norm": 0.572291910648346,
      "learning_rate": 4.5460514785366944e-05,
      "loss": 0.0596,
      "num_input_tokens_seen": 14785584,
      "step": 7795
    },
    {
      "epoch": 1.2724528917529978,
      "grad_norm": 0.9937206506729126,
      "learning_rate": 4.545487189271219e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 14794784,
      "step": 7800
    },
    {
      "epoch": 1.2724528917529978,
      "eval_loss": 0.15411649644374847,
      "eval_runtime": 133.2683,
      "eval_samples_per_second": 20.447,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 14794784,
      "step": 7800
    },
    {
      "epoch": 1.2732686189738152,
      "grad_norm": 0.40805456042289734,
      "learning_rate": 4.544922584565914e-05,
      "loss": 0.1733,
      "num_input_tokens_seen": 14804112,
      "step": 7805
    },
    {
      "epoch": 1.2740843461946325,
      "grad_norm": 0.23557056486606598,
      "learning_rate": 4.544357664507848e-05,
      "loss": 0.0927,
      "num_input_tokens_seen": 14813360,
      "step": 7810
    },
    {
      "epoch": 1.2749000734154499,
      "grad_norm": 0.5620864033699036,
      "learning_rate": 4.54379242918414e-05,
      "loss": 0.1558,
      "num_input_tokens_seen": 14822288,
      "step": 7815
    },
    {
      "epoch": 1.2757158006362672,
      "grad_norm": 0.4835667908191681,
      "learning_rate": 4.543226878681955e-05,
      "loss": 0.183,
      "num_input_tokens_seen": 14832784,
      "step": 7820
    },
    {
      "epoch": 1.2765315278570846,
      "grad_norm": 0.29028910398483276,
      "learning_rate": 4.5426610130885087e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 14841600,
      "step": 7825
    },
    {
      "epoch": 1.277347255077902,
      "grad_norm": 0.2931758165359497,
      "learning_rate": 4.542094832491064e-05,
      "loss": 0.0871,
      "num_input_tokens_seen": 14851984,
      "step": 7830
    },
    {
      "epoch": 1.2781629822987193,
      "grad_norm": 1.2683851718902588,
      "learning_rate": 4.541528336976934e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 14862544,
      "step": 7835
    },
    {
      "epoch": 1.2789787095195366,
      "grad_norm": 1.2336252927780151,
      "learning_rate": 4.540961526633479e-05,
      "loss": 0.2492,
      "num_input_tokens_seen": 14872560,
      "step": 7840
    },
    {
      "epoch": 1.279794436740354,
      "grad_norm": 2.06282901763916,
      "learning_rate": 4.540394401548108e-05,
      "loss": 0.2896,
      "num_input_tokens_seen": 14881872,
      "step": 7845
    },
    {
      "epoch": 1.2806101639611713,
      "grad_norm": 1.139001488685608,
      "learning_rate": 4.539826961808279e-05,
      "loss": 0.1132,
      "num_input_tokens_seen": 14890992,
      "step": 7850
    },
    {
      "epoch": 1.2814258911819887,
      "grad_norm": 0.29101699590682983,
      "learning_rate": 4.5392592075014994e-05,
      "loss": 0.0714,
      "num_input_tokens_seen": 14899888,
      "step": 7855
    },
    {
      "epoch": 1.282241618402806,
      "grad_norm": 0.5185562968254089,
      "learning_rate": 4.538691138715322e-05,
      "loss": 0.0749,
      "num_input_tokens_seen": 14908944,
      "step": 7860
    },
    {
      "epoch": 1.2830573456236234,
      "grad_norm": 0.2862100899219513,
      "learning_rate": 4.5381227555373516e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 14918624,
      "step": 7865
    },
    {
      "epoch": 1.2838730728444407,
      "grad_norm": 0.14514029026031494,
      "learning_rate": 4.537554058055239e-05,
      "loss": 0.0762,
      "num_input_tokens_seen": 14927632,
      "step": 7870
    },
    {
      "epoch": 1.284688800065258,
      "grad_norm": 0.6176927089691162,
      "learning_rate": 4.5369850463566865e-05,
      "loss": 0.1845,
      "num_input_tokens_seen": 14936416,
      "step": 7875
    },
    {
      "epoch": 1.2855045272860757,
      "grad_norm": 0.9871984124183655,
      "learning_rate": 4.5364157205294404e-05,
      "loss": 0.1107,
      "num_input_tokens_seen": 14946496,
      "step": 7880
    },
    {
      "epoch": 1.2863202545068928,
      "grad_norm": 0.21440769731998444,
      "learning_rate": 4.5358460806612996e-05,
      "loss": 0.0711,
      "num_input_tokens_seen": 14956000,
      "step": 7885
    },
    {
      "epoch": 1.2871359817277104,
      "grad_norm": 0.32323578000068665,
      "learning_rate": 4.535276126840109e-05,
      "loss": 0.0675,
      "num_input_tokens_seen": 14966096,
      "step": 7890
    },
    {
      "epoch": 1.2879517089485275,
      "grad_norm": 0.655652642250061,
      "learning_rate": 4.5347058591537626e-05,
      "loss": 0.2719,
      "num_input_tokens_seen": 14975856,
      "step": 7895
    },
    {
      "epoch": 1.288767436169345,
      "grad_norm": 1.4386435747146606,
      "learning_rate": 4.534135277690203e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 14985120,
      "step": 7900
    },
    {
      "epoch": 1.2895831633901622,
      "grad_norm": 0.23518024384975433,
      "learning_rate": 4.533564382537421e-05,
      "loss": 0.1414,
      "num_input_tokens_seen": 14994480,
      "step": 7905
    },
    {
      "epoch": 1.2903988906109798,
      "grad_norm": 1.9502030611038208,
      "learning_rate": 4.532993173783456e-05,
      "loss": 0.3827,
      "num_input_tokens_seen": 15004384,
      "step": 7910
    },
    {
      "epoch": 1.291214617831797,
      "grad_norm": 1.0595723390579224,
      "learning_rate": 4.5324216515163954e-05,
      "loss": 0.1558,
      "num_input_tokens_seen": 15014784,
      "step": 7915
    },
    {
      "epoch": 1.2920303450526145,
      "grad_norm": 0.5841537714004517,
      "learning_rate": 4.531849815824375e-05,
      "loss": 0.0817,
      "num_input_tokens_seen": 15023936,
      "step": 7920
    },
    {
      "epoch": 1.2928460722734318,
      "grad_norm": 0.2769997715950012,
      "learning_rate": 4.5312776667955795e-05,
      "loss": 0.1935,
      "num_input_tokens_seen": 15033696,
      "step": 7925
    },
    {
      "epoch": 1.2936617994942492,
      "grad_norm": 0.0735953152179718,
      "learning_rate": 4.5307052045182405e-05,
      "loss": 0.0731,
      "num_input_tokens_seen": 15043104,
      "step": 7930
    },
    {
      "epoch": 1.2944775267150666,
      "grad_norm": 1.5716427564620972,
      "learning_rate": 4.53013242908064e-05,
      "loss": 0.2732,
      "num_input_tokens_seen": 15052912,
      "step": 7935
    },
    {
      "epoch": 1.295293253935884,
      "grad_norm": 0.10810016095638275,
      "learning_rate": 4.529559340571107e-05,
      "loss": 0.1551,
      "num_input_tokens_seen": 15062032,
      "step": 7940
    },
    {
      "epoch": 1.2961089811567013,
      "grad_norm": 1.9009917974472046,
      "learning_rate": 4.528985939078018e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 15070976,
      "step": 7945
    },
    {
      "epoch": 1.2969247083775186,
      "grad_norm": 1.6652945280075073,
      "learning_rate": 4.5284122246898e-05,
      "loss": 0.2581,
      "num_input_tokens_seen": 15081264,
      "step": 7950
    },
    {
      "epoch": 1.297740435598336,
      "grad_norm": 1.5828083753585815,
      "learning_rate": 4.527838197494926e-05,
      "loss": 0.2076,
      "num_input_tokens_seen": 15090800,
      "step": 7955
    },
    {
      "epoch": 1.2985561628191533,
      "grad_norm": 2.1991236209869385,
      "learning_rate": 4.527263857581918e-05,
      "loss": 0.0882,
      "num_input_tokens_seen": 15100944,
      "step": 7960
    },
    {
      "epoch": 1.2993718900399707,
      "grad_norm": 1.2757130861282349,
      "learning_rate": 4.526689205039347e-05,
      "loss": 0.2756,
      "num_input_tokens_seen": 15111552,
      "step": 7965
    },
    {
      "epoch": 1.300187617260788,
      "grad_norm": 1.063620924949646,
      "learning_rate": 4.5261142399558324e-05,
      "loss": 0.1997,
      "num_input_tokens_seen": 15120816,
      "step": 7970
    },
    {
      "epoch": 1.3010033444816054,
      "grad_norm": 0.19183261692523956,
      "learning_rate": 4.525538962420041e-05,
      "loss": 0.1139,
      "num_input_tokens_seen": 15129600,
      "step": 7975
    },
    {
      "epoch": 1.3018190717024227,
      "grad_norm": 0.5179649591445923,
      "learning_rate": 4.524963372520685e-05,
      "loss": 0.1763,
      "num_input_tokens_seen": 15139408,
      "step": 7980
    },
    {
      "epoch": 1.30263479892324,
      "grad_norm": 1.2430694103240967,
      "learning_rate": 4.524387470346531e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 15148944,
      "step": 7985
    },
    {
      "epoch": 1.3034505261440574,
      "grad_norm": 0.5476223230361938,
      "learning_rate": 4.5238112559863885e-05,
      "loss": 0.1064,
      "num_input_tokens_seen": 15158992,
      "step": 7990
    },
    {
      "epoch": 1.3042662533648748,
      "grad_norm": 1.1742726564407349,
      "learning_rate": 4.5232347295291175e-05,
      "loss": 0.1066,
      "num_input_tokens_seen": 15167408,
      "step": 7995
    },
    {
      "epoch": 1.3050819805856921,
      "grad_norm": 0.5833605527877808,
      "learning_rate": 4.522657891063626e-05,
      "loss": 0.0631,
      "num_input_tokens_seen": 15176240,
      "step": 8000
    },
    {
      "epoch": 1.3050819805856921,
      "eval_loss": 0.1565389782190323,
      "eval_runtime": 133.2689,
      "eval_samples_per_second": 20.447,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 15176240,
      "step": 8000
    },
    {
      "epoch": 1.3058977078065095,
      "grad_norm": 0.30613481998443604,
      "learning_rate": 4.52208074067887e-05,
      "loss": 0.1015,
      "num_input_tokens_seen": 15185600,
      "step": 8005
    },
    {
      "epoch": 1.3067134350273268,
      "grad_norm": 1.0875003337860107,
      "learning_rate": 4.5215032784638516e-05,
      "loss": 0.2404,
      "num_input_tokens_seen": 15193840,
      "step": 8010
    },
    {
      "epoch": 1.3075291622481442,
      "grad_norm": 0.5889846086502075,
      "learning_rate": 4.5209255045076245e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 15203664,
      "step": 8015
    },
    {
      "epoch": 1.3083448894689615,
      "grad_norm": 0.23244711756706238,
      "learning_rate": 4.5203474188992875e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 15213664,
      "step": 8020
    },
    {
      "epoch": 1.309160616689779,
      "grad_norm": 2.036026954650879,
      "learning_rate": 4.51976902172799e-05,
      "loss": 0.2793,
      "num_input_tokens_seen": 15223104,
      "step": 8025
    },
    {
      "epoch": 1.3099763439105963,
      "grad_norm": 0.09426871687173843,
      "learning_rate": 4.519190313082927e-05,
      "loss": 0.1771,
      "num_input_tokens_seen": 15232224,
      "step": 8030
    },
    {
      "epoch": 1.3107920711314136,
      "grad_norm": 0.5773273706436157,
      "learning_rate": 4.518611293053343e-05,
      "loss": 0.1555,
      "num_input_tokens_seen": 15241264,
      "step": 8035
    },
    {
      "epoch": 1.311607798352231,
      "grad_norm": 0.5869739055633545,
      "learning_rate": 4.51803196172853e-05,
      "loss": 0.106,
      "num_input_tokens_seen": 15248736,
      "step": 8040
    },
    {
      "epoch": 1.3124235255730483,
      "grad_norm": 0.649051308631897,
      "learning_rate": 4.517452319197828e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 15257776,
      "step": 8045
    },
    {
      "epoch": 1.3132392527938657,
      "grad_norm": 0.38633695244789124,
      "learning_rate": 4.5168723655506265e-05,
      "loss": 0.0513,
      "num_input_tokens_seen": 15266128,
      "step": 8050
    },
    {
      "epoch": 1.314054980014683,
      "grad_norm": 0.22300750017166138,
      "learning_rate": 4.51629210087636e-05,
      "loss": 0.1172,
      "num_input_tokens_seen": 15274720,
      "step": 8055
    },
    {
      "epoch": 1.3148707072355004,
      "grad_norm": 0.22906287014484406,
      "learning_rate": 4.515711525264513e-05,
      "loss": 0.2294,
      "num_input_tokens_seen": 15283536,
      "step": 8060
    },
    {
      "epoch": 1.315686434456318,
      "grad_norm": 1.092002034187317,
      "learning_rate": 4.5151306388046175e-05,
      "loss": 0.0926,
      "num_input_tokens_seen": 15293824,
      "step": 8065
    },
    {
      "epoch": 1.316502161677135,
      "grad_norm": 0.20033280551433563,
      "learning_rate": 4.514549441586255e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 15302672,
      "step": 8070
    },
    {
      "epoch": 1.3173178888979526,
      "grad_norm": 1.3256404399871826,
      "learning_rate": 4.513967933699051e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 15312432,
      "step": 8075
    },
    {
      "epoch": 1.3181336161187698,
      "grad_norm": 0.402126282453537,
      "learning_rate": 4.513386115232684e-05,
      "loss": 0.0945,
      "num_input_tokens_seen": 15321744,
      "step": 8080
    },
    {
      "epoch": 1.3189493433395874,
      "grad_norm": 0.4388292133808136,
      "learning_rate": 4.5128039862768745e-05,
      "loss": 0.1555,
      "num_input_tokens_seen": 15330624,
      "step": 8085
    },
    {
      "epoch": 1.3197650705604045,
      "grad_norm": 1.060683250427246,
      "learning_rate": 4.512221546921397e-05,
      "loss": 0.1487,
      "num_input_tokens_seen": 15340560,
      "step": 8090
    },
    {
      "epoch": 1.320580797781222,
      "grad_norm": 0.23144178092479706,
      "learning_rate": 4.5116387972560694e-05,
      "loss": 0.0995,
      "num_input_tokens_seen": 15349152,
      "step": 8095
    },
    {
      "epoch": 1.3213965250020392,
      "grad_norm": 0.21089918911457062,
      "learning_rate": 4.511055737370759e-05,
      "loss": 0.0733,
      "num_input_tokens_seen": 15358608,
      "step": 8100
    },
    {
      "epoch": 1.3222122522228568,
      "grad_norm": 0.6240098476409912,
      "learning_rate": 4.510472367355383e-05,
      "loss": 0.1021,
      "num_input_tokens_seen": 15368592,
      "step": 8105
    },
    {
      "epoch": 1.3230279794436741,
      "grad_norm": 0.12820903956890106,
      "learning_rate": 4.509888687299901e-05,
      "loss": 0.1006,
      "num_input_tokens_seen": 15377104,
      "step": 8110
    },
    {
      "epoch": 1.3238437066644915,
      "grad_norm": 2.311434745788574,
      "learning_rate": 4.5093046972943266e-05,
      "loss": 0.3714,
      "num_input_tokens_seen": 15386368,
      "step": 8115
    },
    {
      "epoch": 1.3246594338853088,
      "grad_norm": 0.29251694679260254,
      "learning_rate": 4.508720397428717e-05,
      "loss": 0.2424,
      "num_input_tokens_seen": 15394736,
      "step": 8120
    },
    {
      "epoch": 1.3254751611061262,
      "grad_norm": 0.3033009171485901,
      "learning_rate": 4.508135787793178e-05,
      "loss": 0.1105,
      "num_input_tokens_seen": 15404944,
      "step": 8125
    },
    {
      "epoch": 1.3262908883269435,
      "grad_norm": 0.5313474535942078,
      "learning_rate": 4.5075508684778664e-05,
      "loss": 0.2219,
      "num_input_tokens_seen": 15413760,
      "step": 8130
    },
    {
      "epoch": 1.3271066155477609,
      "grad_norm": 0.5558803081512451,
      "learning_rate": 4.506965639572982e-05,
      "loss": 0.0527,
      "num_input_tokens_seen": 15424544,
      "step": 8135
    },
    {
      "epoch": 1.3279223427685782,
      "grad_norm": 0.8893669843673706,
      "learning_rate": 4.506380101168774e-05,
      "loss": 0.0873,
      "num_input_tokens_seen": 15434368,
      "step": 8140
    },
    {
      "epoch": 1.3287380699893956,
      "grad_norm": 0.19663558900356293,
      "learning_rate": 4.505794253355542e-05,
      "loss": 0.1072,
      "num_input_tokens_seen": 15443296,
      "step": 8145
    },
    {
      "epoch": 1.329553797210213,
      "grad_norm": 0.4706554412841797,
      "learning_rate": 4.5052080962236286e-05,
      "loss": 0.1255,
      "num_input_tokens_seen": 15452656,
      "step": 8150
    },
    {
      "epoch": 1.3303695244310303,
      "grad_norm": 0.38031381368637085,
      "learning_rate": 4.504621629863428e-05,
      "loss": 0.2843,
      "num_input_tokens_seen": 15462848,
      "step": 8155
    },
    {
      "epoch": 1.3311852516518476,
      "grad_norm": 1.0593258142471313,
      "learning_rate": 4.504034854365381e-05,
      "loss": 0.2527,
      "num_input_tokens_seen": 15472288,
      "step": 8160
    },
    {
      "epoch": 1.332000978872665,
      "grad_norm": 0.47114235162734985,
      "learning_rate": 4.503447769819974e-05,
      "loss": 0.0853,
      "num_input_tokens_seen": 15481024,
      "step": 8165
    },
    {
      "epoch": 1.3328167060934824,
      "grad_norm": 0.11015418916940689,
      "learning_rate": 4.502860376317745e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 15490624,
      "step": 8170
    },
    {
      "epoch": 1.3336324333142997,
      "grad_norm": 0.36946752667427063,
      "learning_rate": 4.502272673949276e-05,
      "loss": 0.1786,
      "num_input_tokens_seen": 15499680,
      "step": 8175
    },
    {
      "epoch": 1.334448160535117,
      "grad_norm": 0.34774795174598694,
      "learning_rate": 4.501684662805199e-05,
      "loss": 0.1005,
      "num_input_tokens_seen": 15509280,
      "step": 8180
    },
    {
      "epoch": 1.3352638877559344,
      "grad_norm": 0.32988226413726807,
      "learning_rate": 4.5010963429761924e-05,
      "loss": 0.1995,
      "num_input_tokens_seen": 15519376,
      "step": 8185
    },
    {
      "epoch": 1.3360796149767518,
      "grad_norm": 0.11540359258651733,
      "learning_rate": 4.500507714552982e-05,
      "loss": 0.1043,
      "num_input_tokens_seen": 15529120,
      "step": 8190
    },
    {
      "epoch": 1.3368953421975691,
      "grad_norm": 0.2815401554107666,
      "learning_rate": 4.499918777626342e-05,
      "loss": 0.1372,
      "num_input_tokens_seen": 15538352,
      "step": 8195
    },
    {
      "epoch": 1.3377110694183865,
      "grad_norm": 1.3319008350372314,
      "learning_rate": 4.499329532287093e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 15548080,
      "step": 8200
    },
    {
      "epoch": 1.3377110694183865,
      "eval_loss": 0.15321700274944305,
      "eval_runtime": 133.3478,
      "eval_samples_per_second": 20.435,
      "eval_steps_per_second": 10.221,
      "num_input_tokens_seen": 15548080,
      "step": 8200
    },
    {
      "epoch": 1.3385267966392038,
      "grad_norm": 0.6703736186027527,
      "learning_rate": 4.4987399786261064e-05,
      "loss": 0.0721,
      "num_input_tokens_seen": 15556992,
      "step": 8205
    },
    {
      "epoch": 1.3393425238600212,
      "grad_norm": 0.6942150592803955,
      "learning_rate": 4.498150116734297e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 15565600,
      "step": 8210
    },
    {
      "epoch": 1.3401582510808385,
      "grad_norm": 1.4631931781768799,
      "learning_rate": 4.4975599467026294e-05,
      "loss": 0.2589,
      "num_input_tokens_seen": 15574960,
      "step": 8215
    },
    {
      "epoch": 1.3409739783016559,
      "grad_norm": 1.180650234222412,
      "learning_rate": 4.496969468622114e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 15586208,
      "step": 8220
    },
    {
      "epoch": 1.3417897055224732,
      "grad_norm": 1.1809096336364746,
      "learning_rate": 4.496378682583813e-05,
      "loss": 0.2157,
      "num_input_tokens_seen": 15594960,
      "step": 8225
    },
    {
      "epoch": 1.3426054327432906,
      "grad_norm": 0.45788365602493286,
      "learning_rate": 4.495787588678829e-05,
      "loss": 0.1091,
      "num_input_tokens_seen": 15604144,
      "step": 8230
    },
    {
      "epoch": 1.343421159964108,
      "grad_norm": 0.1266743391752243,
      "learning_rate": 4.4951961869983196e-05,
      "loss": 0.063,
      "num_input_tokens_seen": 15614192,
      "step": 8235
    },
    {
      "epoch": 1.3442368871849253,
      "grad_norm": 1.276055097579956,
      "learning_rate": 4.494604477633485e-05,
      "loss": 0.0918,
      "num_input_tokens_seen": 15624176,
      "step": 8240
    },
    {
      "epoch": 1.3450526144057426,
      "grad_norm": 1.039087176322937,
      "learning_rate": 4.4940124606755734e-05,
      "loss": 0.1217,
      "num_input_tokens_seen": 15633440,
      "step": 8245
    },
    {
      "epoch": 1.3458683416265602,
      "grad_norm": 0.7098714113235474,
      "learning_rate": 4.493420136215882e-05,
      "loss": 0.0973,
      "num_input_tokens_seen": 15641968,
      "step": 8250
    },
    {
      "epoch": 1.3466840688473773,
      "grad_norm": 0.2276516705751419,
      "learning_rate": 4.492827504345756e-05,
      "loss": 0.0402,
      "num_input_tokens_seen": 15651184,
      "step": 8255
    },
    {
      "epoch": 1.347499796068195,
      "grad_norm": 0.8156149983406067,
      "learning_rate": 4.492234565156584e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 15660816,
      "step": 8260
    },
    {
      "epoch": 1.348315523289012,
      "grad_norm": 0.7042011022567749,
      "learning_rate": 4.491641318739807e-05,
      "loss": 0.0644,
      "num_input_tokens_seen": 15669152,
      "step": 8265
    },
    {
      "epoch": 1.3491312505098296,
      "grad_norm": 0.6561802625656128,
      "learning_rate": 4.4910477651869096e-05,
      "loss": 0.1172,
      "num_input_tokens_seen": 15679664,
      "step": 8270
    },
    {
      "epoch": 1.3499469777306468,
      "grad_norm": 0.14083723723888397,
      "learning_rate": 4.4904539045894254e-05,
      "loss": 0.2861,
      "num_input_tokens_seen": 15690352,
      "step": 8275
    },
    {
      "epoch": 1.3507627049514643,
      "grad_norm": 0.282221257686615,
      "learning_rate": 4.4898597370389364e-05,
      "loss": 0.1133,
      "num_input_tokens_seen": 15699984,
      "step": 8280
    },
    {
      "epoch": 1.3515784321722815,
      "grad_norm": 0.9420568346977234,
      "learning_rate": 4.489265262627069e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 15708960,
      "step": 8285
    },
    {
      "epoch": 1.352394159393099,
      "grad_norm": 0.22592097520828247,
      "learning_rate": 4.488670481445499e-05,
      "loss": 0.2475,
      "num_input_tokens_seen": 15717984,
      "step": 8290
    },
    {
      "epoch": 1.3532098866139162,
      "grad_norm": 1.0514631271362305,
      "learning_rate": 4.488075393585951e-05,
      "loss": 0.1352,
      "num_input_tokens_seen": 15727264,
      "step": 8295
    },
    {
      "epoch": 1.3540256138347337,
      "grad_norm": 0.05396819859743118,
      "learning_rate": 4.487479999140193e-05,
      "loss": 0.0637,
      "num_input_tokens_seen": 15736288,
      "step": 8300
    },
    {
      "epoch": 1.354841341055551,
      "grad_norm": 0.07423008978366852,
      "learning_rate": 4.4868842982000425e-05,
      "loss": 0.0864,
      "num_input_tokens_seen": 15745520,
      "step": 8305
    },
    {
      "epoch": 1.3556570682763684,
      "grad_norm": 0.49550268054008484,
      "learning_rate": 4.486288290857365e-05,
      "loss": 0.1733,
      "num_input_tokens_seen": 15754576,
      "step": 8310
    },
    {
      "epoch": 1.3564727954971858,
      "grad_norm": 0.45106297731399536,
      "learning_rate": 4.4856919772040715e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 15764656,
      "step": 8315
    },
    {
      "epoch": 1.3572885227180032,
      "grad_norm": 1.199530005455017,
      "learning_rate": 4.485095357332122e-05,
      "loss": 0.0982,
      "num_input_tokens_seen": 15774016,
      "step": 8320
    },
    {
      "epoch": 1.3581042499388205,
      "grad_norm": 0.6984348297119141,
      "learning_rate": 4.484498431333521e-05,
      "loss": 0.0565,
      "num_input_tokens_seen": 15783792,
      "step": 8325
    },
    {
      "epoch": 1.3589199771596379,
      "grad_norm": 0.21834765374660492,
      "learning_rate": 4.4839011993003245e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 15794160,
      "step": 8330
    },
    {
      "epoch": 1.3597357043804552,
      "grad_norm": 1.0693281888961792,
      "learning_rate": 4.4833036613246305e-05,
      "loss": 0.1954,
      "num_input_tokens_seen": 15804176,
      "step": 8335
    },
    {
      "epoch": 1.3605514316012726,
      "grad_norm": 0.15678833425045013,
      "learning_rate": 4.482705817498589e-05,
      "loss": 0.2058,
      "num_input_tokens_seen": 15814640,
      "step": 8340
    },
    {
      "epoch": 1.36136715882209,
      "grad_norm": 1.6732529401779175,
      "learning_rate": 4.4821076679143934e-05,
      "loss": 0.3217,
      "num_input_tokens_seen": 15824528,
      "step": 8345
    },
    {
      "epoch": 1.3621828860429073,
      "grad_norm": 1.0937225818634033,
      "learning_rate": 4.481509212664288e-05,
      "loss": 0.0535,
      "num_input_tokens_seen": 15833760,
      "step": 8350
    },
    {
      "epoch": 1.3629986132637246,
      "grad_norm": 1.2970434427261353,
      "learning_rate": 4.480910451840559e-05,
      "loss": 0.08,
      "num_input_tokens_seen": 15841744,
      "step": 8355
    },
    {
      "epoch": 1.363814340484542,
      "grad_norm": 0.14115513861179352,
      "learning_rate": 4.480311385535546e-05,
      "loss": 0.1987,
      "num_input_tokens_seen": 15851728,
      "step": 8360
    },
    {
      "epoch": 1.3646300677053593,
      "grad_norm": 0.2181341052055359,
      "learning_rate": 4.47971201384163e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 15860896,
      "step": 8365
    },
    {
      "epoch": 1.3654457949261767,
      "grad_norm": 0.6722909212112427,
      "learning_rate": 4.4791123368512446e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 15870608,
      "step": 8370
    },
    {
      "epoch": 1.366261522146994,
      "grad_norm": 0.72603440284729,
      "learning_rate": 4.478512354656864e-05,
      "loss": 0.1154,
      "num_input_tokens_seen": 15879472,
      "step": 8375
    },
    {
      "epoch": 1.3670772493678114,
      "grad_norm": 0.4626818001270294,
      "learning_rate": 4.477912067351016e-05,
      "loss": 0.2038,
      "num_input_tokens_seen": 15888768,
      "step": 8380
    },
    {
      "epoch": 1.3678929765886287,
      "grad_norm": 0.2091868668794632,
      "learning_rate": 4.477311475026271e-05,
      "loss": 0.2967,
      "num_input_tokens_seen": 15897616,
      "step": 8385
    },
    {
      "epoch": 1.368708703809446,
      "grad_norm": 0.05115637183189392,
      "learning_rate": 4.476710577775248e-05,
      "loss": 0.1421,
      "num_input_tokens_seen": 15906960,
      "step": 8390
    },
    {
      "epoch": 1.3695244310302634,
      "grad_norm": 0.43288668990135193,
      "learning_rate": 4.476109375690612e-05,
      "loss": 0.1839,
      "num_input_tokens_seen": 15916224,
      "step": 8395
    },
    {
      "epoch": 1.3703401582510808,
      "grad_norm": 0.059281088411808014,
      "learning_rate": 4.4755078688650784e-05,
      "loss": 0.0363,
      "num_input_tokens_seen": 15926832,
      "step": 8400
    },
    {
      "epoch": 1.3703401582510808,
      "eval_loss": 0.15287457406520844,
      "eval_runtime": 133.1354,
      "eval_samples_per_second": 20.468,
      "eval_steps_per_second": 10.238,
      "num_input_tokens_seen": 15926832,
      "step": 8400
    },
    {
      "epoch": 1.3711558854718982,
      "grad_norm": 0.28411799669265747,
      "learning_rate": 4.474906057391406e-05,
      "loss": 0.0936,
      "num_input_tokens_seen": 15936512,
      "step": 8405
    },
    {
      "epoch": 1.3719716126927155,
      "grad_norm": 1.381635308265686,
      "learning_rate": 4.4743039413624e-05,
      "loss": 0.5428,
      "num_input_tokens_seen": 15945104,
      "step": 8410
    },
    {
      "epoch": 1.3727873399135329,
      "grad_norm": 0.6589316129684448,
      "learning_rate": 4.473701520870916e-05,
      "loss": 0.2051,
      "num_input_tokens_seen": 15954464,
      "step": 8415
    },
    {
      "epoch": 1.3736030671343502,
      "grad_norm": 0.5425523519515991,
      "learning_rate": 4.4730987960098544e-05,
      "loss": 0.1097,
      "num_input_tokens_seen": 15964096,
      "step": 8420
    },
    {
      "epoch": 1.3744187943551676,
      "grad_norm": 0.2454613447189331,
      "learning_rate": 4.4724957668721635e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 15974352,
      "step": 8425
    },
    {
      "epoch": 1.375234521575985,
      "grad_norm": 0.6615517735481262,
      "learning_rate": 4.471892433550836e-05,
      "loss": 0.0835,
      "num_input_tokens_seen": 15983712,
      "step": 8430
    },
    {
      "epoch": 1.3760502487968023,
      "grad_norm": 1.6326159238815308,
      "learning_rate": 4.471288796138916e-05,
      "loss": 0.2014,
      "num_input_tokens_seen": 15992464,
      "step": 8435
    },
    {
      "epoch": 1.3768659760176196,
      "grad_norm": 0.08496405929327011,
      "learning_rate": 4.470684854729491e-05,
      "loss": 0.1119,
      "num_input_tokens_seen": 16002544,
      "step": 8440
    },
    {
      "epoch": 1.3776817032384372,
      "grad_norm": 0.5950372815132141,
      "learning_rate": 4.4700806094156955e-05,
      "loss": 0.201,
      "num_input_tokens_seen": 16011440,
      "step": 8445
    },
    {
      "epoch": 1.3784974304592543,
      "grad_norm": 2.716418504714966,
      "learning_rate": 4.469476060290713e-05,
      "loss": 0.2759,
      "num_input_tokens_seen": 16021600,
      "step": 8450
    },
    {
      "epoch": 1.379313157680072,
      "grad_norm": 1.6137597560882568,
      "learning_rate": 4.468871207447772e-05,
      "loss": 0.0965,
      "num_input_tokens_seen": 16030704,
      "step": 8455
    },
    {
      "epoch": 1.380128884900889,
      "grad_norm": 0.4342905879020691,
      "learning_rate": 4.4682660509801486e-05,
      "loss": 0.0533,
      "num_input_tokens_seen": 16040000,
      "step": 8460
    },
    {
      "epoch": 1.3809446121217066,
      "grad_norm": 0.17039749026298523,
      "learning_rate": 4.467660590981165e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 16050816,
      "step": 8465
    },
    {
      "epoch": 1.3817603393425237,
      "grad_norm": 0.08786887675523758,
      "learning_rate": 4.467054827544191e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 16059920,
      "step": 8470
    },
    {
      "epoch": 1.3825760665633413,
      "grad_norm": 0.41753390431404114,
      "learning_rate": 4.4664487607626434e-05,
      "loss": 0.1242,
      "num_input_tokens_seen": 16067968,
      "step": 8475
    },
    {
      "epoch": 1.3833917937841584,
      "grad_norm": 0.24492022395133972,
      "learning_rate": 4.4658423907299845e-05,
      "loss": 0.1812,
      "num_input_tokens_seen": 16078032,
      "step": 8480
    },
    {
      "epoch": 1.384207521004976,
      "grad_norm": 0.1550026386976242,
      "learning_rate": 4.465235717539725e-05,
      "loss": 0.1728,
      "num_input_tokens_seen": 16088464,
      "step": 8485
    },
    {
      "epoch": 1.3850232482257934,
      "grad_norm": 0.11043647676706314,
      "learning_rate": 4.464628741285421e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 16097792,
      "step": 8490
    },
    {
      "epoch": 1.3858389754466107,
      "grad_norm": 0.2405678629875183,
      "learning_rate": 4.4640214620606754e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 16108400,
      "step": 8495
    },
    {
      "epoch": 1.386654702667428,
      "grad_norm": 0.4661824107170105,
      "learning_rate": 4.46341387995914e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 16117504,
      "step": 8500
    },
    {
      "epoch": 1.3874704298882454,
      "grad_norm": 1.3757628202438354,
      "learning_rate": 4.4628059950745106e-05,
      "loss": 0.2127,
      "num_input_tokens_seen": 16126752,
      "step": 8505
    },
    {
      "epoch": 1.3882861571090628,
      "grad_norm": 0.16000351309776306,
      "learning_rate": 4.4621978075005297e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 16135648,
      "step": 8510
    },
    {
      "epoch": 1.3891018843298801,
      "grad_norm": 0.45167431235313416,
      "learning_rate": 4.461589317330989e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 16146272,
      "step": 8515
    },
    {
      "epoch": 1.3899176115506975,
      "grad_norm": 0.5646092295646667,
      "learning_rate": 4.460980524659724e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 16155248,
      "step": 8520
    },
    {
      "epoch": 1.3907333387715148,
      "grad_norm": 0.7430686354637146,
      "learning_rate": 4.46037142958062e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 16164144,
      "step": 8525
    },
    {
      "epoch": 1.3915490659923322,
      "grad_norm": 1.0889993906021118,
      "learning_rate": 4.4597620321876046e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 16174304,
      "step": 8530
    },
    {
      "epoch": 1.3923647932131495,
      "grad_norm": 0.6130412817001343,
      "learning_rate": 4.459152332574656e-05,
      "loss": 0.0479,
      "num_input_tokens_seen": 16182032,
      "step": 8535
    },
    {
      "epoch": 1.393180520433967,
      "grad_norm": 0.4934350550174713,
      "learning_rate": 4.4585423308357985e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 16190944,
      "step": 8540
    },
    {
      "epoch": 1.3939962476547842,
      "grad_norm": 0.4237852990627289,
      "learning_rate": 4.457932027065102e-05,
      "loss": 0.2029,
      "num_input_tokens_seen": 16200816,
      "step": 8545
    },
    {
      "epoch": 1.3948119748756016,
      "grad_norm": 0.6494830846786499,
      "learning_rate": 4.45732142135668e-05,
      "loss": 0.0775,
      "num_input_tokens_seen": 16210128,
      "step": 8550
    },
    {
      "epoch": 1.395627702096419,
      "grad_norm": 0.3710777163505554,
      "learning_rate": 4.4567105138046986e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 16220416,
      "step": 8555
    },
    {
      "epoch": 1.3964434293172363,
      "grad_norm": 0.7170553207397461,
      "learning_rate": 4.456099304503365e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 16229984,
      "step": 8560
    },
    {
      "epoch": 1.3972591565380537,
      "grad_norm": 1.2551449537277222,
      "learning_rate": 4.455487793546939e-05,
      "loss": 0.2325,
      "num_input_tokens_seen": 16240144,
      "step": 8565
    },
    {
      "epoch": 1.398074883758871,
      "grad_norm": 0.07494810223579407,
      "learning_rate": 4.454875981029719e-05,
      "loss": 0.2049,
      "num_input_tokens_seen": 16249632,
      "step": 8570
    },
    {
      "epoch": 1.3988906109796884,
      "grad_norm": 0.7916592955589294,
      "learning_rate": 4.454263867046057e-05,
      "loss": 0.0404,
      "num_input_tokens_seen": 16258496,
      "step": 8575
    },
    {
      "epoch": 1.3997063382005057,
      "grad_norm": 0.5290186405181885,
      "learning_rate": 4.4536514516903484e-05,
      "loss": 0.1816,
      "num_input_tokens_seen": 16268000,
      "step": 8580
    },
    {
      "epoch": 1.400522065421323,
      "grad_norm": 0.3936726450920105,
      "learning_rate": 4.453038735057034e-05,
      "loss": 0.1879,
      "num_input_tokens_seen": 16276800,
      "step": 8585
    },
    {
      "epoch": 1.4013377926421404,
      "grad_norm": 0.4640662372112274,
      "learning_rate": 4.4524257172406034e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 16286544,
      "step": 8590
    },
    {
      "epoch": 1.4021535198629578,
      "grad_norm": 0.9134763479232788,
      "learning_rate": 4.451812398335592e-05,
      "loss": 0.3648,
      "num_input_tokens_seen": 16295856,
      "step": 8595
    },
    {
      "epoch": 1.4029692470837751,
      "grad_norm": 0.7919626235961914,
      "learning_rate": 4.4511987784365805e-05,
      "loss": 0.1787,
      "num_input_tokens_seen": 16305344,
      "step": 8600
    },
    {
      "epoch": 1.4029692470837751,
      "eval_loss": 0.15355642139911652,
      "eval_runtime": 133.2956,
      "eval_samples_per_second": 20.443,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 16305344,
      "step": 8600
    },
    {
      "epoch": 1.4037849743045925,
      "grad_norm": 0.22011730074882507,
      "learning_rate": 4.450584857638197e-05,
      "loss": 0.3619,
      "num_input_tokens_seen": 16313648,
      "step": 8605
    },
    {
      "epoch": 1.4046007015254098,
      "grad_norm": 1.1269876956939697,
      "learning_rate": 4.449970636035116e-05,
      "loss": 0.1255,
      "num_input_tokens_seen": 16323568,
      "step": 8610
    },
    {
      "epoch": 1.4054164287462272,
      "grad_norm": 0.4497680962085724,
      "learning_rate": 4.4493561137220574e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 16333024,
      "step": 8615
    },
    {
      "epoch": 1.4062321559670445,
      "grad_norm": 0.8662775158882141,
      "learning_rate": 4.44874129079379e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 16342256,
      "step": 8620
    },
    {
      "epoch": 1.407047883187862,
      "grad_norm": 0.2329082489013672,
      "learning_rate": 4.4481261673451255e-05,
      "loss": 0.1009,
      "num_input_tokens_seen": 16351216,
      "step": 8625
    },
    {
      "epoch": 1.4078636104086795,
      "grad_norm": 0.3819938600063324,
      "learning_rate": 4.4475107434709245e-05,
      "loss": 0.0793,
      "num_input_tokens_seen": 16360768,
      "step": 8630
    },
    {
      "epoch": 1.4086793376294966,
      "grad_norm": 0.4788028597831726,
      "learning_rate": 4.446895019266093e-05,
      "loss": 0.0649,
      "num_input_tokens_seen": 16369552,
      "step": 8635
    },
    {
      "epoch": 1.4094950648503142,
      "grad_norm": 0.34660348296165466,
      "learning_rate": 4.446278994825583e-05,
      "loss": 0.1682,
      "num_input_tokens_seen": 16379168,
      "step": 8640
    },
    {
      "epoch": 1.4103107920711313,
      "grad_norm": 0.041065722703933716,
      "learning_rate": 4.445662670244394e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 16388592,
      "step": 8645
    },
    {
      "epoch": 1.4111265192919489,
      "grad_norm": 0.9065450429916382,
      "learning_rate": 4.44504604561757e-05,
      "loss": 0.1076,
      "num_input_tokens_seen": 16397696,
      "step": 8650
    },
    {
      "epoch": 1.411942246512766,
      "grad_norm": 0.20508907735347748,
      "learning_rate": 4.4444291210402035e-05,
      "loss": 0.2809,
      "num_input_tokens_seen": 16406544,
      "step": 8655
    },
    {
      "epoch": 1.4127579737335836,
      "grad_norm": 0.2782888412475586,
      "learning_rate": 4.443811896607431e-05,
      "loss": 0.1182,
      "num_input_tokens_seen": 16415968,
      "step": 8660
    },
    {
      "epoch": 1.4135737009544007,
      "grad_norm": 0.8788525462150574,
      "learning_rate": 4.443194372414436e-05,
      "loss": 0.0953,
      "num_input_tokens_seen": 16426416,
      "step": 8665
    },
    {
      "epoch": 1.4143894281752183,
      "grad_norm": 0.3672316074371338,
      "learning_rate": 4.442576548556449e-05,
      "loss": 0.2684,
      "num_input_tokens_seen": 16435664,
      "step": 8670
    },
    {
      "epoch": 1.4152051553960356,
      "grad_norm": 0.3707022964954376,
      "learning_rate": 4.441958425128747e-05,
      "loss": 0.0872,
      "num_input_tokens_seen": 16445328,
      "step": 8675
    },
    {
      "epoch": 1.416020882616853,
      "grad_norm": 0.29182925820350647,
      "learning_rate": 4.4413400022266515e-05,
      "loss": 0.0459,
      "num_input_tokens_seen": 16456016,
      "step": 8680
    },
    {
      "epoch": 1.4168366098376703,
      "grad_norm": 0.21974653005599976,
      "learning_rate": 4.4407212799455313e-05,
      "loss": 0.104,
      "num_input_tokens_seen": 16466384,
      "step": 8685
    },
    {
      "epoch": 1.4176523370584877,
      "grad_norm": 0.08936724811792374,
      "learning_rate": 4.4401022583808003e-05,
      "loss": 0.0844,
      "num_input_tokens_seen": 16475264,
      "step": 8690
    },
    {
      "epoch": 1.418468064279305,
      "grad_norm": 0.15556594729423523,
      "learning_rate": 4.439482937627921e-05,
      "loss": 0.1795,
      "num_input_tokens_seen": 16485824,
      "step": 8695
    },
    {
      "epoch": 1.4192837915001224,
      "grad_norm": 0.858793318271637,
      "learning_rate": 4.4388633177824004e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 16494000,
      "step": 8700
    },
    {
      "epoch": 1.4200995187209398,
      "grad_norm": 1.4597536325454712,
      "learning_rate": 4.4382433989397895e-05,
      "loss": 0.1959,
      "num_input_tokens_seen": 16502032,
      "step": 8705
    },
    {
      "epoch": 1.420915245941757,
      "grad_norm": 1.3929907083511353,
      "learning_rate": 4.4376231811956895e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 16511872,
      "step": 8710
    },
    {
      "epoch": 1.4217309731625745,
      "grad_norm": 0.7388080358505249,
      "learning_rate": 4.437002664645745e-05,
      "loss": 0.1906,
      "num_input_tokens_seen": 16522064,
      "step": 8715
    },
    {
      "epoch": 1.4225467003833918,
      "grad_norm": 1.180070161819458,
      "learning_rate": 4.436381849385649e-05,
      "loss": 0.176,
      "num_input_tokens_seen": 16531712,
      "step": 8720
    },
    {
      "epoch": 1.4233624276042092,
      "grad_norm": 1.4696578979492188,
      "learning_rate": 4.435760735511136e-05,
      "loss": 0.2145,
      "num_input_tokens_seen": 16541584,
      "step": 8725
    },
    {
      "epoch": 1.4241781548250265,
      "grad_norm": 0.11643575876951218,
      "learning_rate": 4.435139323117992e-05,
      "loss": 0.2148,
      "num_input_tokens_seen": 16550144,
      "step": 8730
    },
    {
      "epoch": 1.4249938820458439,
      "grad_norm": 1.190036654472351,
      "learning_rate": 4.434517612302046e-05,
      "loss": 0.2036,
      "num_input_tokens_seen": 16561104,
      "step": 8735
    },
    {
      "epoch": 1.4258096092666612,
      "grad_norm": 1.2208508253097534,
      "learning_rate": 4.433895603159174e-05,
      "loss": 0.3073,
      "num_input_tokens_seen": 16572304,
      "step": 8740
    },
    {
      "epoch": 1.4266253364874786,
      "grad_norm": 0.26134830713272095,
      "learning_rate": 4.433273295785296e-05,
      "loss": 0.1699,
      "num_input_tokens_seen": 16582416,
      "step": 8745
    },
    {
      "epoch": 1.427441063708296,
      "grad_norm": 0.19370605051517487,
      "learning_rate": 4.432650690276382e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 16591280,
      "step": 8750
    },
    {
      "epoch": 1.4282567909291133,
      "grad_norm": 0.249551922082901,
      "learning_rate": 4.4320277867284435e-05,
      "loss": 0.1036,
      "num_input_tokens_seen": 16602320,
      "step": 8755
    },
    {
      "epoch": 1.4290725181499306,
      "grad_norm": 0.17512217164039612,
      "learning_rate": 4.431404585237541e-05,
      "loss": 0.0489,
      "num_input_tokens_seen": 16611296,
      "step": 8760
    },
    {
      "epoch": 1.429888245370748,
      "grad_norm": 0.8647394180297852,
      "learning_rate": 4.43078108589978e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 16620544,
      "step": 8765
    },
    {
      "epoch": 1.4307039725915653,
      "grad_norm": 0.46873220801353455,
      "learning_rate": 4.4301572888113116e-05,
      "loss": 0.2148,
      "num_input_tokens_seen": 16630816,
      "step": 8770
    },
    {
      "epoch": 1.4315196998123827,
      "grad_norm": 1.6438896656036377,
      "learning_rate": 4.4295331940683337e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 16639680,
      "step": 8775
    },
    {
      "epoch": 1.4323354270332,
      "grad_norm": 0.40123069286346436,
      "learning_rate": 4.428908801767089e-05,
      "loss": 0.0679,
      "num_input_tokens_seen": 16649504,
      "step": 8780
    },
    {
      "epoch": 1.4331511542540174,
      "grad_norm": 0.05562113970518112,
      "learning_rate": 4.428284112003868e-05,
      "loss": 0.0923,
      "num_input_tokens_seen": 16658448,
      "step": 8785
    },
    {
      "epoch": 1.4339668814748348,
      "grad_norm": 0.09179507195949554,
      "learning_rate": 4.4276591248750033e-05,
      "loss": 0.0385,
      "num_input_tokens_seen": 16667712,
      "step": 8790
    },
    {
      "epoch": 1.434782608695652,
      "grad_norm": 0.6291736364364624,
      "learning_rate": 4.4270338404768774e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 16677600,
      "step": 8795
    },
    {
      "epoch": 1.4355983359164695,
      "grad_norm": 0.09109805524349213,
      "learning_rate": 4.426408258905917e-05,
      "loss": 0.0654,
      "num_input_tokens_seen": 16686528,
      "step": 8800
    },
    {
      "epoch": 1.4355983359164695,
      "eval_loss": 0.15368340909481049,
      "eval_runtime": 133.3737,
      "eval_samples_per_second": 20.431,
      "eval_steps_per_second": 10.219,
      "num_input_tokens_seen": 16686528,
      "step": 8800
    },
    {
      "epoch": 1.4364140631372868,
      "grad_norm": 0.8646674156188965,
      "learning_rate": 4.425782380258594e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 16696240,
      "step": 8805
    },
    {
      "epoch": 1.4372297903581042,
      "grad_norm": 0.6895981431007385,
      "learning_rate": 4.425156204631427e-05,
      "loss": 0.1211,
      "num_input_tokens_seen": 16706432,
      "step": 8810
    },
    {
      "epoch": 1.4380455175789217,
      "grad_norm": 0.15127263963222504,
      "learning_rate": 4.424529732120981e-05,
      "loss": 0.1123,
      "num_input_tokens_seen": 16716272,
      "step": 8815
    },
    {
      "epoch": 1.4388612447997389,
      "grad_norm": 0.39339005947113037,
      "learning_rate": 4.423902962823864e-05,
      "loss": 0.0366,
      "num_input_tokens_seen": 16726064,
      "step": 8820
    },
    {
      "epoch": 1.4396769720205564,
      "grad_norm": 0.9801146984100342,
      "learning_rate": 4.423275896836733e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 16734688,
      "step": 8825
    },
    {
      "epoch": 1.4404926992413736,
      "grad_norm": 1.5569517612457275,
      "learning_rate": 4.42264853425629e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 16745008,
      "step": 8830
    },
    {
      "epoch": 1.4413084264621912,
      "grad_norm": 1.5929551124572754,
      "learning_rate": 4.4220208751792816e-05,
      "loss": 0.3334,
      "num_input_tokens_seen": 16755344,
      "step": 8835
    },
    {
      "epoch": 1.4421241536830083,
      "grad_norm": 0.2340666502714157,
      "learning_rate": 4.421392919702499e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 16764528,
      "step": 8840
    },
    {
      "epoch": 1.4429398809038259,
      "grad_norm": 1.395289421081543,
      "learning_rate": 4.4207646679227846e-05,
      "loss": 0.1893,
      "num_input_tokens_seen": 16772752,
      "step": 8845
    },
    {
      "epoch": 1.443755608124643,
      "grad_norm": 0.7657433748245239,
      "learning_rate": 4.42013611993702e-05,
      "loss": 0.1209,
      "num_input_tokens_seen": 16781984,
      "step": 8850
    },
    {
      "epoch": 1.4445713353454606,
      "grad_norm": 0.6852088570594788,
      "learning_rate": 4.419507275842135e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 16791776,
      "step": 8855
    },
    {
      "epoch": 1.445387062566278,
      "grad_norm": 0.16457779705524445,
      "learning_rate": 4.418878135735106e-05,
      "loss": 0.2042,
      "num_input_tokens_seen": 16798544,
      "step": 8860
    },
    {
      "epoch": 1.4462027897870953,
      "grad_norm": 0.9328606724739075,
      "learning_rate": 4.418248699712955e-05,
      "loss": 0.2105,
      "num_input_tokens_seen": 16808736,
      "step": 8865
    },
    {
      "epoch": 1.4470185170079126,
      "grad_norm": 0.048331938683986664,
      "learning_rate": 4.417618967872748e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 16819232,
      "step": 8870
    },
    {
      "epoch": 1.44783424422873,
      "grad_norm": 0.6048367619514465,
      "learning_rate": 4.4169889403115985e-05,
      "loss": 0.2066,
      "num_input_tokens_seen": 16828960,
      "step": 8875
    },
    {
      "epoch": 1.4486499714495473,
      "grad_norm": 0.4283715784549713,
      "learning_rate": 4.4163586171266627e-05,
      "loss": 0.0786,
      "num_input_tokens_seen": 16838272,
      "step": 8880
    },
    {
      "epoch": 1.4494656986703647,
      "grad_norm": 1.4429333209991455,
      "learning_rate": 4.415727998415147e-05,
      "loss": 0.4363,
      "num_input_tokens_seen": 16848464,
      "step": 8885
    },
    {
      "epoch": 1.450281425891182,
      "grad_norm": 0.1896127611398697,
      "learning_rate": 4.4150970842742985e-05,
      "loss": 0.0507,
      "num_input_tokens_seen": 16857168,
      "step": 8890
    },
    {
      "epoch": 1.4510971531119994,
      "grad_norm": 0.4837021231651306,
      "learning_rate": 4.4144658748014134e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 16867024,
      "step": 8895
    },
    {
      "epoch": 1.4519128803328167,
      "grad_norm": 1.4922345876693726,
      "learning_rate": 4.413834370093831e-05,
      "loss": 0.2057,
      "num_input_tokens_seen": 16876544,
      "step": 8900
    },
    {
      "epoch": 1.452728607553634,
      "grad_norm": 2.5404679775238037,
      "learning_rate": 4.413202570248939e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 16886608,
      "step": 8905
    },
    {
      "epoch": 1.4535443347744514,
      "grad_norm": 1.774273157119751,
      "learning_rate": 4.412570475364167e-05,
      "loss": 0.1829,
      "num_input_tokens_seen": 16897120,
      "step": 8910
    },
    {
      "epoch": 1.4543600619952688,
      "grad_norm": 0.5965388417243958,
      "learning_rate": 4.411938085536994e-05,
      "loss": 0.1246,
      "num_input_tokens_seen": 16906384,
      "step": 8915
    },
    {
      "epoch": 1.4551757892160861,
      "grad_norm": 0.11028746515512466,
      "learning_rate": 4.41130540086494e-05,
      "loss": 0.0582,
      "num_input_tokens_seen": 16915664,
      "step": 8920
    },
    {
      "epoch": 1.4559915164369035,
      "grad_norm": 0.8138248920440674,
      "learning_rate": 4.4106724214455754e-05,
      "loss": 0.1095,
      "num_input_tokens_seen": 16925824,
      "step": 8925
    },
    {
      "epoch": 1.4568072436577209,
      "grad_norm": 0.11898451298475266,
      "learning_rate": 4.4100391473765115e-05,
      "loss": 0.2233,
      "num_input_tokens_seen": 16936320,
      "step": 8930
    },
    {
      "epoch": 1.4576229708785382,
      "grad_norm": 2.2144393920898438,
      "learning_rate": 4.409405578755408e-05,
      "loss": 0.1953,
      "num_input_tokens_seen": 16946144,
      "step": 8935
    },
    {
      "epoch": 1.4584386980993556,
      "grad_norm": 0.7610256671905518,
      "learning_rate": 4.4087717156799705e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 16956608,
      "step": 8940
    },
    {
      "epoch": 1.459254425320173,
      "grad_norm": 0.4362568259239197,
      "learning_rate": 4.408137558247946e-05,
      "loss": 0.187,
      "num_input_tokens_seen": 16966432,
      "step": 8945
    },
    {
      "epoch": 1.4600701525409903,
      "grad_norm": 2.5663201808929443,
      "learning_rate": 4.4075031065571306e-05,
      "loss": 0.3169,
      "num_input_tokens_seen": 16976352,
      "step": 8950
    },
    {
      "epoch": 1.4608858797618076,
      "grad_norm": 0.744220495223999,
      "learning_rate": 4.406868360705366e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 16985760,
      "step": 8955
    },
    {
      "epoch": 1.461701606982625,
      "grad_norm": 0.6440314054489136,
      "learning_rate": 4.406233320790536e-05,
      "loss": 0.192,
      "num_input_tokens_seen": 16995312,
      "step": 8960
    },
    {
      "epoch": 1.4625173342034423,
      "grad_norm": 0.7419310212135315,
      "learning_rate": 4.4055979869105734e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 17005776,
      "step": 8965
    },
    {
      "epoch": 1.4633330614242597,
      "grad_norm": 0.12780769169330597,
      "learning_rate": 4.404962359163454e-05,
      "loss": 0.1271,
      "num_input_tokens_seen": 17014848,
      "step": 8970
    },
    {
      "epoch": 1.464148788645077,
      "grad_norm": 0.11650644242763519,
      "learning_rate": 4.404326437647199e-05,
      "loss": 0.0748,
      "num_input_tokens_seen": 17024064,
      "step": 8975
    },
    {
      "epoch": 1.4649645158658944,
      "grad_norm": 0.1500168889760971,
      "learning_rate": 4.403690222459877e-05,
      "loss": 0.0844,
      "num_input_tokens_seen": 17033584,
      "step": 8980
    },
    {
      "epoch": 1.4657802430867117,
      "grad_norm": 0.11435669660568237,
      "learning_rate": 4.4030537136995984e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 17043712,
      "step": 8985
    },
    {
      "epoch": 1.466595970307529,
      "grad_norm": 0.2530133128166199,
      "learning_rate": 4.402416911464523e-05,
      "loss": 0.2222,
      "num_input_tokens_seen": 17052848,
      "step": 8990
    },
    {
      "epoch": 1.4674116975283464,
      "grad_norm": 1.2783663272857666,
      "learning_rate": 4.4017798158528516e-05,
      "loss": 0.1104,
      "num_input_tokens_seen": 17063280,
      "step": 8995
    },
    {
      "epoch": 1.468227424749164,
      "grad_norm": 1.7204605340957642,
      "learning_rate": 4.401142426962834e-05,
      "loss": 0.1872,
      "num_input_tokens_seen": 17073648,
      "step": 9000
    },
    {
      "epoch": 1.468227424749164,
      "eval_loss": 0.1503739356994629,
      "eval_runtime": 133.2777,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 17073648,
      "step": 9000
    },
    {
      "epoch": 1.4690431519699811,
      "grad_norm": 0.5554591417312622,
      "learning_rate": 4.400504744892763e-05,
      "loss": 0.0914,
      "num_input_tokens_seen": 17082832,
      "step": 9005
    },
    {
      "epoch": 1.4698588791907987,
      "grad_norm": 0.7005682587623596,
      "learning_rate": 4.399866769740975e-05,
      "loss": 0.0994,
      "num_input_tokens_seen": 17091104,
      "step": 9010
    },
    {
      "epoch": 1.4706746064116158,
      "grad_norm": 0.6032952666282654,
      "learning_rate": 4.399228501605859e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 17098544,
      "step": 9015
    },
    {
      "epoch": 1.4714903336324334,
      "grad_norm": 0.570702850818634,
      "learning_rate": 4.398589940585839e-05,
      "loss": 0.2291,
      "num_input_tokens_seen": 17108640,
      "step": 9020
    },
    {
      "epoch": 1.4723060608532506,
      "grad_norm": 0.14274759590625763,
      "learning_rate": 4.3979510867793917e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 17117504,
      "step": 9025
    },
    {
      "epoch": 1.4731217880740681,
      "grad_norm": 0.7756927013397217,
      "learning_rate": 4.3973119402850346e-05,
      "loss": 0.1146,
      "num_input_tokens_seen": 17125792,
      "step": 9030
    },
    {
      "epoch": 1.4739375152948853,
      "grad_norm": 1.0868374109268188,
      "learning_rate": 4.396672501201334e-05,
      "loss": 0.1837,
      "num_input_tokens_seen": 17135040,
      "step": 9035
    },
    {
      "epoch": 1.4747532425157028,
      "grad_norm": 0.3438683748245239,
      "learning_rate": 4.396032769626899e-05,
      "loss": 0.0754,
      "num_input_tokens_seen": 17143472,
      "step": 9040
    },
    {
      "epoch": 1.4755689697365202,
      "grad_norm": 0.9690688848495483,
      "learning_rate": 4.395392745660384e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 17153472,
      "step": 9045
    },
    {
      "epoch": 1.4763846969573375,
      "grad_norm": 1.3243076801300049,
      "learning_rate": 4.394752429400488e-05,
      "loss": 0.248,
      "num_input_tokens_seen": 17163584,
      "step": 9050
    },
    {
      "epoch": 1.477200424178155,
      "grad_norm": 0.07478995621204376,
      "learning_rate": 4.394111820945957e-05,
      "loss": 0.0822,
      "num_input_tokens_seen": 17174704,
      "step": 9055
    },
    {
      "epoch": 1.4780161513989722,
      "grad_norm": 0.16229937970638275,
      "learning_rate": 4.393470920395579e-05,
      "loss": 0.2295,
      "num_input_tokens_seen": 17183232,
      "step": 9060
    },
    {
      "epoch": 1.4788318786197896,
      "grad_norm": 0.17252646386623383,
      "learning_rate": 4.392829727848192e-05,
      "loss": 0.176,
      "num_input_tokens_seen": 17192640,
      "step": 9065
    },
    {
      "epoch": 1.479647605840607,
      "grad_norm": 1.6974482536315918,
      "learning_rate": 4.392188243402673e-05,
      "loss": 0.2216,
      "num_input_tokens_seen": 17202432,
      "step": 9070
    },
    {
      "epoch": 1.4804633330614243,
      "grad_norm": 0.7701851725578308,
      "learning_rate": 4.391546467157949e-05,
      "loss": 0.0285,
      "num_input_tokens_seen": 17212496,
      "step": 9075
    },
    {
      "epoch": 1.4812790602822417,
      "grad_norm": 0.30414053797721863,
      "learning_rate": 4.390904399212988e-05,
      "loss": 0.1157,
      "num_input_tokens_seen": 17223056,
      "step": 9080
    },
    {
      "epoch": 1.482094787503059,
      "grad_norm": 0.2663792669773102,
      "learning_rate": 4.390262039666807e-05,
      "loss": 0.0533,
      "num_input_tokens_seen": 17233760,
      "step": 9085
    },
    {
      "epoch": 1.4829105147238764,
      "grad_norm": 0.7342919111251831,
      "learning_rate": 4.389619388618464e-05,
      "loss": 0.1246,
      "num_input_tokens_seen": 17242592,
      "step": 9090
    },
    {
      "epoch": 1.4837262419446937,
      "grad_norm": 0.9883701801300049,
      "learning_rate": 4.3889764461670655e-05,
      "loss": 0.1119,
      "num_input_tokens_seen": 17252512,
      "step": 9095
    },
    {
      "epoch": 1.484541969165511,
      "grad_norm": 0.35924220085144043,
      "learning_rate": 4.38833321241176e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 17262784,
      "step": 9100
    },
    {
      "epoch": 1.4853576963863284,
      "grad_norm": 1.7199273109436035,
      "learning_rate": 4.3876896874517434e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 17272224,
      "step": 9105
    },
    {
      "epoch": 1.4861734236071458,
      "grad_norm": 0.18580040335655212,
      "learning_rate": 4.3870458713862554e-05,
      "loss": 0.1075,
      "num_input_tokens_seen": 17281984,
      "step": 9110
    },
    {
      "epoch": 1.4869891508279631,
      "grad_norm": 0.7650999426841736,
      "learning_rate": 4.386401764314579e-05,
      "loss": 0.2469,
      "num_input_tokens_seen": 17290848,
      "step": 9115
    },
    {
      "epoch": 1.4878048780487805,
      "grad_norm": 1.446846604347229,
      "learning_rate": 4.385757366336045e-05,
      "loss": 0.2552,
      "num_input_tokens_seen": 17301184,
      "step": 9120
    },
    {
      "epoch": 1.4886206052695978,
      "grad_norm": 1.1035858392715454,
      "learning_rate": 4.385112677550027e-05,
      "loss": 0.3517,
      "num_input_tokens_seen": 17311632,
      "step": 9125
    },
    {
      "epoch": 1.4894363324904152,
      "grad_norm": 0.9619936347007751,
      "learning_rate": 4.384467698055945e-05,
      "loss": 0.0741,
      "num_input_tokens_seen": 17321360,
      "step": 9130
    },
    {
      "epoch": 1.4902520597112325,
      "grad_norm": 0.9024060368537903,
      "learning_rate": 4.383822427953261e-05,
      "loss": 0.0812,
      "num_input_tokens_seen": 17331680,
      "step": 9135
    },
    {
      "epoch": 1.4910677869320499,
      "grad_norm": 2.1097490787506104,
      "learning_rate": 4.3831768673414864e-05,
      "loss": 0.1871,
      "num_input_tokens_seen": 17342320,
      "step": 9140
    },
    {
      "epoch": 1.4918835141528672,
      "grad_norm": 0.04759790748357773,
      "learning_rate": 4.382531016320173e-05,
      "loss": 0.0855,
      "num_input_tokens_seen": 17353024,
      "step": 9145
    },
    {
      "epoch": 1.4926992413736846,
      "grad_norm": 0.14702722430229187,
      "learning_rate": 4.3818848749889184e-05,
      "loss": 0.0922,
      "num_input_tokens_seen": 17364560,
      "step": 9150
    },
    {
      "epoch": 1.493514968594502,
      "grad_norm": 0.382320374250412,
      "learning_rate": 4.381238443447368e-05,
      "loss": 0.1883,
      "num_input_tokens_seen": 17373424,
      "step": 9155
    },
    {
      "epoch": 1.4943306958153193,
      "grad_norm": 0.2512596845626831,
      "learning_rate": 4.380591721795208e-05,
      "loss": 0.1936,
      "num_input_tokens_seen": 17382992,
      "step": 9160
    },
    {
      "epoch": 1.4951464230361367,
      "grad_norm": 0.0660560354590416,
      "learning_rate": 4.3799447101321723e-05,
      "loss": 0.337,
      "num_input_tokens_seen": 17392160,
      "step": 9165
    },
    {
      "epoch": 1.495962150256954,
      "grad_norm": 0.6863941550254822,
      "learning_rate": 4.379297408558036e-05,
      "loss": 0.1894,
      "num_input_tokens_seen": 17400592,
      "step": 9170
    },
    {
      "epoch": 1.4967778774777714,
      "grad_norm": 0.2587314546108246,
      "learning_rate": 4.378649817172624e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 17409744,
      "step": 9175
    },
    {
      "epoch": 1.4975936046985887,
      "grad_norm": 0.3921236991882324,
      "learning_rate": 4.378001936075801e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 17420688,
      "step": 9180
    },
    {
      "epoch": 1.4984093319194063,
      "grad_norm": 0.5093157291412354,
      "learning_rate": 4.377353765367479e-05,
      "loss": 0.1887,
      "num_input_tokens_seen": 17430288,
      "step": 9185
    },
    {
      "epoch": 1.4992250591402234,
      "grad_norm": 1.1277450323104858,
      "learning_rate": 4.376705305147614e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 17439216,
      "step": 9190
    },
    {
      "epoch": 1.500040786361041,
      "grad_norm": 0.08460885286331177,
      "learning_rate": 4.376056555516206e-05,
      "loss": 0.102,
      "num_input_tokens_seen": 17448000,
      "step": 9195
    },
    {
      "epoch": 1.5008565135818581,
      "grad_norm": 0.1937606781721115,
      "learning_rate": 4.375407516573302e-05,
      "loss": 0.0465,
      "num_input_tokens_seen": 17457952,
      "step": 9200
    },
    {
      "epoch": 1.5008565135818581,
      "eval_loss": 0.15147331357002258,
      "eval_runtime": 133.1618,
      "eval_samples_per_second": 20.464,
      "eval_steps_per_second": 10.236,
      "num_input_tokens_seen": 17457952,
      "step": 9200
    },
    {
      "epoch": 1.5016722408026757,
      "grad_norm": 0.4141722023487091,
      "learning_rate": 4.3747581884189913e-05,
      "loss": 0.0932,
      "num_input_tokens_seen": 17467488,
      "step": 9205
    },
    {
      "epoch": 1.5024879680234928,
      "grad_norm": 0.22016359865665436,
      "learning_rate": 4.374108571153408e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 17477440,
      "step": 9210
    },
    {
      "epoch": 1.5033036952443104,
      "grad_norm": 0.3062012493610382,
      "learning_rate": 4.3734586648767316e-05,
      "loss": 0.082,
      "num_input_tokens_seen": 17486448,
      "step": 9215
    },
    {
      "epoch": 1.5041194224651275,
      "grad_norm": 0.6065188050270081,
      "learning_rate": 4.372808469689186e-05,
      "loss": 0.1216,
      "num_input_tokens_seen": 17496656,
      "step": 9220
    },
    {
      "epoch": 1.504935149685945,
      "grad_norm": 0.16671256721019745,
      "learning_rate": 4.372157985691039e-05,
      "loss": 0.028,
      "num_input_tokens_seen": 17505248,
      "step": 9225
    },
    {
      "epoch": 1.5057508769067622,
      "grad_norm": 0.37062713503837585,
      "learning_rate": 4.371507212982603e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 17514912,
      "step": 9230
    },
    {
      "epoch": 1.5065666041275798,
      "grad_norm": 0.6626244783401489,
      "learning_rate": 4.370856151664236e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 17525248,
      "step": 9235
    },
    {
      "epoch": 1.507382331348397,
      "grad_norm": 0.7688473463058472,
      "learning_rate": 4.3702048018363404e-05,
      "loss": 0.0729,
      "num_input_tokens_seen": 17534288,
      "step": 9240
    },
    {
      "epoch": 1.5081980585692145,
      "grad_norm": 0.060627281665802,
      "learning_rate": 4.369553163599362e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 17544992,
      "step": 9245
    },
    {
      "epoch": 1.5090137857900316,
      "grad_norm": 2.2579240798950195,
      "learning_rate": 4.3689012370537904e-05,
      "loss": 0.3233,
      "num_input_tokens_seen": 17553744,
      "step": 9250
    },
    {
      "epoch": 1.5098295130108492,
      "grad_norm": 1.381569504737854,
      "learning_rate": 4.368249022300164e-05,
      "loss": 0.2838,
      "num_input_tokens_seen": 17563792,
      "step": 9255
    },
    {
      "epoch": 1.5106452402316666,
      "grad_norm": 0.9691140055656433,
      "learning_rate": 4.367596519439059e-05,
      "loss": 0.1169,
      "num_input_tokens_seen": 17572512,
      "step": 9260
    },
    {
      "epoch": 1.511460967452484,
      "grad_norm": 0.25864893198013306,
      "learning_rate": 4.366943728571101e-05,
      "loss": 0.2137,
      "num_input_tokens_seen": 17580544,
      "step": 9265
    },
    {
      "epoch": 1.5122766946733013,
      "grad_norm": 0.18843324482440948,
      "learning_rate": 4.366290649796959e-05,
      "loss": 0.0615,
      "num_input_tokens_seen": 17589264,
      "step": 9270
    },
    {
      "epoch": 1.5130924218941186,
      "grad_norm": 3.4266324043273926,
      "learning_rate": 4.3656372832173456e-05,
      "loss": 0.2316,
      "num_input_tokens_seen": 17599024,
      "step": 9275
    },
    {
      "epoch": 1.513908149114936,
      "grad_norm": 0.2100258469581604,
      "learning_rate": 4.364983628933017e-05,
      "loss": 0.1068,
      "num_input_tokens_seen": 17608544,
      "step": 9280
    },
    {
      "epoch": 1.5147238763357533,
      "grad_norm": 1.3646409511566162,
      "learning_rate": 4.364329687044777e-05,
      "loss": 0.1246,
      "num_input_tokens_seen": 17617424,
      "step": 9285
    },
    {
      "epoch": 1.5155396035565707,
      "grad_norm": 0.40810835361480713,
      "learning_rate": 4.36367545765347e-05,
      "loss": 0.0621,
      "num_input_tokens_seen": 17625776,
      "step": 9290
    },
    {
      "epoch": 1.516355330777388,
      "grad_norm": 0.4452776312828064,
      "learning_rate": 4.363020940859988e-05,
      "loss": 0.2639,
      "num_input_tokens_seen": 17634752,
      "step": 9295
    },
    {
      "epoch": 1.5171710579982054,
      "grad_norm": 0.7932855486869812,
      "learning_rate": 4.362366136765263e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 17644256,
      "step": 9300
    },
    {
      "epoch": 1.5179867852190227,
      "grad_norm": 2.3757386207580566,
      "learning_rate": 4.361711045470278e-05,
      "loss": 0.2421,
      "num_input_tokens_seen": 17653328,
      "step": 9305
    },
    {
      "epoch": 1.51880251243984,
      "grad_norm": 0.4240168035030365,
      "learning_rate": 4.3610556670760524e-05,
      "loss": 0.0933,
      "num_input_tokens_seen": 17663408,
      "step": 9310
    },
    {
      "epoch": 1.5196182396606575,
      "grad_norm": 0.2419804483652115,
      "learning_rate": 4.360400001683657e-05,
      "loss": 0.1061,
      "num_input_tokens_seen": 17673296,
      "step": 9315
    },
    {
      "epoch": 1.5204339668814748,
      "grad_norm": 0.3404192626476288,
      "learning_rate": 4.3597440493942e-05,
      "loss": 0.0527,
      "num_input_tokens_seen": 17682736,
      "step": 9320
    },
    {
      "epoch": 1.5212496941022922,
      "grad_norm": 0.04683191701769829,
      "learning_rate": 4.3590878103088405e-05,
      "loss": 0.1114,
      "num_input_tokens_seen": 17692304,
      "step": 9325
    },
    {
      "epoch": 1.5220654213231095,
      "grad_norm": 0.5178383588790894,
      "learning_rate": 4.358431284528779e-05,
      "loss": 0.1025,
      "num_input_tokens_seen": 17701536,
      "step": 9330
    },
    {
      "epoch": 1.5228811485439269,
      "grad_norm": 0.5885770916938782,
      "learning_rate": 4.357774472155257e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 17710720,
      "step": 9335
    },
    {
      "epoch": 1.5236968757647442,
      "grad_norm": 0.3403792381286621,
      "learning_rate": 4.3571173732895664e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 17719568,
      "step": 9340
    },
    {
      "epoch": 1.5245126029855616,
      "grad_norm": 0.24179403483867645,
      "learning_rate": 4.356459988033039e-05,
      "loss": 0.0509,
      "num_input_tokens_seen": 17727904,
      "step": 9345
    },
    {
      "epoch": 1.5253283302063791,
      "grad_norm": 0.2711013853549957,
      "learning_rate": 4.355802316487051e-05,
      "loss": 0.3421,
      "num_input_tokens_seen": 17735936,
      "step": 9350
    },
    {
      "epoch": 1.5261440574271963,
      "grad_norm": 0.299312561750412,
      "learning_rate": 4.355144358753025e-05,
      "loss": 0.0948,
      "num_input_tokens_seen": 17745520,
      "step": 9355
    },
    {
      "epoch": 1.5269597846480139,
      "grad_norm": 2.000023126602173,
      "learning_rate": 4.354486114932425e-05,
      "loss": 0.2418,
      "num_input_tokens_seen": 17755600,
      "step": 9360
    },
    {
      "epoch": 1.527775511868831,
      "grad_norm": 0.686103343963623,
      "learning_rate": 4.353827585126762e-05,
      "loss": 0.0794,
      "num_input_tokens_seen": 17765296,
      "step": 9365
    },
    {
      "epoch": 1.5285912390896486,
      "grad_norm": 1.2450051307678223,
      "learning_rate": 4.353168769437588e-05,
      "loss": 0.1097,
      "num_input_tokens_seen": 17774240,
      "step": 9370
    },
    {
      "epoch": 1.5294069663104657,
      "grad_norm": 0.14063233137130737,
      "learning_rate": 4.3525096679665014e-05,
      "loss": 0.0893,
      "num_input_tokens_seen": 17783600,
      "step": 9375
    },
    {
      "epoch": 1.5302226935312833,
      "grad_norm": 0.14145971834659576,
      "learning_rate": 4.351850280815144e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 17793040,
      "step": 9380
    },
    {
      "epoch": 1.5310384207521004,
      "grad_norm": 0.2045295387506485,
      "learning_rate": 4.3511906080852014e-05,
      "loss": 0.0926,
      "num_input_tokens_seen": 17802144,
      "step": 9385
    },
    {
      "epoch": 1.531854147972918,
      "grad_norm": 0.4853304624557495,
      "learning_rate": 4.350530649878404e-05,
      "loss": 0.0966,
      "num_input_tokens_seen": 17812272,
      "step": 9390
    },
    {
      "epoch": 1.532669875193735,
      "grad_norm": 1.3871865272521973,
      "learning_rate": 4.3498704062965246e-05,
      "loss": 0.1763,
      "num_input_tokens_seen": 17821488,
      "step": 9395
    },
    {
      "epoch": 1.5334856024145527,
      "grad_norm": 0.6763354539871216,
      "learning_rate": 4.3492098774413815e-05,
      "loss": 0.2969,
      "num_input_tokens_seen": 17831104,
      "step": 9400
    },
    {
      "epoch": 1.5334856024145527,
      "eval_loss": 0.15347765386104584,
      "eval_runtime": 133.2603,
      "eval_samples_per_second": 20.449,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 17831104,
      "step": 9400
    },
    {
      "epoch": 1.5343013296353698,
      "grad_norm": 1.7858043909072876,
      "learning_rate": 4.3485490634148375e-05,
      "loss": 0.3466,
      "num_input_tokens_seen": 17841200,
      "step": 9405
    },
    {
      "epoch": 1.5351170568561874,
      "grad_norm": 0.11198416352272034,
      "learning_rate": 4.347887964318797e-05,
      "loss": 0.1503,
      "num_input_tokens_seen": 17849904,
      "step": 9410
    },
    {
      "epoch": 1.5359327840770045,
      "grad_norm": 0.8223466873168945,
      "learning_rate": 4.34722658025521e-05,
      "loss": 0.1405,
      "num_input_tokens_seen": 17859840,
      "step": 9415
    },
    {
      "epoch": 1.536748511297822,
      "grad_norm": 0.2614806890487671,
      "learning_rate": 4.346564911326071e-05,
      "loss": 0.1248,
      "num_input_tokens_seen": 17870336,
      "step": 9420
    },
    {
      "epoch": 1.5375642385186392,
      "grad_norm": 0.19580702483654022,
      "learning_rate": 4.345902957633418e-05,
      "loss": 0.0447,
      "num_input_tokens_seen": 17880304,
      "step": 9425
    },
    {
      "epoch": 1.5383799657394568,
      "grad_norm": 1.5845788717269897,
      "learning_rate": 4.345240719279331e-05,
      "loss": 0.2333,
      "num_input_tokens_seen": 17889008,
      "step": 9430
    },
    {
      "epoch": 1.539195692960274,
      "grad_norm": 0.3135697543621063,
      "learning_rate": 4.3445781963659374e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 17898992,
      "step": 9435
    },
    {
      "epoch": 1.5400114201810915,
      "grad_norm": 1.3650370836257935,
      "learning_rate": 4.3439153889954045e-05,
      "loss": 0.1231,
      "num_input_tokens_seen": 17909168,
      "step": 9440
    },
    {
      "epoch": 1.5408271474019088,
      "grad_norm": 0.4534240961074829,
      "learning_rate": 4.343252297269946e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 17918720,
      "step": 9445
    },
    {
      "epoch": 1.5416428746227262,
      "grad_norm": 0.24356012046337128,
      "learning_rate": 4.342588921291821e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 17929136,
      "step": 9450
    },
    {
      "epoch": 1.5424586018435436,
      "grad_norm": 0.376162052154541,
      "learning_rate": 4.341925261163328e-05,
      "loss": 0.0957,
      "num_input_tokens_seen": 17938032,
      "step": 9455
    },
    {
      "epoch": 1.543274329064361,
      "grad_norm": 0.48543480038642883,
      "learning_rate": 4.341261316986813e-05,
      "loss": 0.0865,
      "num_input_tokens_seen": 17947840,
      "step": 9460
    },
    {
      "epoch": 1.5440900562851783,
      "grad_norm": 0.8626775145530701,
      "learning_rate": 4.340597088864664e-05,
      "loss": 0.0555,
      "num_input_tokens_seen": 17958224,
      "step": 9465
    },
    {
      "epoch": 1.5449057835059956,
      "grad_norm": 0.5767843723297119,
      "learning_rate": 4.339932576899313e-05,
      "loss": 0.111,
      "num_input_tokens_seen": 17968144,
      "step": 9470
    },
    {
      "epoch": 1.545721510726813,
      "grad_norm": 0.516126275062561,
      "learning_rate": 4.3392677811932375e-05,
      "loss": 0.2011,
      "num_input_tokens_seen": 17978176,
      "step": 9475
    },
    {
      "epoch": 1.5465372379476303,
      "grad_norm": 0.055376194417476654,
      "learning_rate": 4.338602701848956e-05,
      "loss": 0.0599,
      "num_input_tokens_seen": 17987040,
      "step": 9480
    },
    {
      "epoch": 1.5473529651684477,
      "grad_norm": 0.18649710714817047,
      "learning_rate": 4.337937338969033e-05,
      "loss": 0.1784,
      "num_input_tokens_seen": 17995904,
      "step": 9485
    },
    {
      "epoch": 1.548168692389265,
      "grad_norm": 1.2794703245162964,
      "learning_rate": 4.337271692656075e-05,
      "loss": 0.1859,
      "num_input_tokens_seen": 18006832,
      "step": 9490
    },
    {
      "epoch": 1.5489844196100824,
      "grad_norm": 0.5801422595977783,
      "learning_rate": 4.336605763012733e-05,
      "loss": 0.26,
      "num_input_tokens_seen": 18016400,
      "step": 9495
    },
    {
      "epoch": 1.5498001468308997,
      "grad_norm": 1.0992947816848755,
      "learning_rate": 4.3359395501417026e-05,
      "loss": 0.3239,
      "num_input_tokens_seen": 18026976,
      "step": 9500
    },
    {
      "epoch": 1.550615874051717,
      "grad_norm": 0.4882197976112366,
      "learning_rate": 4.335273054145722e-05,
      "loss": 0.243,
      "num_input_tokens_seen": 18037520,
      "step": 9505
    },
    {
      "epoch": 1.5514316012725344,
      "grad_norm": 0.3303312063217163,
      "learning_rate": 4.334606275127572e-05,
      "loss": 0.0742,
      "num_input_tokens_seen": 18047264,
      "step": 9510
    },
    {
      "epoch": 1.5522473284933518,
      "grad_norm": 0.5504597425460815,
      "learning_rate": 4.33393921319008e-05,
      "loss": 0.0932,
      "num_input_tokens_seen": 18056320,
      "step": 9515
    },
    {
      "epoch": 1.5530630557141691,
      "grad_norm": 1.1761449575424194,
      "learning_rate": 4.3332718684361146e-05,
      "loss": 0.2304,
      "num_input_tokens_seen": 18065968,
      "step": 9520
    },
    {
      "epoch": 1.5538787829349865,
      "grad_norm": 0.7201607823371887,
      "learning_rate": 4.332604240968588e-05,
      "loss": 0.0835,
      "num_input_tokens_seen": 18074704,
      "step": 9525
    },
    {
      "epoch": 1.5546945101558038,
      "grad_norm": 1.5084795951843262,
      "learning_rate": 4.331936330890459e-05,
      "loss": 0.1237,
      "num_input_tokens_seen": 18083808,
      "step": 9530
    },
    {
      "epoch": 1.5555102373766214,
      "grad_norm": 0.28152167797088623,
      "learning_rate": 4.331268138304725e-05,
      "loss": 0.0664,
      "num_input_tokens_seen": 18093920,
      "step": 9535
    },
    {
      "epoch": 1.5563259645974385,
      "grad_norm": 1.0916935205459595,
      "learning_rate": 4.330599663314431e-05,
      "loss": 0.107,
      "num_input_tokens_seen": 18104016,
      "step": 9540
    },
    {
      "epoch": 1.5571416918182561,
      "grad_norm": 0.07125560939311981,
      "learning_rate": 4.329930906022665e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 18113264,
      "step": 9545
    },
    {
      "epoch": 1.5579574190390733,
      "grad_norm": 0.051106132566928864,
      "learning_rate": 4.3292618665325564e-05,
      "loss": 0.0795,
      "num_input_tokens_seen": 18120912,
      "step": 9550
    },
    {
      "epoch": 1.5587731462598908,
      "grad_norm": 1.4396352767944336,
      "learning_rate": 4.3285925449472796e-05,
      "loss": 0.2941,
      "num_input_tokens_seen": 18130816,
      "step": 9555
    },
    {
      "epoch": 1.559588873480708,
      "grad_norm": 0.550153374671936,
      "learning_rate": 4.327922941370054e-05,
      "loss": 0.1179,
      "num_input_tokens_seen": 18140128,
      "step": 9560
    },
    {
      "epoch": 1.5604046007015255,
      "grad_norm": 0.6566241383552551,
      "learning_rate": 4.3272530559041384e-05,
      "loss": 0.1161,
      "num_input_tokens_seen": 18150000,
      "step": 9565
    },
    {
      "epoch": 1.5612203279223427,
      "grad_norm": 0.7337068915367126,
      "learning_rate": 4.32658288865284e-05,
      "loss": 0.1215,
      "num_input_tokens_seen": 18159712,
      "step": 9570
    },
    {
      "epoch": 1.5620360551431602,
      "grad_norm": 1.9657013416290283,
      "learning_rate": 4.325912439719505e-05,
      "loss": 0.2383,
      "num_input_tokens_seen": 18169936,
      "step": 9575
    },
    {
      "epoch": 1.5628517823639774,
      "grad_norm": 0.15071837604045868,
      "learning_rate": 4.3252417092075266e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 18179120,
      "step": 9580
    },
    {
      "epoch": 1.563667509584795,
      "grad_norm": 0.593845009803772,
      "learning_rate": 4.3245706972203385e-05,
      "loss": 0.1433,
      "num_input_tokens_seen": 18189504,
      "step": 9585
    },
    {
      "epoch": 1.564483236805612,
      "grad_norm": 1.4379785060882568,
      "learning_rate": 4.323899403861421e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 18197088,
      "step": 9590
    },
    {
      "epoch": 1.5652989640264297,
      "grad_norm": 0.6419689059257507,
      "learning_rate": 4.3232278292342935e-05,
      "loss": 0.2145,
      "num_input_tokens_seen": 18206048,
      "step": 9595
    },
    {
      "epoch": 1.5661146912472468,
      "grad_norm": 0.05928565561771393,
      "learning_rate": 4.322555973442524e-05,
      "loss": 0.1203,
      "num_input_tokens_seen": 18215168,
      "step": 9600
    },
    {
      "epoch": 1.5661146912472468,
      "eval_loss": 0.16439342498779297,
      "eval_runtime": 133.1009,
      "eval_samples_per_second": 20.473,
      "eval_steps_per_second": 10.24,
      "num_input_tokens_seen": 18215168,
      "step": 9600
    },
    {
      "epoch": 1.5669304184680644,
      "grad_norm": 0.2015770971775055,
      "learning_rate": 4.3218838365897184e-05,
      "loss": 0.0475,
      "num_input_tokens_seen": 18223648,
      "step": 9605
    },
    {
      "epoch": 1.5677461456888815,
      "grad_norm": 0.06495405733585358,
      "learning_rate": 4.3212114187795306e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 18233632,
      "step": 9610
    },
    {
      "epoch": 1.568561872909699,
      "grad_norm": 1.0088409185409546,
      "learning_rate": 4.320538720115656e-05,
      "loss": 0.1809,
      "num_input_tokens_seen": 18242288,
      "step": 9615
    },
    {
      "epoch": 1.5693776001305162,
      "grad_norm": 0.6335096955299377,
      "learning_rate": 4.319865740701831e-05,
      "loss": 0.0438,
      "num_input_tokens_seen": 18251456,
      "step": 9620
    },
    {
      "epoch": 1.5701933273513338,
      "grad_norm": 0.7588249444961548,
      "learning_rate": 4.3191924806418396e-05,
      "loss": 0.2637,
      "num_input_tokens_seen": 18262208,
      "step": 9625
    },
    {
      "epoch": 1.5710090545721511,
      "grad_norm": 1.1061667203903198,
      "learning_rate": 4.318518940039507e-05,
      "loss": 0.1027,
      "num_input_tokens_seen": 18270976,
      "step": 9630
    },
    {
      "epoch": 1.5718247817929685,
      "grad_norm": 0.5319994688034058,
      "learning_rate": 4.3178451189987e-05,
      "loss": 0.1923,
      "num_input_tokens_seen": 18281152,
      "step": 9635
    },
    {
      "epoch": 1.5726405090137858,
      "grad_norm": 0.5007250308990479,
      "learning_rate": 4.3171710176233315e-05,
      "loss": 0.0712,
      "num_input_tokens_seen": 18290544,
      "step": 9640
    },
    {
      "epoch": 1.5734562362346032,
      "grad_norm": 1.4879257678985596,
      "learning_rate": 4.316496636017355e-05,
      "loss": 0.2221,
      "num_input_tokens_seen": 18300000,
      "step": 9645
    },
    {
      "epoch": 1.5742719634554205,
      "grad_norm": 0.7183653712272644,
      "learning_rate": 4.315821974284771e-05,
      "loss": 0.242,
      "num_input_tokens_seen": 18310240,
      "step": 9650
    },
    {
      "epoch": 1.5750876906762379,
      "grad_norm": 2.1082940101623535,
      "learning_rate": 4.315147032529619e-05,
      "loss": 0.1467,
      "num_input_tokens_seen": 18318576,
      "step": 9655
    },
    {
      "epoch": 1.5759034178970552,
      "grad_norm": 0.350856214761734,
      "learning_rate": 4.3144718108559845e-05,
      "loss": 0.0972,
      "num_input_tokens_seen": 18326592,
      "step": 9660
    },
    {
      "epoch": 1.5767191451178726,
      "grad_norm": 0.4726054072380066,
      "learning_rate": 4.3137963093679945e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 18336784,
      "step": 9665
    },
    {
      "epoch": 1.57753487233869,
      "grad_norm": 1.7025309801101685,
      "learning_rate": 4.31312052816982e-05,
      "loss": 0.2818,
      "num_input_tokens_seen": 18345040,
      "step": 9670
    },
    {
      "epoch": 1.5783505995595073,
      "grad_norm": 1.3877065181732178,
      "learning_rate": 4.312444467365675e-05,
      "loss": 0.2377,
      "num_input_tokens_seen": 18354256,
      "step": 9675
    },
    {
      "epoch": 1.5791663267803246,
      "grad_norm": 0.14755016565322876,
      "learning_rate": 4.311768127059816e-05,
      "loss": 0.0566,
      "num_input_tokens_seen": 18363856,
      "step": 9680
    },
    {
      "epoch": 1.579982054001142,
      "grad_norm": 0.040116287767887115,
      "learning_rate": 4.3110915073565444e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 18374288,
      "step": 9685
    },
    {
      "epoch": 1.5807977812219594,
      "grad_norm": 0.1244005337357521,
      "learning_rate": 4.310414608360203e-05,
      "loss": 0.1173,
      "num_input_tokens_seen": 18383680,
      "step": 9690
    },
    {
      "epoch": 1.5816135084427767,
      "grad_norm": 0.2252495437860489,
      "learning_rate": 4.309737430175177e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 18393360,
      "step": 9695
    },
    {
      "epoch": 1.582429235663594,
      "grad_norm": 0.3002210855484009,
      "learning_rate": 4.309059972905897e-05,
      "loss": 0.1163,
      "num_input_tokens_seen": 18403744,
      "step": 9700
    },
    {
      "epoch": 1.5832449628844114,
      "grad_norm": 0.3839481472969055,
      "learning_rate": 4.308382236656836e-05,
      "loss": 0.0646,
      "num_input_tokens_seen": 18413632,
      "step": 9705
    },
    {
      "epoch": 1.5840606901052288,
      "grad_norm": 0.11340807378292084,
      "learning_rate": 4.307704221532507e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 18423120,
      "step": 9710
    },
    {
      "epoch": 1.5848764173260461,
      "grad_norm": 1.560112714767456,
      "learning_rate": 4.307025927637471e-05,
      "loss": 0.1847,
      "num_input_tokens_seen": 18432096,
      "step": 9715
    },
    {
      "epoch": 1.5856921445468637,
      "grad_norm": 0.1367829591035843,
      "learning_rate": 4.306347355076328e-05,
      "loss": 0.049,
      "num_input_tokens_seen": 18442160,
      "step": 9720
    },
    {
      "epoch": 1.5865078717676808,
      "grad_norm": 1.5593241453170776,
      "learning_rate": 4.305668503953724e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 18450672,
      "step": 9725
    },
    {
      "epoch": 1.5873235989884984,
      "grad_norm": 0.669302761554718,
      "learning_rate": 4.3049893743743436e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 18459312,
      "step": 9730
    },
    {
      "epoch": 1.5881393262093155,
      "grad_norm": 0.6259718537330627,
      "learning_rate": 4.304309966442919e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 18468192,
      "step": 9735
    },
    {
      "epoch": 1.588955053430133,
      "grad_norm": 1.6533403396606445,
      "learning_rate": 4.303630280264224e-05,
      "loss": 0.22,
      "num_input_tokens_seen": 18478992,
      "step": 9740
    },
    {
      "epoch": 1.5897707806509502,
      "grad_norm": 1.0256680250167847,
      "learning_rate": 4.302950315943074e-05,
      "loss": 0.2144,
      "num_input_tokens_seen": 18489344,
      "step": 9745
    },
    {
      "epoch": 1.5905865078717678,
      "grad_norm": 0.2032400518655777,
      "learning_rate": 4.3022700735843275e-05,
      "loss": 0.2428,
      "num_input_tokens_seen": 18497952,
      "step": 9750
    },
    {
      "epoch": 1.591402235092585,
      "grad_norm": 0.7120448350906372,
      "learning_rate": 4.301589553292887e-05,
      "loss": 0.265,
      "num_input_tokens_seen": 18507264,
      "step": 9755
    },
    {
      "epoch": 1.5922179623134025,
      "grad_norm": 0.8511444330215454,
      "learning_rate": 4.300908755173697e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 18517648,
      "step": 9760
    },
    {
      "epoch": 1.5930336895342196,
      "grad_norm": 1.4336150884628296,
      "learning_rate": 4.300227679331745e-05,
      "loss": 0.1181,
      "num_input_tokens_seen": 18526704,
      "step": 9765
    },
    {
      "epoch": 1.5938494167550372,
      "grad_norm": 0.22641518712043762,
      "learning_rate": 4.299546325872063e-05,
      "loss": 0.1166,
      "num_input_tokens_seen": 18535888,
      "step": 9770
    },
    {
      "epoch": 1.5946651439758543,
      "grad_norm": 1.0813686847686768,
      "learning_rate": 4.2988646948997225e-05,
      "loss": 0.1984,
      "num_input_tokens_seen": 18545280,
      "step": 9775
    },
    {
      "epoch": 1.595480871196672,
      "grad_norm": 1.7747336626052856,
      "learning_rate": 4.29818278651984e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 18555056,
      "step": 9780
    },
    {
      "epoch": 1.596296598417489,
      "grad_norm": 0.3232709467411041,
      "learning_rate": 4.297500600837574e-05,
      "loss": 0.1979,
      "num_input_tokens_seen": 18564368,
      "step": 9785
    },
    {
      "epoch": 1.5971123256383066,
      "grad_norm": 0.9473532438278198,
      "learning_rate": 4.2968181379581276e-05,
      "loss": 0.1441,
      "num_input_tokens_seen": 18575008,
      "step": 9790
    },
    {
      "epoch": 1.5979280528591238,
      "grad_norm": 0.4023412764072418,
      "learning_rate": 4.296135397986743e-05,
      "loss": 0.1074,
      "num_input_tokens_seen": 18583328,
      "step": 9795
    },
    {
      "epoch": 1.5987437800799413,
      "grad_norm": 0.8890153765678406,
      "learning_rate": 4.295452381028709e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 18592816,
      "step": 9800
    },
    {
      "epoch": 1.5987437800799413,
      "eval_loss": 0.1581130027770996,
      "eval_runtime": 133.2341,
      "eval_samples_per_second": 20.453,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 18592816,
      "step": 9800
    },
    {
      "epoch": 1.5995595073007585,
      "grad_norm": 0.865346372127533,
      "learning_rate": 4.294769087189354e-05,
      "loss": 0.1911,
      "num_input_tokens_seen": 18603504,
      "step": 9805
    },
    {
      "epoch": 1.600375234521576,
      "grad_norm": 0.2187448889017105,
      "learning_rate": 4.294085516574052e-05,
      "loss": 0.1515,
      "num_input_tokens_seen": 18612992,
      "step": 9810
    },
    {
      "epoch": 1.6011909617423934,
      "grad_norm": 0.18107940256595612,
      "learning_rate": 4.2934016692882176e-05,
      "loss": 0.0841,
      "num_input_tokens_seen": 18623152,
      "step": 9815
    },
    {
      "epoch": 1.6020066889632107,
      "grad_norm": 0.40761950612068176,
      "learning_rate": 4.292717545437308e-05,
      "loss": 0.1127,
      "num_input_tokens_seen": 18631696,
      "step": 9820
    },
    {
      "epoch": 1.602822416184028,
      "grad_norm": 1.9460551738739014,
      "learning_rate": 4.292033145126825e-05,
      "loss": 0.2779,
      "num_input_tokens_seen": 18641376,
      "step": 9825
    },
    {
      "epoch": 1.6036381434048455,
      "grad_norm": 0.8123272657394409,
      "learning_rate": 4.29134846846231e-05,
      "loss": 0.2258,
      "num_input_tokens_seen": 18651344,
      "step": 9830
    },
    {
      "epoch": 1.6044538706256628,
      "grad_norm": 0.916522204875946,
      "learning_rate": 4.29066351554935e-05,
      "loss": 0.0843,
      "num_input_tokens_seen": 18660880,
      "step": 9835
    },
    {
      "epoch": 1.6052695978464802,
      "grad_norm": 0.29722630977630615,
      "learning_rate": 4.289978286493574e-05,
      "loss": 0.1066,
      "num_input_tokens_seen": 18669744,
      "step": 9840
    },
    {
      "epoch": 1.6060853250672975,
      "grad_norm": 0.7242155075073242,
      "learning_rate": 4.28929278140065e-05,
      "loss": 0.146,
      "num_input_tokens_seen": 18679776,
      "step": 9845
    },
    {
      "epoch": 1.6069010522881149,
      "grad_norm": 0.07545256614685059,
      "learning_rate": 4.288607000376295e-05,
      "loss": 0.2676,
      "num_input_tokens_seen": 18690128,
      "step": 9850
    },
    {
      "epoch": 1.6077167795089322,
      "grad_norm": 0.955350935459137,
      "learning_rate": 4.2879209435262624e-05,
      "loss": 0.0987,
      "num_input_tokens_seen": 18699568,
      "step": 9855
    },
    {
      "epoch": 1.6085325067297496,
      "grad_norm": 0.8701638579368591,
      "learning_rate": 4.287234610956353e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 18708832,
      "step": 9860
    },
    {
      "epoch": 1.609348233950567,
      "grad_norm": 1.3338693380355835,
      "learning_rate": 4.2865480027724056e-05,
      "loss": 0.0993,
      "num_input_tokens_seen": 18717376,
      "step": 9865
    },
    {
      "epoch": 1.6101639611713843,
      "grad_norm": 0.7149238586425781,
      "learning_rate": 4.285861119080306e-05,
      "loss": 0.1003,
      "num_input_tokens_seen": 18727168,
      "step": 9870
    },
    {
      "epoch": 1.6109796883922016,
      "grad_norm": 0.7870262861251831,
      "learning_rate": 4.2851739599859784e-05,
      "loss": 0.1277,
      "num_input_tokens_seen": 18737056,
      "step": 9875
    },
    {
      "epoch": 1.611795415613019,
      "grad_norm": 0.6687918901443481,
      "learning_rate": 4.2844865255953934e-05,
      "loss": 0.1109,
      "num_input_tokens_seen": 18745264,
      "step": 9880
    },
    {
      "epoch": 1.6126111428338363,
      "grad_norm": 1.5205010175704956,
      "learning_rate": 4.2837988160145605e-05,
      "loss": 0.3034,
      "num_input_tokens_seen": 18754576,
      "step": 9885
    },
    {
      "epoch": 1.6134268700546537,
      "grad_norm": 1.2761268615722656,
      "learning_rate": 4.2831108313495336e-05,
      "loss": 0.2127,
      "num_input_tokens_seen": 18765440,
      "step": 9890
    },
    {
      "epoch": 1.614242597275471,
      "grad_norm": 1.596524953842163,
      "learning_rate": 4.282422571706408e-05,
      "loss": 0.4129,
      "num_input_tokens_seen": 18775200,
      "step": 9895
    },
    {
      "epoch": 1.6150583244962884,
      "grad_norm": 0.35886842012405396,
      "learning_rate": 4.281734037191323e-05,
      "loss": 0.1139,
      "num_input_tokens_seen": 18784496,
      "step": 9900
    },
    {
      "epoch": 1.615874051717106,
      "grad_norm": 0.6994949579238892,
      "learning_rate": 4.281045227910459e-05,
      "loss": 0.0589,
      "num_input_tokens_seen": 18793344,
      "step": 9905
    },
    {
      "epoch": 1.616689778937923,
      "grad_norm": 0.20159998536109924,
      "learning_rate": 4.280356143970038e-05,
      "loss": 0.099,
      "num_input_tokens_seen": 18801952,
      "step": 9910
    },
    {
      "epoch": 1.6175055061587407,
      "grad_norm": 1.4246608018875122,
      "learning_rate": 4.279666785476327e-05,
      "loss": 0.3032,
      "num_input_tokens_seen": 18812064,
      "step": 9915
    },
    {
      "epoch": 1.6183212333795578,
      "grad_norm": 1.2903294563293457,
      "learning_rate": 4.2789771525356325e-05,
      "loss": 0.2428,
      "num_input_tokens_seen": 18820704,
      "step": 9920
    },
    {
      "epoch": 1.6191369606003754,
      "grad_norm": 1.0478806495666504,
      "learning_rate": 4.2782872452543056e-05,
      "loss": 0.2322,
      "num_input_tokens_seen": 18830272,
      "step": 9925
    },
    {
      "epoch": 1.6199526878211925,
      "grad_norm": 0.13347740471363068,
      "learning_rate": 4.2775970637387376e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 18839872,
      "step": 9930
    },
    {
      "epoch": 1.62076841504201,
      "grad_norm": 1.53057861328125,
      "learning_rate": 4.276906608095363e-05,
      "loss": 0.1839,
      "num_input_tokens_seen": 18850144,
      "step": 9935
    },
    {
      "epoch": 1.6215841422628272,
      "grad_norm": 0.02900485321879387,
      "learning_rate": 4.276215878430661e-05,
      "loss": 0.1874,
      "num_input_tokens_seen": 18859424,
      "step": 9940
    },
    {
      "epoch": 1.6223998694836448,
      "grad_norm": 0.10117574781179428,
      "learning_rate": 4.275524874851149e-05,
      "loss": 0.0882,
      "num_input_tokens_seen": 18869088,
      "step": 9945
    },
    {
      "epoch": 1.623215596704462,
      "grad_norm": 0.3327276110649109,
      "learning_rate": 4.274833597463388e-05,
      "loss": 0.0872,
      "num_input_tokens_seen": 18878144,
      "step": 9950
    },
    {
      "epoch": 1.6240313239252795,
      "grad_norm": 1.5277458429336548,
      "learning_rate": 4.2741420463739824e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 18887328,
      "step": 9955
    },
    {
      "epoch": 1.6248470511460966,
      "grad_norm": 0.21347741782665253,
      "learning_rate": 4.273450221689578e-05,
      "loss": 0.0892,
      "num_input_tokens_seen": 18895968,
      "step": 9960
    },
    {
      "epoch": 1.6256627783669142,
      "grad_norm": 1.5371249914169312,
      "learning_rate": 4.272758123516863e-05,
      "loss": 0.255,
      "num_input_tokens_seen": 18904672,
      "step": 9965
    },
    {
      "epoch": 1.6264785055877313,
      "grad_norm": 1.150486707687378,
      "learning_rate": 4.272065751962567e-05,
      "loss": 0.1016,
      "num_input_tokens_seen": 18914480,
      "step": 9970
    },
    {
      "epoch": 1.627294232808549,
      "grad_norm": 2.12373948097229,
      "learning_rate": 4.271373107133464e-05,
      "loss": 0.241,
      "num_input_tokens_seen": 18924448,
      "step": 9975
    },
    {
      "epoch": 1.628109960029366,
      "grad_norm": 1.8168290853500366,
      "learning_rate": 4.270680189136366e-05,
      "loss": 0.2847,
      "num_input_tokens_seen": 18933888,
      "step": 9980
    },
    {
      "epoch": 1.6289256872501836,
      "grad_norm": 1.04997718334198,
      "learning_rate": 4.269986998078132e-05,
      "loss": 0.1298,
      "num_input_tokens_seen": 18943472,
      "step": 9985
    },
    {
      "epoch": 1.6297414144710007,
      "grad_norm": 0.9330211877822876,
      "learning_rate": 4.2692935340656595e-05,
      "loss": 0.1284,
      "num_input_tokens_seen": 18952624,
      "step": 9990
    },
    {
      "epoch": 1.6305571416918183,
      "grad_norm": 0.5364879369735718,
      "learning_rate": 4.26859979720589e-05,
      "loss": 0.0923,
      "num_input_tokens_seen": 18963456,
      "step": 9995
    },
    {
      "epoch": 1.6313728689126357,
      "grad_norm": 0.9622782468795776,
      "learning_rate": 4.267905787605806e-05,
      "loss": 0.1982,
      "num_input_tokens_seen": 18972864,
      "step": 10000
    },
    {
      "epoch": 1.6313728689126357,
      "eval_loss": 0.14815010130405426,
      "eval_runtime": 133.3656,
      "eval_samples_per_second": 20.433,
      "eval_steps_per_second": 10.22,
      "num_input_tokens_seen": 18972864,
      "step": 10000
    },
    {
      "epoch": 1.632188596133453,
      "grad_norm": 0.4572129547595978,
      "learning_rate": 4.267211505372433e-05,
      "loss": 0.0541,
      "num_input_tokens_seen": 18982160,
      "step": 10005
    },
    {
      "epoch": 1.6330043233542704,
      "grad_norm": 0.39383068680763245,
      "learning_rate": 4.266516950612837e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 18992384,
      "step": 10010
    },
    {
      "epoch": 1.6338200505750877,
      "grad_norm": 0.7858657240867615,
      "learning_rate": 4.265822123434128e-05,
      "loss": 0.1211,
      "num_input_tokens_seen": 19002448,
      "step": 10015
    },
    {
      "epoch": 1.634635777795905,
      "grad_norm": 1.1828116178512573,
      "learning_rate": 4.265127023943457e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 19012416,
      "step": 10020
    },
    {
      "epoch": 1.6354515050167224,
      "grad_norm": 0.9318610429763794,
      "learning_rate": 4.2644316522480176e-05,
      "loss": 0.1528,
      "num_input_tokens_seen": 19021472,
      "step": 10025
    },
    {
      "epoch": 1.6362672322375398,
      "grad_norm": 0.4835682511329651,
      "learning_rate": 4.263736008455044e-05,
      "loss": 0.094,
      "num_input_tokens_seen": 19030720,
      "step": 10030
    },
    {
      "epoch": 1.6370829594583571,
      "grad_norm": 0.16232718527317047,
      "learning_rate": 4.2630400926718125e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 19041136,
      "step": 10035
    },
    {
      "epoch": 1.6378986866791745,
      "grad_norm": 0.2556120753288269,
      "learning_rate": 4.262343905005644e-05,
      "loss": 0.0505,
      "num_input_tokens_seen": 19049696,
      "step": 10040
    },
    {
      "epoch": 1.6387144138999918,
      "grad_norm": 0.24442051351070404,
      "learning_rate": 4.261647445563897e-05,
      "loss": 0.0559,
      "num_input_tokens_seen": 19059504,
      "step": 10045
    },
    {
      "epoch": 1.6395301411208092,
      "grad_norm": 0.24141043424606323,
      "learning_rate": 4.260950714453976e-05,
      "loss": 0.1752,
      "num_input_tokens_seen": 19068480,
      "step": 10050
    },
    {
      "epoch": 1.6403458683416265,
      "grad_norm": 0.29382407665252686,
      "learning_rate": 4.2602537117833266e-05,
      "loss": 0.0747,
      "num_input_tokens_seen": 19078560,
      "step": 10055
    },
    {
      "epoch": 1.641161595562444,
      "grad_norm": 0.9659916162490845,
      "learning_rate": 4.259556437659433e-05,
      "loss": 0.1127,
      "num_input_tokens_seen": 19088080,
      "step": 10060
    },
    {
      "epoch": 1.6419773227832613,
      "grad_norm": 0.6874339580535889,
      "learning_rate": 4.258858892189825e-05,
      "loss": 0.0637,
      "num_input_tokens_seen": 19097136,
      "step": 10065
    },
    {
      "epoch": 1.6427930500040786,
      "grad_norm": 0.07783155888319016,
      "learning_rate": 4.2581610754820725e-05,
      "loss": 0.0601,
      "num_input_tokens_seen": 19105680,
      "step": 10070
    },
    {
      "epoch": 1.643608777224896,
      "grad_norm": 1.4940106868743896,
      "learning_rate": 4.2574629876437876e-05,
      "loss": 0.3008,
      "num_input_tokens_seen": 19114240,
      "step": 10075
    },
    {
      "epoch": 1.6444245044457133,
      "grad_norm": 1.8580517768859863,
      "learning_rate": 4.256764628782625e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 19124464,
      "step": 10080
    },
    {
      "epoch": 1.6452402316665307,
      "grad_norm": 1.738180160522461,
      "learning_rate": 4.256065999006279e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 19134112,
      "step": 10085
    },
    {
      "epoch": 1.6460559588873482,
      "grad_norm": 0.07201196253299713,
      "learning_rate": 4.2553670984224885e-05,
      "loss": 0.3633,
      "num_input_tokens_seen": 19143952,
      "step": 10090
    },
    {
      "epoch": 1.6468716861081654,
      "grad_norm": 0.40311041474342346,
      "learning_rate": 4.254667927139032e-05,
      "loss": 0.2006,
      "num_input_tokens_seen": 19153632,
      "step": 10095
    },
    {
      "epoch": 1.647687413328983,
      "grad_norm": 0.026100026443600655,
      "learning_rate": 4.2539684852637295e-05,
      "loss": 0.0613,
      "num_input_tokens_seen": 19163568,
      "step": 10100
    },
    {
      "epoch": 1.6485031405498,
      "grad_norm": 1.1207398176193237,
      "learning_rate": 4.253268772904446e-05,
      "loss": 0.3732,
      "num_input_tokens_seen": 19174128,
      "step": 10105
    },
    {
      "epoch": 1.6493188677706176,
      "grad_norm": 0.7150819897651672,
      "learning_rate": 4.252568790169085e-05,
      "loss": 0.3752,
      "num_input_tokens_seen": 19183680,
      "step": 10110
    },
    {
      "epoch": 1.6501345949914348,
      "grad_norm": 1.3861331939697266,
      "learning_rate": 4.251868537165592e-05,
      "loss": 0.2501,
      "num_input_tokens_seen": 19194272,
      "step": 10115
    },
    {
      "epoch": 1.6509503222122524,
      "grad_norm": 0.5700134634971619,
      "learning_rate": 4.251168014001955e-05,
      "loss": 0.1431,
      "num_input_tokens_seen": 19203024,
      "step": 10120
    },
    {
      "epoch": 1.6517660494330695,
      "grad_norm": 2.023714780807495,
      "learning_rate": 4.250467220786204e-05,
      "loss": 0.2383,
      "num_input_tokens_seen": 19213536,
      "step": 10125
    },
    {
      "epoch": 1.652581776653887,
      "grad_norm": 0.18492989242076874,
      "learning_rate": 4.249766157626409e-05,
      "loss": 0.2419,
      "num_input_tokens_seen": 19222752,
      "step": 10130
    },
    {
      "epoch": 1.6533975038747042,
      "grad_norm": 0.8741475343704224,
      "learning_rate": 4.249064824630684e-05,
      "loss": 0.1713,
      "num_input_tokens_seen": 19232336,
      "step": 10135
    },
    {
      "epoch": 1.6542132310955218,
      "grad_norm": 0.8039097189903259,
      "learning_rate": 4.248363221907183e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 19241936,
      "step": 10140
    },
    {
      "epoch": 1.655028958316339,
      "grad_norm": 0.11416496336460114,
      "learning_rate": 4.2476613495641026e-05,
      "loss": 0.1598,
      "num_input_tokens_seen": 19251888,
      "step": 10145
    },
    {
      "epoch": 1.6558446855371565,
      "grad_norm": 0.26467621326446533,
      "learning_rate": 4.246959207709679e-05,
      "loss": 0.1144,
      "num_input_tokens_seen": 19261120,
      "step": 10150
    },
    {
      "epoch": 1.6566604127579736,
      "grad_norm": 1.1643755435943604,
      "learning_rate": 4.246256796452192e-05,
      "loss": 0.0908,
      "num_input_tokens_seen": 19270176,
      "step": 10155
    },
    {
      "epoch": 1.6574761399787912,
      "grad_norm": 1.0855470895767212,
      "learning_rate": 4.245554115899962e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 19279584,
      "step": 10160
    },
    {
      "epoch": 1.6582918671996083,
      "grad_norm": 0.050762567669153214,
      "learning_rate": 4.2448511661613514e-05,
      "loss": 0.1912,
      "num_input_tokens_seen": 19288496,
      "step": 10165
    },
    {
      "epoch": 1.6591075944204259,
      "grad_norm": 0.18884317576885223,
      "learning_rate": 4.2441479473447635e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 19297088,
      "step": 10170
    },
    {
      "epoch": 1.659923321641243,
      "grad_norm": 1.5820082426071167,
      "learning_rate": 4.243444459558644e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 19306688,
      "step": 10175
    },
    {
      "epoch": 1.6607390488620606,
      "grad_norm": 0.8419724106788635,
      "learning_rate": 4.24274070291148e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 19315184,
      "step": 10180
    },
    {
      "epoch": 1.661554776082878,
      "grad_norm": 0.07893728464841843,
      "learning_rate": 4.242036677511798e-05,
      "loss": 0.131,
      "num_input_tokens_seen": 19323648,
      "step": 10185
    },
    {
      "epoch": 1.6623705033036953,
      "grad_norm": 0.2567532956600189,
      "learning_rate": 4.241332383468169e-05,
      "loss": 0.21,
      "num_input_tokens_seen": 19331616,
      "step": 10190
    },
    {
      "epoch": 1.6631862305245126,
      "grad_norm": 0.46000930666923523,
      "learning_rate": 4.2406278208892034e-05,
      "loss": 0.1611,
      "num_input_tokens_seen": 19340960,
      "step": 10195
    },
    {
      "epoch": 1.66400195774533,
      "grad_norm": 0.22759266197681427,
      "learning_rate": 4.2399229898835536e-05,
      "loss": 0.0933,
      "num_input_tokens_seen": 19350160,
      "step": 10200
    },
    {
      "epoch": 1.66400195774533,
      "eval_loss": 0.14996075630187988,
      "eval_runtime": 133.2853,
      "eval_samples_per_second": 20.445,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 19350160,
      "step": 10200
    },
    {
      "epoch": 1.6648176849661473,
      "grad_norm": 0.11948738247156143,
      "learning_rate": 4.239217890559914e-05,
      "loss": 0.0238,
      "num_input_tokens_seen": 19360160,
      "step": 10205
    },
    {
      "epoch": 1.6656334121869647,
      "grad_norm": 0.37732720375061035,
      "learning_rate": 4.238512523027019e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 19369696,
      "step": 10210
    },
    {
      "epoch": 1.666449139407782,
      "grad_norm": 0.19359534978866577,
      "learning_rate": 4.237806887393645e-05,
      "loss": 0.1001,
      "num_input_tokens_seen": 19379152,
      "step": 10215
    },
    {
      "epoch": 1.6672648666285994,
      "grad_norm": 2.146620512008667,
      "learning_rate": 4.237100983768611e-05,
      "loss": 0.3407,
      "num_input_tokens_seen": 19389616,
      "step": 10220
    },
    {
      "epoch": 1.6680805938494168,
      "grad_norm": 0.16933868825435638,
      "learning_rate": 4.2363948122607756e-05,
      "loss": 0.0338,
      "num_input_tokens_seen": 19398848,
      "step": 10225
    },
    {
      "epoch": 1.6688963210702341,
      "grad_norm": 1.950792908668518,
      "learning_rate": 4.235688372979039e-05,
      "loss": 0.2246,
      "num_input_tokens_seen": 19408576,
      "step": 10230
    },
    {
      "epoch": 1.6697120482910515,
      "grad_norm": 0.3222319781780243,
      "learning_rate": 4.234981666032343e-05,
      "loss": 0.2613,
      "num_input_tokens_seen": 19417824,
      "step": 10235
    },
    {
      "epoch": 1.6705277755118688,
      "grad_norm": 0.06488750874996185,
      "learning_rate": 4.2342746915296704e-05,
      "loss": 0.1169,
      "num_input_tokens_seen": 19427776,
      "step": 10240
    },
    {
      "epoch": 1.6713435027326862,
      "grad_norm": 0.21755969524383545,
      "learning_rate": 4.233567449580047e-05,
      "loss": 0.1242,
      "num_input_tokens_seen": 19438448,
      "step": 10245
    },
    {
      "epoch": 1.6721592299535035,
      "grad_norm": 0.09093452990055084,
      "learning_rate": 4.232859940292537e-05,
      "loss": 0.2255,
      "num_input_tokens_seen": 19446592,
      "step": 10250
    },
    {
      "epoch": 1.6729749571743209,
      "grad_norm": 1.5399898290634155,
      "learning_rate": 4.232152163776248e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 19456672,
      "step": 10255
    },
    {
      "epoch": 1.6737906843951382,
      "grad_norm": 0.38192421197891235,
      "learning_rate": 4.231444120140328e-05,
      "loss": 0.1111,
      "num_input_tokens_seen": 19466960,
      "step": 10260
    },
    {
      "epoch": 1.6746064116159556,
      "grad_norm": 1.3347219228744507,
      "learning_rate": 4.230735809493967e-05,
      "loss": 0.2199,
      "num_input_tokens_seen": 19476688,
      "step": 10265
    },
    {
      "epoch": 1.675422138836773,
      "grad_norm": 0.1335144191980362,
      "learning_rate": 4.2300272319463926e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 19486592,
      "step": 10270
    },
    {
      "epoch": 1.6762378660575903,
      "grad_norm": 0.25552886724472046,
      "learning_rate": 4.2293183876068786e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 19496128,
      "step": 10275
    },
    {
      "epoch": 1.6770535932784076,
      "grad_norm": 0.5884513258934021,
      "learning_rate": 4.228609276584737e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 19506640,
      "step": 10280
    },
    {
      "epoch": 1.6778693204992252,
      "grad_norm": 0.15040068328380585,
      "learning_rate": 4.227899898989323e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 19515648,
      "step": 10285
    },
    {
      "epoch": 1.6786850477200423,
      "grad_norm": 0.840946614742279,
      "learning_rate": 4.2271902549300293e-05,
      "loss": 0.1119,
      "num_input_tokens_seen": 19524400,
      "step": 10290
    },
    {
      "epoch": 1.67950077494086,
      "grad_norm": 0.16937638819217682,
      "learning_rate": 4.226480344516294e-05,
      "loss": 0.3501,
      "num_input_tokens_seen": 19534608,
      "step": 10295
    },
    {
      "epoch": 1.680316502161677,
      "grad_norm": 0.49192148447036743,
      "learning_rate": 4.2257701678575925e-05,
      "loss": 0.1048,
      "num_input_tokens_seen": 19543936,
      "step": 10300
    },
    {
      "epoch": 1.6811322293824946,
      "grad_norm": 0.7017384171485901,
      "learning_rate": 4.225059725063444e-05,
      "loss": 0.1313,
      "num_input_tokens_seen": 19553200,
      "step": 10305
    },
    {
      "epoch": 1.6819479566033118,
      "grad_norm": 2.192016363143921,
      "learning_rate": 4.2243490162434074e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 19562736,
      "step": 10310
    },
    {
      "epoch": 1.6827636838241293,
      "grad_norm": 0.18217048048973083,
      "learning_rate": 4.223638041507083e-05,
      "loss": 0.0677,
      "num_input_tokens_seen": 19572016,
      "step": 10315
    },
    {
      "epoch": 1.6835794110449465,
      "grad_norm": 0.19525107741355896,
      "learning_rate": 4.2229268009641124e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 19581888,
      "step": 10320
    },
    {
      "epoch": 1.684395138265764,
      "grad_norm": 0.8850154280662537,
      "learning_rate": 4.222215294724177e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 19591264,
      "step": 10325
    },
    {
      "epoch": 1.6852108654865812,
      "grad_norm": 1.3964226245880127,
      "learning_rate": 4.2215035228970005e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 19599568,
      "step": 10330
    },
    {
      "epoch": 1.6860265927073987,
      "grad_norm": 0.5038401484489441,
      "learning_rate": 4.2207914855923464e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 19609504,
      "step": 10335
    },
    {
      "epoch": 1.6868423199282159,
      "grad_norm": 0.3546307384967804,
      "learning_rate": 4.220079182920021e-05,
      "loss": 0.1578,
      "num_input_tokens_seen": 19617536,
      "step": 10340
    },
    {
      "epoch": 1.6876580471490334,
      "grad_norm": 1.3845500946044922,
      "learning_rate": 4.2193666149898705e-05,
      "loss": 0.1124,
      "num_input_tokens_seen": 19627936,
      "step": 10345
    },
    {
      "epoch": 1.6884737743698506,
      "grad_norm": 0.5887707471847534,
      "learning_rate": 4.21865378191178e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 19637552,
      "step": 10350
    },
    {
      "epoch": 1.6892895015906682,
      "grad_norm": 0.2771652042865753,
      "learning_rate": 4.217940683795678e-05,
      "loss": 0.1894,
      "num_input_tokens_seen": 19647424,
      "step": 10355
    },
    {
      "epoch": 1.6901052288114853,
      "grad_norm": 1.567273497581482,
      "learning_rate": 4.217227320751534e-05,
      "loss": 0.1682,
      "num_input_tokens_seen": 19657664,
      "step": 10360
    },
    {
      "epoch": 1.6909209560323029,
      "grad_norm": 0.8360516428947449,
      "learning_rate": 4.216513692889358e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 19666432,
      "step": 10365
    },
    {
      "epoch": 1.6917366832531202,
      "grad_norm": 0.7747339606285095,
      "learning_rate": 4.215799800319199e-05,
      "loss": 0.0789,
      "num_input_tokens_seen": 19675600,
      "step": 10370
    },
    {
      "epoch": 1.6925524104739376,
      "grad_norm": 0.7826972603797913,
      "learning_rate": 4.2150856431511485e-05,
      "loss": 0.1817,
      "num_input_tokens_seen": 19685216,
      "step": 10375
    },
    {
      "epoch": 1.693368137694755,
      "grad_norm": 0.8482260704040527,
      "learning_rate": 4.214371221495339e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 19694544,
      "step": 10380
    },
    {
      "epoch": 1.6941838649155723,
      "grad_norm": 0.07187929004430771,
      "learning_rate": 4.213656535461942e-05,
      "loss": 0.1147,
      "num_input_tokens_seen": 19704864,
      "step": 10385
    },
    {
      "epoch": 1.6949995921363896,
      "grad_norm": 0.3628086447715759,
      "learning_rate": 4.2129415851611734e-05,
      "loss": 0.0564,
      "num_input_tokens_seen": 19714992,
      "step": 10390
    },
    {
      "epoch": 1.695815319357207,
      "grad_norm": 0.5589928030967712,
      "learning_rate": 4.2122263707032855e-05,
      "loss": 0.124,
      "num_input_tokens_seen": 19724784,
      "step": 10395
    },
    {
      "epoch": 1.6966310465780243,
      "grad_norm": 1.3613295555114746,
      "learning_rate": 4.211510892198574e-05,
      "loss": 0.2432,
      "num_input_tokens_seen": 19735024,
      "step": 10400
    },
    {
      "epoch": 1.6966310465780243,
      "eval_loss": 0.1490197479724884,
      "eval_runtime": 133.187,
      "eval_samples_per_second": 20.46,
      "eval_steps_per_second": 10.234,
      "num_input_tokens_seen": 19735024,
      "step": 10400
    },
    {
      "epoch": 1.6974467737988417,
      "grad_norm": 1.9754207134246826,
      "learning_rate": 4.210795149757375e-05,
      "loss": 0.1682,
      "num_input_tokens_seen": 19743360,
      "step": 10405
    },
    {
      "epoch": 1.698262501019659,
      "grad_norm": 0.15372611582279205,
      "learning_rate": 4.210079143490065e-05,
      "loss": 0.3483,
      "num_input_tokens_seen": 19751760,
      "step": 10410
    },
    {
      "epoch": 1.6990782282404764,
      "grad_norm": 0.09468267112970352,
      "learning_rate": 4.2093628735070604e-05,
      "loss": 0.0624,
      "num_input_tokens_seen": 19760416,
      "step": 10415
    },
    {
      "epoch": 1.6998939554612937,
      "grad_norm": 0.4475596845149994,
      "learning_rate": 4.208646339918819e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 19769808,
      "step": 10420
    },
    {
      "epoch": 1.700709682682111,
      "grad_norm": 0.5174076557159424,
      "learning_rate": 4.2079295428358414e-05,
      "loss": 0.1083,
      "num_input_tokens_seen": 19779200,
      "step": 10425
    },
    {
      "epoch": 1.7015254099029284,
      "grad_norm": 0.23780672252178192,
      "learning_rate": 4.207212482368664e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 19789712,
      "step": 10430
    },
    {
      "epoch": 1.7023411371237458,
      "grad_norm": 0.3810199499130249,
      "learning_rate": 4.206495158627867e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 19799808,
      "step": 10435
    },
    {
      "epoch": 1.7031568643445631,
      "grad_norm": 1.02205228805542,
      "learning_rate": 4.205777571724073e-05,
      "loss": 0.0743,
      "num_input_tokens_seen": 19809760,
      "step": 10440
    },
    {
      "epoch": 1.7039725915653805,
      "grad_norm": 0.5876002907752991,
      "learning_rate": 4.20505972176794e-05,
      "loss": 0.1127,
      "num_input_tokens_seen": 19819200,
      "step": 10445
    },
    {
      "epoch": 1.7047883187861979,
      "grad_norm": 2.146831750869751,
      "learning_rate": 4.204341608870171e-05,
      "loss": 0.2205,
      "num_input_tokens_seen": 19829744,
      "step": 10450
    },
    {
      "epoch": 1.7056040460070152,
      "grad_norm": 0.1984856128692627,
      "learning_rate": 4.203623233141508e-05,
      "loss": 0.1128,
      "num_input_tokens_seen": 19839504,
      "step": 10455
    },
    {
      "epoch": 1.7064197732278326,
      "grad_norm": 0.4034910500049591,
      "learning_rate": 4.2029045946927334e-05,
      "loss": 0.2672,
      "num_input_tokens_seen": 19848176,
      "step": 10460
    },
    {
      "epoch": 1.70723550044865,
      "grad_norm": 0.2905881106853485,
      "learning_rate": 4.20218569363467e-05,
      "loss": 0.1008,
      "num_input_tokens_seen": 19856288,
      "step": 10465
    },
    {
      "epoch": 1.7080512276694675,
      "grad_norm": 0.056312497705221176,
      "learning_rate": 4.2014665300781834e-05,
      "loss": 0.1963,
      "num_input_tokens_seen": 19865808,
      "step": 10470
    },
    {
      "epoch": 1.7088669548902846,
      "grad_norm": 0.3845035135746002,
      "learning_rate": 4.200747104134174e-05,
      "loss": 0.131,
      "num_input_tokens_seen": 19874128,
      "step": 10475
    },
    {
      "epoch": 1.7096826821111022,
      "grad_norm": 0.3887147605419159,
      "learning_rate": 4.200027415913588e-05,
      "loss": 0.1656,
      "num_input_tokens_seen": 19882864,
      "step": 10480
    },
    {
      "epoch": 1.7104984093319193,
      "grad_norm": 1.0988065004348755,
      "learning_rate": 4.1993074655274126e-05,
      "loss": 0.365,
      "num_input_tokens_seen": 19892208,
      "step": 10485
    },
    {
      "epoch": 1.711314136552737,
      "grad_norm": 0.7104158997535706,
      "learning_rate": 4.198587253086669e-05,
      "loss": 0.1745,
      "num_input_tokens_seen": 19901856,
      "step": 10490
    },
    {
      "epoch": 1.712129863773554,
      "grad_norm": 0.07464124262332916,
      "learning_rate": 4.197866778702426e-05,
      "loss": 0.1827,
      "num_input_tokens_seen": 19911776,
      "step": 10495
    },
    {
      "epoch": 1.7129455909943716,
      "grad_norm": 0.2885219156742096,
      "learning_rate": 4.197146042485789e-05,
      "loss": 0.0865,
      "num_input_tokens_seen": 19921088,
      "step": 10500
    },
    {
      "epoch": 1.7137613182151887,
      "grad_norm": 0.22198447585105896,
      "learning_rate": 4.1964250445479046e-05,
      "loss": 0.1942,
      "num_input_tokens_seen": 19929840,
      "step": 10505
    },
    {
      "epoch": 1.7145770454360063,
      "grad_norm": 0.8723790049552917,
      "learning_rate": 4.19570378499996e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 19939168,
      "step": 10510
    },
    {
      "epoch": 1.7153927726568234,
      "grad_norm": 0.8107682466506958,
      "learning_rate": 4.194982263953182e-05,
      "loss": 0.1413,
      "num_input_tokens_seen": 19948768,
      "step": 10515
    },
    {
      "epoch": 1.716208499877641,
      "grad_norm": 0.8251574039459229,
      "learning_rate": 4.194260481518838e-05,
      "loss": 0.3302,
      "num_input_tokens_seen": 19957616,
      "step": 10520
    },
    {
      "epoch": 1.7170242270984581,
      "grad_norm": 1.8366056680679321,
      "learning_rate": 4.1935384378082366e-05,
      "loss": 0.2056,
      "num_input_tokens_seen": 19967216,
      "step": 10525
    },
    {
      "epoch": 1.7178399543192757,
      "grad_norm": 1.3611857891082764,
      "learning_rate": 4.1928161329327267e-05,
      "loss": 0.1934,
      "num_input_tokens_seen": 19977472,
      "step": 10530
    },
    {
      "epoch": 1.7186556815400928,
      "grad_norm": 0.20907758176326752,
      "learning_rate": 4.1920935670036945e-05,
      "loss": 0.1087,
      "num_input_tokens_seen": 19987904,
      "step": 10535
    },
    {
      "epoch": 1.7194714087609104,
      "grad_norm": 0.27140146493911743,
      "learning_rate": 4.1913707401325705e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 19997632,
      "step": 10540
    },
    {
      "epoch": 1.7202871359817276,
      "grad_norm": 0.16559520363807678,
      "learning_rate": 4.1906476524308235e-05,
      "loss": 0.1761,
      "num_input_tokens_seen": 20008112,
      "step": 10545
    },
    {
      "epoch": 1.7211028632025451,
      "grad_norm": 0.34841158986091614,
      "learning_rate": 4.189924304009962e-05,
      "loss": 0.2686,
      "num_input_tokens_seen": 20017504,
      "step": 10550
    },
    {
      "epoch": 1.7219185904233625,
      "grad_norm": 0.7411243915557861,
      "learning_rate": 4.189200694981537e-05,
      "loss": 0.181,
      "num_input_tokens_seen": 20026288,
      "step": 10555
    },
    {
      "epoch": 1.7227343176441798,
      "grad_norm": 0.3348053991794586,
      "learning_rate": 4.188476825457136e-05,
      "loss": 0.0457,
      "num_input_tokens_seen": 20035968,
      "step": 10560
    },
    {
      "epoch": 1.7235500448649972,
      "grad_norm": 0.4819759130477905,
      "learning_rate": 4.18775269554839e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 20044848,
      "step": 10565
    },
    {
      "epoch": 1.7243657720858145,
      "grad_norm": 0.5932174921035767,
      "learning_rate": 4.187028305366969e-05,
      "loss": 0.1107,
      "num_input_tokens_seen": 20052288,
      "step": 10570
    },
    {
      "epoch": 1.725181499306632,
      "grad_norm": 0.7227306365966797,
      "learning_rate": 4.1863036550245824e-05,
      "loss": 0.1512,
      "num_input_tokens_seen": 20061728,
      "step": 10575
    },
    {
      "epoch": 1.7259972265274492,
      "grad_norm": 1.3508647680282593,
      "learning_rate": 4.1855787446329806e-05,
      "loss": 0.2217,
      "num_input_tokens_seen": 20071680,
      "step": 10580
    },
    {
      "epoch": 1.7268129537482666,
      "grad_norm": 0.9116230010986328,
      "learning_rate": 4.184853574303955e-05,
      "loss": 0.3619,
      "num_input_tokens_seen": 20081456,
      "step": 10585
    },
    {
      "epoch": 1.727628680969084,
      "grad_norm": 0.1878964751958847,
      "learning_rate": 4.184128144149334e-05,
      "loss": 0.0857,
      "num_input_tokens_seen": 20090512,
      "step": 10590
    },
    {
      "epoch": 1.7284444081899013,
      "grad_norm": 0.3664260506629944,
      "learning_rate": 4.1834024542809896e-05,
      "loss": 0.1014,
      "num_input_tokens_seen": 20099552,
      "step": 10595
    },
    {
      "epoch": 1.7292601354107187,
      "grad_norm": 0.6272090673446655,
      "learning_rate": 4.1826765048108315e-05,
      "loss": 0.238,
      "num_input_tokens_seen": 20108768,
      "step": 10600
    },
    {
      "epoch": 1.7292601354107187,
      "eval_loss": 0.14931465685367584,
      "eval_runtime": 133.3566,
      "eval_samples_per_second": 20.434,
      "eval_steps_per_second": 10.221,
      "num_input_tokens_seen": 20108768,
      "step": 10600
    },
    {
      "epoch": 1.730075862631536,
      "grad_norm": 0.2470075488090515,
      "learning_rate": 4.181950295850811e-05,
      "loss": 0.2176,
      "num_input_tokens_seen": 20117600,
      "step": 10605
    },
    {
      "epoch": 1.7308915898523534,
      "grad_norm": 0.3148192763328552,
      "learning_rate": 4.181223827512918e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 20127632,
      "step": 10610
    },
    {
      "epoch": 1.7317073170731707,
      "grad_norm": 0.6537553071975708,
      "learning_rate": 4.180497099909183e-05,
      "loss": 0.1867,
      "num_input_tokens_seen": 20135632,
      "step": 10615
    },
    {
      "epoch": 1.732523044293988,
      "grad_norm": 0.6319558620452881,
      "learning_rate": 4.179770113151677e-05,
      "loss": 0.1157,
      "num_input_tokens_seen": 20144096,
      "step": 10620
    },
    {
      "epoch": 1.7333387715148054,
      "grad_norm": 0.4934304356575012,
      "learning_rate": 4.179042867352511e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 20154288,
      "step": 10625
    },
    {
      "epoch": 1.7341544987356228,
      "grad_norm": 0.2862880527973175,
      "learning_rate": 4.1783153626238334e-05,
      "loss": 0.1818,
      "num_input_tokens_seen": 20165824,
      "step": 10630
    },
    {
      "epoch": 1.7349702259564401,
      "grad_norm": 0.9960951209068298,
      "learning_rate": 4.177587599077836e-05,
      "loss": 0.1234,
      "num_input_tokens_seen": 20176384,
      "step": 10635
    },
    {
      "epoch": 1.7357859531772575,
      "grad_norm": 0.15226903557777405,
      "learning_rate": 4.1768595768267494e-05,
      "loss": 0.1157,
      "num_input_tokens_seen": 20185664,
      "step": 10640
    },
    {
      "epoch": 1.7366016803980748,
      "grad_norm": 0.46054112911224365,
      "learning_rate": 4.176131295982843e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 20195344,
      "step": 10645
    },
    {
      "epoch": 1.7374174076188922,
      "grad_norm": 1.4196138381958008,
      "learning_rate": 4.1754027566584276e-05,
      "loss": 0.3133,
      "num_input_tokens_seen": 20204112,
      "step": 10650
    },
    {
      "epoch": 1.7382331348397098,
      "grad_norm": 1.2795538902282715,
      "learning_rate": 4.174673958965852e-05,
      "loss": 0.2137,
      "num_input_tokens_seen": 20213472,
      "step": 10655
    },
    {
      "epoch": 1.739048862060527,
      "grad_norm": 0.3008710443973541,
      "learning_rate": 4.173944903017507e-05,
      "loss": 0.2041,
      "num_input_tokens_seen": 20223280,
      "step": 10660
    },
    {
      "epoch": 1.7398645892813445,
      "grad_norm": 1.8269128799438477,
      "learning_rate": 4.173215588925822e-05,
      "loss": 0.1738,
      "num_input_tokens_seen": 20231696,
      "step": 10665
    },
    {
      "epoch": 1.7406803165021616,
      "grad_norm": 0.9538857340812683,
      "learning_rate": 4.172486016803266e-05,
      "loss": 0.0707,
      "num_input_tokens_seen": 20240704,
      "step": 10670
    },
    {
      "epoch": 1.7414960437229792,
      "grad_norm": 0.8727644681930542,
      "learning_rate": 4.171756186762349e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 20249792,
      "step": 10675
    },
    {
      "epoch": 1.7423117709437963,
      "grad_norm": 0.3607631325721741,
      "learning_rate": 4.171026098915619e-05,
      "loss": 0.2704,
      "num_input_tokens_seen": 20259120,
      "step": 10680
    },
    {
      "epoch": 1.7431274981646139,
      "grad_norm": 1.6060717105865479,
      "learning_rate": 4.170295753375665e-05,
      "loss": 0.2772,
      "num_input_tokens_seen": 20268288,
      "step": 10685
    },
    {
      "epoch": 1.743943225385431,
      "grad_norm": 1.6714982986450195,
      "learning_rate": 4.169565150255117e-05,
      "loss": 0.1984,
      "num_input_tokens_seen": 20277024,
      "step": 10690
    },
    {
      "epoch": 1.7447589526062486,
      "grad_norm": 0.9225578308105469,
      "learning_rate": 4.16883428966664e-05,
      "loss": 0.0996,
      "num_input_tokens_seen": 20286560,
      "step": 10695
    },
    {
      "epoch": 1.7455746798270657,
      "grad_norm": 0.31928718090057373,
      "learning_rate": 4.168103171722944e-05,
      "loss": 0.0843,
      "num_input_tokens_seen": 20296688,
      "step": 10700
    },
    {
      "epoch": 1.7463904070478833,
      "grad_norm": 0.6330808401107788,
      "learning_rate": 4.167371796536777e-05,
      "loss": 0.0904,
      "num_input_tokens_seen": 20307024,
      "step": 10705
    },
    {
      "epoch": 1.7472061342687004,
      "grad_norm": 0.20690128207206726,
      "learning_rate": 4.166640164220924e-05,
      "loss": 0.1828,
      "num_input_tokens_seen": 20316336,
      "step": 10710
    },
    {
      "epoch": 1.748021861489518,
      "grad_norm": 0.4073238968849182,
      "learning_rate": 4.1659082748882144e-05,
      "loss": 0.0253,
      "num_input_tokens_seen": 20325280,
      "step": 10715
    },
    {
      "epoch": 1.7488375887103351,
      "grad_norm": 0.9167855978012085,
      "learning_rate": 4.1651761286515135e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 20335392,
      "step": 10720
    },
    {
      "epoch": 1.7496533159311527,
      "grad_norm": 0.21964780986309052,
      "learning_rate": 4.164443725623728e-05,
      "loss": 0.1428,
      "num_input_tokens_seen": 20343936,
      "step": 10725
    },
    {
      "epoch": 1.7504690431519698,
      "grad_norm": 1.7121567726135254,
      "learning_rate": 4.163711065917802e-05,
      "loss": 0.1879,
      "num_input_tokens_seen": 20352704,
      "step": 10730
    },
    {
      "epoch": 1.7512847703727874,
      "grad_norm": 0.47327572107315063,
      "learning_rate": 4.1629781496467234e-05,
      "loss": 0.0928,
      "num_input_tokens_seen": 20361696,
      "step": 10735
    },
    {
      "epoch": 1.7521004975936045,
      "grad_norm": 0.25349563360214233,
      "learning_rate": 4.1622449769235164e-05,
      "loss": 0.0841,
      "num_input_tokens_seen": 20371248,
      "step": 10740
    },
    {
      "epoch": 1.752916224814422,
      "grad_norm": 0.19022178649902344,
      "learning_rate": 4.161511547861243e-05,
      "loss": 0.186,
      "num_input_tokens_seen": 20381376,
      "step": 10745
    },
    {
      "epoch": 1.7537319520352395,
      "grad_norm": 0.5431305170059204,
      "learning_rate": 4.1607778625730104e-05,
      "loss": 0.1181,
      "num_input_tokens_seen": 20390688,
      "step": 10750
    },
    {
      "epoch": 1.7545476792560568,
      "grad_norm": 0.8762412667274475,
      "learning_rate": 4.160043921171961e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 20399504,
      "step": 10755
    },
    {
      "epoch": 1.7553634064768742,
      "grad_norm": 0.42410585284233093,
      "learning_rate": 4.159309723771276e-05,
      "loss": 0.1233,
      "num_input_tokens_seen": 20410240,
      "step": 10760
    },
    {
      "epoch": 1.7561791336976915,
      "grad_norm": 0.37146660685539246,
      "learning_rate": 4.158575270484181e-05,
      "loss": 0.1364,
      "num_input_tokens_seen": 20420688,
      "step": 10765
    },
    {
      "epoch": 1.7569948609185089,
      "grad_norm": 1.106939673423767,
      "learning_rate": 4.157840561423936e-05,
      "loss": 0.1247,
      "num_input_tokens_seen": 20431856,
      "step": 10770
    },
    {
      "epoch": 1.7578105881393262,
      "grad_norm": 0.7576045393943787,
      "learning_rate": 4.1571055967038416e-05,
      "loss": 0.102,
      "num_input_tokens_seen": 20441168,
      "step": 10775
    },
    {
      "epoch": 1.7586263153601436,
      "grad_norm": 0.5191263556480408,
      "learning_rate": 4.156370376437241e-05,
      "loss": 0.0469,
      "num_input_tokens_seen": 20450336,
      "step": 10780
    },
    {
      "epoch": 1.759442042580961,
      "grad_norm": 0.1781008541584015,
      "learning_rate": 4.155634900737513e-05,
      "loss": 0.2258,
      "num_input_tokens_seen": 20460496,
      "step": 10785
    },
    {
      "epoch": 1.7602577698017783,
      "grad_norm": 0.7258459329605103,
      "learning_rate": 4.1548991697180764e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 20470032,
      "step": 10790
    },
    {
      "epoch": 1.7610734970225956,
      "grad_norm": 1.5235613584518433,
      "learning_rate": 4.1541631834923914e-05,
      "loss": 0.2521,
      "num_input_tokens_seen": 20480208,
      "step": 10795
    },
    {
      "epoch": 1.761889224243413,
      "grad_norm": 1.3083686828613281,
      "learning_rate": 4.153426942173956e-05,
      "loss": 0.276,
      "num_input_tokens_seen": 20489424,
      "step": 10800
    },
    {
      "epoch": 1.761889224243413,
      "eval_loss": 0.14713002741336823,
      "eval_runtime": 133.2293,
      "eval_samples_per_second": 20.453,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 20489424,
      "step": 10800
    },
    {
      "epoch": 1.7627049514642303,
      "grad_norm": 0.8450349569320679,
      "learning_rate": 4.152690445876308e-05,
      "loss": 0.2642,
      "num_input_tokens_seen": 20498464,
      "step": 10805
    },
    {
      "epoch": 1.7635206786850477,
      "grad_norm": 0.5372847318649292,
      "learning_rate": 4.1519536947130245e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 20507328,
      "step": 10810
    },
    {
      "epoch": 1.764336405905865,
      "grad_norm": 0.09897629171609879,
      "learning_rate": 4.151216688797722e-05,
      "loss": 0.1097,
      "num_input_tokens_seen": 20516704,
      "step": 10815
    },
    {
      "epoch": 1.7651521331266824,
      "grad_norm": 0.9360846877098083,
      "learning_rate": 4.150479428244054e-05,
      "loss": 0.1811,
      "num_input_tokens_seen": 20527296,
      "step": 10820
    },
    {
      "epoch": 1.7659678603474998,
      "grad_norm": 0.5590841770172119,
      "learning_rate": 4.1497419131657176e-05,
      "loss": 0.0509,
      "num_input_tokens_seen": 20536384,
      "step": 10825
    },
    {
      "epoch": 1.766783587568317,
      "grad_norm": 0.5867722034454346,
      "learning_rate": 4.149004143676447e-05,
      "loss": 0.0886,
      "num_input_tokens_seen": 20547536,
      "step": 10830
    },
    {
      "epoch": 1.7675993147891345,
      "grad_norm": 0.1341889351606369,
      "learning_rate": 4.148266119890015e-05,
      "loss": 0.1273,
      "num_input_tokens_seen": 20555680,
      "step": 10835
    },
    {
      "epoch": 1.768415042009952,
      "grad_norm": 0.17609836161136627,
      "learning_rate": 4.1475278419202324e-05,
      "loss": 0.1003,
      "num_input_tokens_seen": 20563824,
      "step": 10840
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 0.19718626141548157,
      "learning_rate": 4.146789309880953e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 20573840,
      "step": 10845
    },
    {
      "epoch": 1.7700464964515867,
      "grad_norm": 0.6379894018173218,
      "learning_rate": 4.146050523886068e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 20582480,
      "step": 10850
    },
    {
      "epoch": 1.7708622236724039,
      "grad_norm": 0.06384234130382538,
      "learning_rate": 4.1453114840495055e-05,
      "loss": 0.2447,
      "num_input_tokens_seen": 20592640,
      "step": 10855
    },
    {
      "epoch": 1.7716779508932214,
      "grad_norm": 0.2788175940513611,
      "learning_rate": 4.1445721904852364e-05,
      "loss": 0.1816,
      "num_input_tokens_seen": 20602384,
      "step": 10860
    },
    {
      "epoch": 1.7724936781140386,
      "grad_norm": 0.47166043519973755,
      "learning_rate": 4.143832643307269e-05,
      "loss": 0.09,
      "num_input_tokens_seen": 20610976,
      "step": 10865
    },
    {
      "epoch": 1.7733094053348561,
      "grad_norm": 0.46872085332870483,
      "learning_rate": 4.1430928426296503e-05,
      "loss": 0.1281,
      "num_input_tokens_seen": 20620752,
      "step": 10870
    },
    {
      "epoch": 1.7741251325556733,
      "grad_norm": 0.03537541627883911,
      "learning_rate": 4.142352788566466e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 20630016,
      "step": 10875
    },
    {
      "epoch": 1.7749408597764909,
      "grad_norm": 0.26459869742393494,
      "learning_rate": 4.1416124812318424e-05,
      "loss": 0.107,
      "num_input_tokens_seen": 20640896,
      "step": 10880
    },
    {
      "epoch": 1.775756586997308,
      "grad_norm": 0.15815170109272003,
      "learning_rate": 4.1408719207399453e-05,
      "loss": 0.1132,
      "num_input_tokens_seen": 20651040,
      "step": 10885
    },
    {
      "epoch": 1.7765723142181256,
      "grad_norm": 0.2396436184644699,
      "learning_rate": 4.140131107204978e-05,
      "loss": 0.1049,
      "num_input_tokens_seen": 20660192,
      "step": 10890
    },
    {
      "epoch": 1.7773880414389427,
      "grad_norm": 0.5938466787338257,
      "learning_rate": 4.139390040741182e-05,
      "loss": 0.1387,
      "num_input_tokens_seen": 20669728,
      "step": 10895
    },
    {
      "epoch": 1.7782037686597603,
      "grad_norm": 0.5671244263648987,
      "learning_rate": 4.1386487214628396e-05,
      "loss": 0.1933,
      "num_input_tokens_seen": 20679920,
      "step": 10900
    },
    {
      "epoch": 1.7790194958805774,
      "grad_norm": 0.8705153465270996,
      "learning_rate": 4.137907149484272e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 20689040,
      "step": 10905
    },
    {
      "epoch": 1.779835223101395,
      "grad_norm": 1.0119428634643555,
      "learning_rate": 4.137165324919839e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 20698496,
      "step": 10910
    },
    {
      "epoch": 1.780650950322212,
      "grad_norm": 0.1698477417230606,
      "learning_rate": 4.136423247883939e-05,
      "loss": 0.0699,
      "num_input_tokens_seen": 20707440,
      "step": 10915
    },
    {
      "epoch": 1.7814666775430297,
      "grad_norm": 1.0312727689743042,
      "learning_rate": 4.135680918491009e-05,
      "loss": 0.1352,
      "num_input_tokens_seen": 20717008,
      "step": 10920
    },
    {
      "epoch": 1.7822824047638468,
      "grad_norm": 0.4837890863418579,
      "learning_rate": 4.1349383368555265e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 20726480,
      "step": 10925
    },
    {
      "epoch": 1.7830981319846644,
      "grad_norm": 0.05641445517539978,
      "learning_rate": 4.1341955030920065e-05,
      "loss": 0.0835,
      "num_input_tokens_seen": 20735472,
      "step": 10930
    },
    {
      "epoch": 1.7839138592054817,
      "grad_norm": 0.1869477927684784,
      "learning_rate": 4.1334524173150036e-05,
      "loss": 0.0452,
      "num_input_tokens_seen": 20744720,
      "step": 10935
    },
    {
      "epoch": 1.784729586426299,
      "grad_norm": 0.6498836278915405,
      "learning_rate": 4.13270907963911e-05,
      "loss": 0.1757,
      "num_input_tokens_seen": 20753600,
      "step": 10940
    },
    {
      "epoch": 1.7855453136471164,
      "grad_norm": 0.0932241901755333,
      "learning_rate": 4.131965490178959e-05,
      "loss": 0.0826,
      "num_input_tokens_seen": 20763024,
      "step": 10945
    },
    {
      "epoch": 1.7863610408679338,
      "grad_norm": 0.3776271939277649,
      "learning_rate": 4.131221649049222e-05,
      "loss": 0.2115,
      "num_input_tokens_seen": 20773120,
      "step": 10950
    },
    {
      "epoch": 1.7871767680887511,
      "grad_norm": 0.2512197494506836,
      "learning_rate": 4.130477556364606e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 20781984,
      "step": 10955
    },
    {
      "epoch": 1.7879924953095685,
      "grad_norm": 0.8879039287567139,
      "learning_rate": 4.129733212239861e-05,
      "loss": 0.0928,
      "num_input_tokens_seen": 20790688,
      "step": 10960
    },
    {
      "epoch": 1.7888082225303858,
      "grad_norm": 1.5104135274887085,
      "learning_rate": 4.128988616789774e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 20801008,
      "step": 10965
    },
    {
      "epoch": 1.7896239497512032,
      "grad_norm": 0.9439749717712402,
      "learning_rate": 4.1282437701291724e-05,
      "loss": 0.1082,
      "num_input_tokens_seen": 20811328,
      "step": 10970
    },
    {
      "epoch": 1.7904396769720206,
      "grad_norm": 0.31335532665252686,
      "learning_rate": 4.1274986723729184e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 20821376,
      "step": 10975
    },
    {
      "epoch": 1.791255404192838,
      "grad_norm": 1.2726085186004639,
      "learning_rate": 4.126753323635917e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 20831936,
      "step": 10980
    },
    {
      "epoch": 1.7920711314136553,
      "grad_norm": 0.3384057581424713,
      "learning_rate": 4.12600772403311e-05,
      "loss": 0.1179,
      "num_input_tokens_seen": 20842208,
      "step": 10985
    },
    {
      "epoch": 1.7928868586344726,
      "grad_norm": 0.24463185667991638,
      "learning_rate": 4.125261873679479e-05,
      "loss": 0.0545,
      "num_input_tokens_seen": 20852288,
      "step": 10990
    },
    {
      "epoch": 1.79370258585529,
      "grad_norm": 0.6875919103622437,
      "learning_rate": 4.124515772690042e-05,
      "loss": 0.214,
      "num_input_tokens_seen": 20861536,
      "step": 10995
    },
    {
      "epoch": 1.7945183130761073,
      "grad_norm": 0.10059333592653275,
      "learning_rate": 4.123769421179858e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 20870832,
      "step": 11000
    },
    {
      "epoch": 1.7945183130761073,
      "eval_loss": 0.150941401720047,
      "eval_runtime": 133.3017,
      "eval_samples_per_second": 20.442,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 20870832,
      "step": 11000
    },
    {
      "epoch": 1.7953340402969247,
      "grad_norm": 1.0939390659332275,
      "learning_rate": 4.1230228192640236e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 20879216,
      "step": 11005
    },
    {
      "epoch": 1.796149767517742,
      "grad_norm": 1.1364253759384155,
      "learning_rate": 4.122275967057675e-05,
      "loss": 0.1692,
      "num_input_tokens_seen": 20888320,
      "step": 11010
    },
    {
      "epoch": 1.7969654947385594,
      "grad_norm": 0.23759251832962036,
      "learning_rate": 4.1215288646759846e-05,
      "loss": 0.1141,
      "num_input_tokens_seen": 20898064,
      "step": 11015
    },
    {
      "epoch": 1.7977812219593767,
      "grad_norm": 0.4529772102832794,
      "learning_rate": 4.120781512234166e-05,
      "loss": 0.2745,
      "num_input_tokens_seen": 20906784,
      "step": 11020
    },
    {
      "epoch": 1.7985969491801943,
      "grad_norm": 0.9014625549316406,
      "learning_rate": 4.120033909847471e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 20916736,
      "step": 11025
    },
    {
      "epoch": 1.7994126764010114,
      "grad_norm": 1.2991846799850464,
      "learning_rate": 4.119286057631187e-05,
      "loss": 0.1689,
      "num_input_tokens_seen": 20926112,
      "step": 11030
    },
    {
      "epoch": 1.800228403621829,
      "grad_norm": 3.5448355674743652,
      "learning_rate": 4.118537955700646e-05,
      "loss": 0.1154,
      "num_input_tokens_seen": 20937104,
      "step": 11035
    },
    {
      "epoch": 1.8010441308426461,
      "grad_norm": 0.5988321900367737,
      "learning_rate": 4.11778960417121e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 20947040,
      "step": 11040
    },
    {
      "epoch": 1.8018598580634637,
      "grad_norm": 1.595685362815857,
      "learning_rate": 4.117041003158288e-05,
      "loss": 0.1688,
      "num_input_tokens_seen": 20956096,
      "step": 11045
    },
    {
      "epoch": 1.8026755852842808,
      "grad_norm": 0.4753856062889099,
      "learning_rate": 4.1162921527773215e-05,
      "loss": 0.0559,
      "num_input_tokens_seen": 20965456,
      "step": 11050
    },
    {
      "epoch": 1.8034913125050984,
      "grad_norm": 0.43115344643592834,
      "learning_rate": 4.115543053143794e-05,
      "loss": 0.1433,
      "num_input_tokens_seen": 20974688,
      "step": 11055
    },
    {
      "epoch": 1.8043070397259156,
      "grad_norm": 0.15722495317459106,
      "learning_rate": 4.114793704373226e-05,
      "loss": 0.0901,
      "num_input_tokens_seen": 20984464,
      "step": 11060
    },
    {
      "epoch": 1.8051227669467331,
      "grad_norm": 0.16579017043113708,
      "learning_rate": 4.114044106581175e-05,
      "loss": 0.0998,
      "num_input_tokens_seen": 20992928,
      "step": 11065
    },
    {
      "epoch": 1.8059384941675503,
      "grad_norm": 0.10362575948238373,
      "learning_rate": 4.11329425988324e-05,
      "loss": 0.0531,
      "num_input_tokens_seen": 21002208,
      "step": 11070
    },
    {
      "epoch": 1.8067542213883678,
      "grad_norm": 0.5644795298576355,
      "learning_rate": 4.112544164395056e-05,
      "loss": 0.1987,
      "num_input_tokens_seen": 21011808,
      "step": 11075
    },
    {
      "epoch": 1.807569948609185,
      "grad_norm": 0.5403379797935486,
      "learning_rate": 4.111793820232297e-05,
      "loss": 0.0265,
      "num_input_tokens_seen": 21021360,
      "step": 11080
    },
    {
      "epoch": 1.8083856758300025,
      "grad_norm": 0.8461185693740845,
      "learning_rate": 4.1110432275106767e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 21030032,
      "step": 11085
    },
    {
      "epoch": 1.8092014030508197,
      "grad_norm": 0.14606541395187378,
      "learning_rate": 4.110292386345944e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 21039520,
      "step": 11090
    },
    {
      "epoch": 1.8100171302716372,
      "grad_norm": 0.17347830533981323,
      "learning_rate": 4.109541296853891e-05,
      "loss": 0.1092,
      "num_input_tokens_seen": 21048128,
      "step": 11095
    },
    {
      "epoch": 1.8108328574924544,
      "grad_norm": 0.8031986355781555,
      "learning_rate": 4.108789959150341e-05,
      "loss": 0.0728,
      "num_input_tokens_seen": 21058096,
      "step": 11100
    },
    {
      "epoch": 1.811648584713272,
      "grad_norm": 0.6355143785476685,
      "learning_rate": 4.108038373351163e-05,
      "loss": 0.2794,
      "num_input_tokens_seen": 21065968,
      "step": 11105
    },
    {
      "epoch": 1.812464311934089,
      "grad_norm": 0.6886758804321289,
      "learning_rate": 4.10728653957226e-05,
      "loss": 0.2114,
      "num_input_tokens_seen": 21074448,
      "step": 11110
    },
    {
      "epoch": 1.8132800391549067,
      "grad_norm": 0.2560984790325165,
      "learning_rate": 4.106534457929575e-05,
      "loss": 0.0929,
      "num_input_tokens_seen": 21083312,
      "step": 11115
    },
    {
      "epoch": 1.814095766375724,
      "grad_norm": 0.20294424891471863,
      "learning_rate": 4.105782128539086e-05,
      "loss": 0.1107,
      "num_input_tokens_seen": 21091376,
      "step": 11120
    },
    {
      "epoch": 1.8149114935965414,
      "grad_norm": 0.2982413172721863,
      "learning_rate": 4.1050295515168144e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 21100192,
      "step": 11125
    },
    {
      "epoch": 1.8157272208173587,
      "grad_norm": 0.40975290536880493,
      "learning_rate": 4.1042767269788155e-05,
      "loss": 0.1883,
      "num_input_tokens_seen": 21109056,
      "step": 11130
    },
    {
      "epoch": 1.816542948038176,
      "grad_norm": 1.174809217453003,
      "learning_rate": 4.103523655041185e-05,
      "loss": 0.1619,
      "num_input_tokens_seen": 21117392,
      "step": 11135
    },
    {
      "epoch": 1.8173586752589934,
      "grad_norm": 0.19011132419109344,
      "learning_rate": 4.102770335820055e-05,
      "loss": 0.3158,
      "num_input_tokens_seen": 21126112,
      "step": 11140
    },
    {
      "epoch": 1.8181744024798108,
      "grad_norm": 0.32907795906066895,
      "learning_rate": 4.1020167694315984e-05,
      "loss": 0.0738,
      "num_input_tokens_seen": 21135008,
      "step": 11145
    },
    {
      "epoch": 1.8189901297006281,
      "grad_norm": 0.5854259729385376,
      "learning_rate": 4.101262955992023e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 21143216,
      "step": 11150
    },
    {
      "epoch": 1.8198058569214455,
      "grad_norm": 0.2176334261894226,
      "learning_rate": 4.100508895617578e-05,
      "loss": 0.0496,
      "num_input_tokens_seen": 21152464,
      "step": 11155
    },
    {
      "epoch": 1.8206215841422628,
      "grad_norm": 0.4404340088367462,
      "learning_rate": 4.099754588424547e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 21162352,
      "step": 11160
    },
    {
      "epoch": 1.8214373113630802,
      "grad_norm": 0.049887895584106445,
      "learning_rate": 4.0990000345292546e-05,
      "loss": 0.1132,
      "num_input_tokens_seen": 21172864,
      "step": 11165
    },
    {
      "epoch": 1.8222530385838975,
      "grad_norm": 1.686362385749817,
      "learning_rate": 4.098245234048064e-05,
      "loss": 0.2669,
      "num_input_tokens_seen": 21182208,
      "step": 11170
    },
    {
      "epoch": 1.8230687658047149,
      "grad_norm": 1.1982375383377075,
      "learning_rate": 4.0974901870973726e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 21192128,
      "step": 11175
    },
    {
      "epoch": 1.8238844930255322,
      "grad_norm": 0.6244966387748718,
      "learning_rate": 4.096734893793619e-05,
      "loss": 0.3029,
      "num_input_tokens_seen": 21201248,
      "step": 11180
    },
    {
      "epoch": 1.8247002202463496,
      "grad_norm": 1.5777251720428467,
      "learning_rate": 4.095979354253279e-05,
      "loss": 0.204,
      "num_input_tokens_seen": 21211216,
      "step": 11185
    },
    {
      "epoch": 1.825515947467167,
      "grad_norm": 0.3634169101715088,
      "learning_rate": 4.0952235685928656e-05,
      "loss": 0.0534,
      "num_input_tokens_seen": 21221168,
      "step": 11190
    },
    {
      "epoch": 1.8263316746879843,
      "grad_norm": 0.28046277165412903,
      "learning_rate": 4.094467536928932e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 21231472,
      "step": 11195
    },
    {
      "epoch": 1.8271474019088016,
      "grad_norm": 0.17639657855033875,
      "learning_rate": 4.093711259378067e-05,
      "loss": 0.1532,
      "num_input_tokens_seen": 21240960,
      "step": 11200
    },
    {
      "epoch": 1.8271474019088016,
      "eval_loss": 0.15010398626327515,
      "eval_runtime": 133.2947,
      "eval_samples_per_second": 20.443,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 21240960,
      "step": 11200
    },
    {
      "epoch": 1.827963129129619,
      "grad_norm": 0.4159243702888489,
      "learning_rate": 4.092954736056897e-05,
      "loss": 0.1813,
      "num_input_tokens_seen": 21250784,
      "step": 11205
    },
    {
      "epoch": 1.8287788563504366,
      "grad_norm": 0.5452386736869812,
      "learning_rate": 4.09219796708209e-05,
      "loss": 0.1111,
      "num_input_tokens_seen": 21260432,
      "step": 11210
    },
    {
      "epoch": 1.8295945835712537,
      "grad_norm": 0.329264760017395,
      "learning_rate": 4.0914409525703464e-05,
      "loss": 0.0915,
      "num_input_tokens_seen": 21268688,
      "step": 11215
    },
    {
      "epoch": 1.8304103107920713,
      "grad_norm": 0.05948200821876526,
      "learning_rate": 4.090683692638408e-05,
      "loss": 0.0425,
      "num_input_tokens_seen": 21277792,
      "step": 11220
    },
    {
      "epoch": 1.8312260380128884,
      "grad_norm": 0.5961433053016663,
      "learning_rate": 4.089926187403056e-05,
      "loss": 0.1745,
      "num_input_tokens_seen": 21287312,
      "step": 11225
    },
    {
      "epoch": 1.832041765233706,
      "grad_norm": 0.1069825142621994,
      "learning_rate": 4.0891684369811044e-05,
      "loss": 0.2905,
      "num_input_tokens_seen": 21295792,
      "step": 11230
    },
    {
      "epoch": 1.8328574924545231,
      "grad_norm": 0.5212803483009338,
      "learning_rate": 4.0884104414894107e-05,
      "loss": 0.1034,
      "num_input_tokens_seen": 21306064,
      "step": 11235
    },
    {
      "epoch": 1.8336732196753407,
      "grad_norm": 0.9101600646972656,
      "learning_rate": 4.087652201044864e-05,
      "loss": 0.1006,
      "num_input_tokens_seen": 21316080,
      "step": 11240
    },
    {
      "epoch": 1.8344889468961578,
      "grad_norm": 0.125034898519516,
      "learning_rate": 4.086893715764397e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 21325568,
      "step": 11245
    },
    {
      "epoch": 1.8353046741169754,
      "grad_norm": 0.7725772857666016,
      "learning_rate": 4.086134985764977e-05,
      "loss": 0.1704,
      "num_input_tokens_seen": 21335296,
      "step": 11250
    },
    {
      "epoch": 1.8361204013377925,
      "grad_norm": 0.17533046007156372,
      "learning_rate": 4.0853760111636085e-05,
      "loss": 0.1033,
      "num_input_tokens_seen": 21344720,
      "step": 11255
    },
    {
      "epoch": 1.83693612855861,
      "grad_norm": 0.47086620330810547,
      "learning_rate": 4.084616792077337e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 21354320,
      "step": 11260
    },
    {
      "epoch": 1.8377518557794272,
      "grad_norm": 0.05439858138561249,
      "learning_rate": 4.083857328623243e-05,
      "loss": 0.0996,
      "num_input_tokens_seen": 21363312,
      "step": 11265
    },
    {
      "epoch": 1.8385675830002448,
      "grad_norm": 0.03492393344640732,
      "learning_rate": 4.083097620918444e-05,
      "loss": 0.2494,
      "num_input_tokens_seen": 21373024,
      "step": 11270
    },
    {
      "epoch": 1.839383310221062,
      "grad_norm": 0.46162015199661255,
      "learning_rate": 4.082337669080097e-05,
      "loss": 0.1034,
      "num_input_tokens_seen": 21382192,
      "step": 11275
    },
    {
      "epoch": 1.8401990374418795,
      "grad_norm": 0.1278483271598816,
      "learning_rate": 4.081577473225398e-05,
      "loss": 0.2156,
      "num_input_tokens_seen": 21392416,
      "step": 11280
    },
    {
      "epoch": 1.8410147646626966,
      "grad_norm": 2.492196559906006,
      "learning_rate": 4.080817033471577e-05,
      "loss": 0.2506,
      "num_input_tokens_seen": 21400576,
      "step": 11285
    },
    {
      "epoch": 1.8418304918835142,
      "grad_norm": 0.8138436079025269,
      "learning_rate": 4.080056349935903e-05,
      "loss": 0.1853,
      "num_input_tokens_seen": 21410096,
      "step": 11290
    },
    {
      "epoch": 1.8426462191043314,
      "grad_norm": 0.3311351239681244,
      "learning_rate": 4.079295422735684e-05,
      "loss": 0.1741,
      "num_input_tokens_seen": 21418576,
      "step": 11295
    },
    {
      "epoch": 1.843461946325149,
      "grad_norm": 0.4613144099712372,
      "learning_rate": 4.078534251988264e-05,
      "loss": 0.113,
      "num_input_tokens_seen": 21428032,
      "step": 11300
    },
    {
      "epoch": 1.8442776735459663,
      "grad_norm": 0.18174828588962555,
      "learning_rate": 4.077772837811025e-05,
      "loss": 0.0854,
      "num_input_tokens_seen": 21436528,
      "step": 11305
    },
    {
      "epoch": 1.8450934007667836,
      "grad_norm": 1.2112890481948853,
      "learning_rate": 4.0770111803213874e-05,
      "loss": 0.2546,
      "num_input_tokens_seen": 21447072,
      "step": 11310
    },
    {
      "epoch": 1.845909127987601,
      "grad_norm": 0.5004761815071106,
      "learning_rate": 4.076249279636807e-05,
      "loss": 0.0647,
      "num_input_tokens_seen": 21456000,
      "step": 11315
    },
    {
      "epoch": 1.8467248552084183,
      "grad_norm": 0.13962674140930176,
      "learning_rate": 4.075487135874781e-05,
      "loss": 0.151,
      "num_input_tokens_seen": 21465888,
      "step": 11320
    },
    {
      "epoch": 1.8475405824292357,
      "grad_norm": 0.6733920574188232,
      "learning_rate": 4.074724749152837e-05,
      "loss": 0.0807,
      "num_input_tokens_seen": 21475616,
      "step": 11325
    },
    {
      "epoch": 1.848356309650053,
      "grad_norm": 0.5490584969520569,
      "learning_rate": 4.07396211958855e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 21485216,
      "step": 11330
    },
    {
      "epoch": 1.8491720368708704,
      "grad_norm": 0.3838418126106262,
      "learning_rate": 4.073199247299523e-05,
      "loss": 0.146,
      "num_input_tokens_seen": 21493824,
      "step": 11335
    },
    {
      "epoch": 1.8499877640916877,
      "grad_norm": 0.09403187036514282,
      "learning_rate": 4.072436132403403e-05,
      "loss": 0.0885,
      "num_input_tokens_seen": 21503552,
      "step": 11340
    },
    {
      "epoch": 1.850803491312505,
      "grad_norm": 0.29000434279441833,
      "learning_rate": 4.0716727750178704e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 21512544,
      "step": 11345
    },
    {
      "epoch": 1.8516192185333225,
      "grad_norm": 0.640967071056366,
      "learning_rate": 4.0709091752606455e-05,
      "loss": 0.2354,
      "num_input_tokens_seen": 21522128,
      "step": 11350
    },
    {
      "epoch": 1.8524349457541398,
      "grad_norm": 0.15910044312477112,
      "learning_rate": 4.070145333249484e-05,
      "loss": 0.1431,
      "num_input_tokens_seen": 21531920,
      "step": 11355
    },
    {
      "epoch": 1.8532506729749572,
      "grad_norm": 0.5346929430961609,
      "learning_rate": 4.069381249102181e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 21542016,
      "step": 11360
    },
    {
      "epoch": 1.8540664001957745,
      "grad_norm": 0.816213846206665,
      "learning_rate": 4.0686169229365665e-05,
      "loss": 0.1027,
      "num_input_tokens_seen": 21551808,
      "step": 11365
    },
    {
      "epoch": 1.8548821274165919,
      "grad_norm": 0.3126137852668762,
      "learning_rate": 4.067852354870511e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 21560944,
      "step": 11370
    },
    {
      "epoch": 1.8556978546374092,
      "grad_norm": 0.23667143285274506,
      "learning_rate": 4.067087545021919e-05,
      "loss": 0.0937,
      "num_input_tokens_seen": 21569664,
      "step": 11375
    },
    {
      "epoch": 1.8565135818582266,
      "grad_norm": 1.328459620475769,
      "learning_rate": 4.066322493508734e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 21579552,
      "step": 11380
    },
    {
      "epoch": 1.857329309079044,
      "grad_norm": 1.4422142505645752,
      "learning_rate": 4.065557200448937e-05,
      "loss": 0.2219,
      "num_input_tokens_seen": 21587424,
      "step": 11385
    },
    {
      "epoch": 1.8581450362998613,
      "grad_norm": 0.03237544000148773,
      "learning_rate": 4.064791665960546e-05,
      "loss": 0.3421,
      "num_input_tokens_seen": 21597296,
      "step": 11390
    },
    {
      "epoch": 1.8589607635206788,
      "grad_norm": 0.7242356538772583,
      "learning_rate": 4.064025890161615e-05,
      "loss": 0.1655,
      "num_input_tokens_seen": 21605904,
      "step": 11395
    },
    {
      "epoch": 1.859776490741496,
      "grad_norm": 0.7654532194137573,
      "learning_rate": 4.0632598731702373e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 21615744,
      "step": 11400
    },
    {
      "epoch": 1.859776490741496,
      "eval_loss": 0.14717352390289307,
      "eval_runtime": 133.2855,
      "eval_samples_per_second": 20.445,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 21615744,
      "step": 11400
    },
    {
      "epoch": 1.8605922179623136,
      "grad_norm": 1.0018669366836548,
      "learning_rate": 4.0624936151045426e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 21625008,
      "step": 11405
    },
    {
      "epoch": 1.8614079451831307,
      "grad_norm": 0.7419418692588806,
      "learning_rate": 4.061727116082696e-05,
      "loss": 0.1467,
      "num_input_tokens_seen": 21634192,
      "step": 11410
    },
    {
      "epoch": 1.8622236724039483,
      "grad_norm": 0.25655439496040344,
      "learning_rate": 4.060960376222903e-05,
      "loss": 0.0571,
      "num_input_tokens_seen": 21642256,
      "step": 11415
    },
    {
      "epoch": 1.8630393996247654,
      "grad_norm": 1.894083857536316,
      "learning_rate": 4.0601933956434034e-05,
      "loss": 0.1027,
      "num_input_tokens_seen": 21650128,
      "step": 11420
    },
    {
      "epoch": 1.863855126845583,
      "grad_norm": 0.3042335510253906,
      "learning_rate": 4.059426174462476e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 21660080,
      "step": 11425
    },
    {
      "epoch": 1.8646708540664,
      "grad_norm": 0.33941882848739624,
      "learning_rate": 4.058658712798435e-05,
      "loss": 0.0632,
      "num_input_tokens_seen": 21670480,
      "step": 11430
    },
    {
      "epoch": 1.8654865812872177,
      "grad_norm": 0.35730862617492676,
      "learning_rate": 4.0578910107696336e-05,
      "loss": 0.0772,
      "num_input_tokens_seen": 21679680,
      "step": 11435
    },
    {
      "epoch": 1.8663023085080348,
      "grad_norm": 0.612936794757843,
      "learning_rate": 4.05712306849446e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 21688528,
      "step": 11440
    },
    {
      "epoch": 1.8671180357288524,
      "grad_norm": 1.8345434665679932,
      "learning_rate": 4.0563548860913415e-05,
      "loss": 0.1918,
      "num_input_tokens_seen": 21697504,
      "step": 11445
    },
    {
      "epoch": 1.8679337629496695,
      "grad_norm": 0.22472675144672394,
      "learning_rate": 4.0555864636787414e-05,
      "loss": 0.0466,
      "num_input_tokens_seen": 21707296,
      "step": 11450
    },
    {
      "epoch": 1.868749490170487,
      "grad_norm": 0.6556540131568909,
      "learning_rate": 4.054817801375159e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 21716672,
      "step": 11455
    },
    {
      "epoch": 1.8695652173913042,
      "grad_norm": 0.031320083886384964,
      "learning_rate": 4.054048899299134e-05,
      "loss": 0.1776,
      "num_input_tokens_seen": 21725616,
      "step": 11460
    },
    {
      "epoch": 1.8703809446121218,
      "grad_norm": 0.2635636627674103,
      "learning_rate": 4.0532797575692385e-05,
      "loss": 0.1961,
      "num_input_tokens_seen": 21735296,
      "step": 11465
    },
    {
      "epoch": 1.871196671832939,
      "grad_norm": 0.9978274703025818,
      "learning_rate": 4.052510376304085e-05,
      "loss": 0.1883,
      "num_input_tokens_seen": 21744480,
      "step": 11470
    },
    {
      "epoch": 1.8720123990537565,
      "grad_norm": 1.6238669157028198,
      "learning_rate": 4.051740755622321e-05,
      "loss": 0.2411,
      "num_input_tokens_seen": 21753376,
      "step": 11475
    },
    {
      "epoch": 1.8728281262745736,
      "grad_norm": 1.6721055507659912,
      "learning_rate": 4.050970895642632e-05,
      "loss": 0.3687,
      "num_input_tokens_seen": 21761312,
      "step": 11480
    },
    {
      "epoch": 1.8736438534953912,
      "grad_norm": 0.12104020267724991,
      "learning_rate": 4.050200796483741e-05,
      "loss": 0.0806,
      "num_input_tokens_seen": 21770304,
      "step": 11485
    },
    {
      "epoch": 1.8744595807162086,
      "grad_norm": 1.6078588962554932,
      "learning_rate": 4.049430458264405e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 21778496,
      "step": 11490
    },
    {
      "epoch": 1.875275307937026,
      "grad_norm": 0.08209791779518127,
      "learning_rate": 4.048659881103422e-05,
      "loss": 0.1094,
      "num_input_tokens_seen": 21789056,
      "step": 11495
    },
    {
      "epoch": 1.8760910351578433,
      "grad_norm": 0.15113170444965363,
      "learning_rate": 4.0478890651196235e-05,
      "loss": 0.1097,
      "num_input_tokens_seen": 21799664,
      "step": 11500
    },
    {
      "epoch": 1.8769067623786606,
      "grad_norm": 0.7393327355384827,
      "learning_rate": 4.047118010431879e-05,
      "loss": 0.1438,
      "num_input_tokens_seen": 21809376,
      "step": 11505
    },
    {
      "epoch": 1.877722489599478,
      "grad_norm": 0.16186939179897308,
      "learning_rate": 4.046346717159094e-05,
      "loss": 0.0738,
      "num_input_tokens_seen": 21819456,
      "step": 11510
    },
    {
      "epoch": 1.8785382168202953,
      "grad_norm": 0.4545779824256897,
      "learning_rate": 4.045575185420214e-05,
      "loss": 0.1175,
      "num_input_tokens_seen": 21828976,
      "step": 11515
    },
    {
      "epoch": 1.8793539440411127,
      "grad_norm": 0.9493061900138855,
      "learning_rate": 4.0448034153342165e-05,
      "loss": 0.2443,
      "num_input_tokens_seen": 21837632,
      "step": 11520
    },
    {
      "epoch": 1.88016967126193,
      "grad_norm": 1.3527309894561768,
      "learning_rate": 4.0440314070201194e-05,
      "loss": 0.2595,
      "num_input_tokens_seen": 21846944,
      "step": 11525
    },
    {
      "epoch": 1.8809853984827474,
      "grad_norm": 0.6718131899833679,
      "learning_rate": 4.043259160596976e-05,
      "loss": 0.1081,
      "num_input_tokens_seen": 21855728,
      "step": 11530
    },
    {
      "epoch": 1.8818011257035647,
      "grad_norm": 0.6728736162185669,
      "learning_rate": 4.0424866761838767e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 21866512,
      "step": 11535
    },
    {
      "epoch": 1.882616852924382,
      "grad_norm": 0.5904411673545837,
      "learning_rate": 4.041713953899948e-05,
      "loss": 0.2176,
      "num_input_tokens_seen": 21875616,
      "step": 11540
    },
    {
      "epoch": 1.8834325801451994,
      "grad_norm": 1.3861593008041382,
      "learning_rate": 4.0409409938643515e-05,
      "loss": 0.1656,
      "num_input_tokens_seen": 21883632,
      "step": 11545
    },
    {
      "epoch": 1.8842483073660168,
      "grad_norm": 0.466372013092041,
      "learning_rate": 4.0401677961962904e-05,
      "loss": 0.1001,
      "num_input_tokens_seen": 21892544,
      "step": 11550
    },
    {
      "epoch": 1.8850640345868341,
      "grad_norm": 1.6274160146713257,
      "learning_rate": 4.039394361015001e-05,
      "loss": 0.1451,
      "num_input_tokens_seen": 21903408,
      "step": 11555
    },
    {
      "epoch": 1.8858797618076515,
      "grad_norm": 0.4691822826862335,
      "learning_rate": 4.038620688439755e-05,
      "loss": 0.0599,
      "num_input_tokens_seen": 21911744,
      "step": 11560
    },
    {
      "epoch": 1.8866954890284688,
      "grad_norm": 0.2745802700519562,
      "learning_rate": 4.037846778589862e-05,
      "loss": 0.0583,
      "num_input_tokens_seen": 21922016,
      "step": 11565
    },
    {
      "epoch": 1.8875112162492862,
      "grad_norm": 0.898066520690918,
      "learning_rate": 4.0370726315846715e-05,
      "loss": 0.1009,
      "num_input_tokens_seen": 21932912,
      "step": 11570
    },
    {
      "epoch": 1.8883269434701035,
      "grad_norm": 0.11612845212221146,
      "learning_rate": 4.036298247543565e-05,
      "loss": 0.1009,
      "num_input_tokens_seen": 21942864,
      "step": 11575
    },
    {
      "epoch": 1.8891426706909211,
      "grad_norm": 2.216395854949951,
      "learning_rate": 4.035523626585962e-05,
      "loss": 0.219,
      "num_input_tokens_seen": 21953296,
      "step": 11580
    },
    {
      "epoch": 1.8899583979117383,
      "grad_norm": 2.83276104927063,
      "learning_rate": 4.0347487688313194e-05,
      "loss": 0.2751,
      "num_input_tokens_seen": 21962480,
      "step": 11585
    },
    {
      "epoch": 1.8907741251325558,
      "grad_norm": 1.2178764343261719,
      "learning_rate": 4.0339736743991296e-05,
      "loss": 0.1553,
      "num_input_tokens_seen": 21972368,
      "step": 11590
    },
    {
      "epoch": 1.891589852353373,
      "grad_norm": 0.07029769569635391,
      "learning_rate": 4.0331983434089227e-05,
      "loss": 0.2029,
      "num_input_tokens_seen": 21981632,
      "step": 11595
    },
    {
      "epoch": 1.8924055795741905,
      "grad_norm": 1.1574082374572754,
      "learning_rate": 4.032422775980264e-05,
      "loss": 0.2225,
      "num_input_tokens_seen": 21991984,
      "step": 11600
    },
    {
      "epoch": 1.8924055795741905,
      "eval_loss": 0.1478807032108307,
      "eval_runtime": 133.2966,
      "eval_samples_per_second": 20.443,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 21991984,
      "step": 11600
    },
    {
      "epoch": 1.8932213067950077,
      "grad_norm": 0.04732925444841385,
      "learning_rate": 4.031646972232754e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 22001280,
      "step": 11605
    },
    {
      "epoch": 1.8940370340158252,
      "grad_norm": 0.8467029333114624,
      "learning_rate": 4.0308709322860344e-05,
      "loss": 0.0628,
      "num_input_tokens_seen": 22010336,
      "step": 11610
    },
    {
      "epoch": 1.8948527612366424,
      "grad_norm": 1.957285761833191,
      "learning_rate": 4.0300946562597784e-05,
      "loss": 0.2289,
      "num_input_tokens_seen": 22020720,
      "step": 11615
    },
    {
      "epoch": 1.89566848845746,
      "grad_norm": 0.24859103560447693,
      "learning_rate": 4.029318144273698e-05,
      "loss": 0.0495,
      "num_input_tokens_seen": 22028400,
      "step": 11620
    },
    {
      "epoch": 1.896484215678277,
      "grad_norm": 0.43284785747528076,
      "learning_rate": 4.0285413964475415e-05,
      "loss": 0.1567,
      "num_input_tokens_seen": 22038448,
      "step": 11625
    },
    {
      "epoch": 1.8972999428990946,
      "grad_norm": 0.45979979634284973,
      "learning_rate": 4.0277644129010927e-05,
      "loss": 0.1902,
      "num_input_tokens_seen": 22048512,
      "step": 11630
    },
    {
      "epoch": 1.8981156701199118,
      "grad_norm": 0.37100523710250854,
      "learning_rate": 4.0269871937541724e-05,
      "loss": 0.0711,
      "num_input_tokens_seen": 22057440,
      "step": 11635
    },
    {
      "epoch": 1.8989313973407294,
      "grad_norm": 1.4955861568450928,
      "learning_rate": 4.026209739126637e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 22066208,
      "step": 11640
    },
    {
      "epoch": 1.8997471245615465,
      "grad_norm": 0.17293769121170044,
      "learning_rate": 4.025432049138381e-05,
      "loss": 0.1821,
      "num_input_tokens_seen": 22075792,
      "step": 11645
    },
    {
      "epoch": 1.900562851782364,
      "grad_norm": 0.19491130113601685,
      "learning_rate": 4.0246541239093325e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 22084752,
      "step": 11650
    },
    {
      "epoch": 1.9013785790031812,
      "grad_norm": 1.418987512588501,
      "learning_rate": 4.023875963559459e-05,
      "loss": 0.1969,
      "num_input_tokens_seen": 22093568,
      "step": 11655
    },
    {
      "epoch": 1.9021943062239988,
      "grad_norm": 0.5643376708030701,
      "learning_rate": 4.023097568208761e-05,
      "loss": 0.3287,
      "num_input_tokens_seen": 22103296,
      "step": 11660
    },
    {
      "epoch": 1.903010033444816,
      "grad_norm": 0.8512290716171265,
      "learning_rate": 4.022318937977277e-05,
      "loss": 0.0838,
      "num_input_tokens_seen": 22113792,
      "step": 11665
    },
    {
      "epoch": 1.9038257606656335,
      "grad_norm": 0.6387978196144104,
      "learning_rate": 4.021540072985084e-05,
      "loss": 0.1216,
      "num_input_tokens_seen": 22123648,
      "step": 11670
    },
    {
      "epoch": 1.9046414878864508,
      "grad_norm": 0.16309775412082672,
      "learning_rate": 4.020760973352289e-05,
      "loss": 0.1604,
      "num_input_tokens_seen": 22133568,
      "step": 11675
    },
    {
      "epoch": 1.9054572151072682,
      "grad_norm": 0.05641605705022812,
      "learning_rate": 4.019981639199042e-05,
      "loss": 0.0124,
      "num_input_tokens_seen": 22143152,
      "step": 11680
    },
    {
      "epoch": 1.9062729423280855,
      "grad_norm": 0.49189749360084534,
      "learning_rate": 4.0192020706455245e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 22152144,
      "step": 11685
    },
    {
      "epoch": 1.9070886695489029,
      "grad_norm": 1.3718374967575073,
      "learning_rate": 4.018422267811956e-05,
      "loss": 0.1695,
      "num_input_tokens_seen": 22161888,
      "step": 11690
    },
    {
      "epoch": 1.9079043967697202,
      "grad_norm": 0.23225638270378113,
      "learning_rate": 4.017642230818592e-05,
      "loss": 0.0864,
      "num_input_tokens_seen": 22171072,
      "step": 11695
    },
    {
      "epoch": 1.9087201239905376,
      "grad_norm": 0.23266135156154633,
      "learning_rate": 4.0168619597857246e-05,
      "loss": 0.1108,
      "num_input_tokens_seen": 22179632,
      "step": 11700
    },
    {
      "epoch": 1.909535851211355,
      "grad_norm": 1.3649471998214722,
      "learning_rate": 4.016081454833681e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 22189072,
      "step": 11705
    },
    {
      "epoch": 1.9103515784321723,
      "grad_norm": 1.097195029258728,
      "learning_rate": 4.0153007160828245e-05,
      "loss": 0.2358,
      "num_input_tokens_seen": 22197632,
      "step": 11710
    },
    {
      "epoch": 1.9111673056529896,
      "grad_norm": 0.32613393664360046,
      "learning_rate": 4.0145197436535555e-05,
      "loss": 0.135,
      "num_input_tokens_seen": 22207888,
      "step": 11715
    },
    {
      "epoch": 1.911983032873807,
      "grad_norm": 0.5806831121444702,
      "learning_rate": 4.0137385376663095e-05,
      "loss": 0.2222,
      "num_input_tokens_seen": 22216288,
      "step": 11720
    },
    {
      "epoch": 1.9127987600946244,
      "grad_norm": 0.15963977575302124,
      "learning_rate": 4.012957098241558e-05,
      "loss": 0.0467,
      "num_input_tokens_seen": 22225664,
      "step": 11725
    },
    {
      "epoch": 1.9136144873154417,
      "grad_norm": 0.30686116218566895,
      "learning_rate": 4.0121754254998076e-05,
      "loss": 0.1119,
      "num_input_tokens_seen": 22235200,
      "step": 11730
    },
    {
      "epoch": 1.914430214536259,
      "grad_norm": 1.2839051485061646,
      "learning_rate": 4.011393519561606e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 22243216,
      "step": 11735
    },
    {
      "epoch": 1.9152459417570764,
      "grad_norm": 0.20227676630020142,
      "learning_rate": 4.010611380547529e-05,
      "loss": 0.1452,
      "num_input_tokens_seen": 22251968,
      "step": 11740
    },
    {
      "epoch": 1.9160616689778938,
      "grad_norm": 0.2446291297674179,
      "learning_rate": 4.009829008578192e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 22261632,
      "step": 11745
    },
    {
      "epoch": 1.9168773961987111,
      "grad_norm": 0.032251372933387756,
      "learning_rate": 4.00904640377425e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 22270448,
      "step": 11750
    },
    {
      "epoch": 1.9176931234195285,
      "grad_norm": 0.16704899072647095,
      "learning_rate": 4.0082635662563886e-05,
      "loss": 0.1187,
      "num_input_tokens_seen": 22279792,
      "step": 11755
    },
    {
      "epoch": 1.9185088506403458,
      "grad_norm": 1.5235897302627563,
      "learning_rate": 4.007480496145331e-05,
      "loss": 0.2809,
      "num_input_tokens_seen": 22290160,
      "step": 11760
    },
    {
      "epoch": 1.9193245778611632,
      "grad_norm": 0.12901902198791504,
      "learning_rate": 4.006697193561837e-05,
      "loss": 0.0682,
      "num_input_tokens_seen": 22300848,
      "step": 11765
    },
    {
      "epoch": 1.9201403050819805,
      "grad_norm": 0.45974838733673096,
      "learning_rate": 4.005913658626701e-05,
      "loss": 0.1333,
      "num_input_tokens_seen": 22309184,
      "step": 11770
    },
    {
      "epoch": 1.920956032302798,
      "grad_norm": 0.46754634380340576,
      "learning_rate": 4.005129891460754e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 22319632,
      "step": 11775
    },
    {
      "epoch": 1.9217717595236152,
      "grad_norm": 0.14386780560016632,
      "learning_rate": 4.004345892184864e-05,
      "loss": 0.2282,
      "num_input_tokens_seen": 22327552,
      "step": 11780
    },
    {
      "epoch": 1.9225874867444328,
      "grad_norm": 0.6693107485771179,
      "learning_rate": 4.003561660919932e-05,
      "loss": 0.2028,
      "num_input_tokens_seen": 22338112,
      "step": 11785
    },
    {
      "epoch": 1.92340321396525,
      "grad_norm": 0.24888060986995697,
      "learning_rate": 4.002777197786897e-05,
      "loss": 0.0657,
      "num_input_tokens_seen": 22346576,
      "step": 11790
    },
    {
      "epoch": 1.9242189411860675,
      "grad_norm": 2.4043707847595215,
      "learning_rate": 4.0019925029067326e-05,
      "loss": 0.2657,
      "num_input_tokens_seen": 22355984,
      "step": 11795
    },
    {
      "epoch": 1.9250346684068846,
      "grad_norm": 1.4188827276229858,
      "learning_rate": 4.0012075764004495e-05,
      "loss": 0.0627,
      "num_input_tokens_seen": 22366624,
      "step": 11800
    },
    {
      "epoch": 1.9250346684068846,
      "eval_loss": 0.1523992270231247,
      "eval_runtime": 133.3253,
      "eval_samples_per_second": 20.439,
      "eval_steps_per_second": 10.223,
      "num_input_tokens_seen": 22366624,
      "step": 11800
    },
    {
      "epoch": 1.9258503956277022,
      "grad_norm": 0.890707790851593,
      "learning_rate": 4.000422418389094e-05,
      "loss": 0.0648,
      "num_input_tokens_seen": 22376304,
      "step": 11805
    },
    {
      "epoch": 1.9266661228485193,
      "grad_norm": 1.251540184020996,
      "learning_rate": 3.999637028993744e-05,
      "loss": 0.1951,
      "num_input_tokens_seen": 22386448,
      "step": 11810
    },
    {
      "epoch": 1.927481850069337,
      "grad_norm": 2.250267505645752,
      "learning_rate": 3.99885140833552e-05,
      "loss": 0.2421,
      "num_input_tokens_seen": 22396912,
      "step": 11815
    },
    {
      "epoch": 1.928297577290154,
      "grad_norm": 0.6842126846313477,
      "learning_rate": 3.998065556535572e-05,
      "loss": 0.0615,
      "num_input_tokens_seen": 22404944,
      "step": 11820
    },
    {
      "epoch": 1.9291133045109716,
      "grad_norm": 0.5061923861503601,
      "learning_rate": 3.9972794737150895e-05,
      "loss": 0.1539,
      "num_input_tokens_seen": 22414848,
      "step": 11825
    },
    {
      "epoch": 1.9299290317317888,
      "grad_norm": 1.974499225616455,
      "learning_rate": 3.996493159995297e-05,
      "loss": 0.1655,
      "num_input_tokens_seen": 22424256,
      "step": 11830
    },
    {
      "epoch": 1.9307447589526063,
      "grad_norm": 0.6728468537330627,
      "learning_rate": 3.995706615497453e-05,
      "loss": 0.0867,
      "num_input_tokens_seen": 22433488,
      "step": 11835
    },
    {
      "epoch": 1.9315604861734235,
      "grad_norm": 0.7735950946807861,
      "learning_rate": 3.994919840342852e-05,
      "loss": 0.2123,
      "num_input_tokens_seen": 22442704,
      "step": 11840
    },
    {
      "epoch": 1.932376213394241,
      "grad_norm": 0.14590641856193542,
      "learning_rate": 3.994132834652825e-05,
      "loss": 0.0658,
      "num_input_tokens_seen": 22451056,
      "step": 11845
    },
    {
      "epoch": 1.9331919406150582,
      "grad_norm": 0.06835173070430756,
      "learning_rate": 3.99334559854874e-05,
      "loss": 0.028,
      "num_input_tokens_seen": 22462016,
      "step": 11850
    },
    {
      "epoch": 1.9340076678358757,
      "grad_norm": 1.0998111963272095,
      "learning_rate": 3.9925581321519955e-05,
      "loss": 0.2767,
      "num_input_tokens_seen": 22470128,
      "step": 11855
    },
    {
      "epoch": 1.934823395056693,
      "grad_norm": 1.0528508424758911,
      "learning_rate": 3.991770435584031e-05,
      "loss": 0.2299,
      "num_input_tokens_seen": 22478720,
      "step": 11860
    },
    {
      "epoch": 1.9356391222775104,
      "grad_norm": 0.8749222159385681,
      "learning_rate": 3.990982508966319e-05,
      "loss": 0.0949,
      "num_input_tokens_seen": 22488416,
      "step": 11865
    },
    {
      "epoch": 1.9364548494983278,
      "grad_norm": 0.1679299771785736,
      "learning_rate": 3.990194352420367e-05,
      "loss": 0.191,
      "num_input_tokens_seen": 22497328,
      "step": 11870
    },
    {
      "epoch": 1.9372705767191452,
      "grad_norm": 1.4510211944580078,
      "learning_rate": 3.9894059660677184e-05,
      "loss": 0.1889,
      "num_input_tokens_seen": 22506064,
      "step": 11875
    },
    {
      "epoch": 1.9380863039399625,
      "grad_norm": 0.49503910541534424,
      "learning_rate": 3.9886173500299526e-05,
      "loss": 0.268,
      "num_input_tokens_seen": 22515760,
      "step": 11880
    },
    {
      "epoch": 1.9389020311607799,
      "grad_norm": 0.26566189527511597,
      "learning_rate": 3.987828504428685e-05,
      "loss": 0.223,
      "num_input_tokens_seen": 22525888,
      "step": 11885
    },
    {
      "epoch": 1.9397177583815972,
      "grad_norm": 1.4375351667404175,
      "learning_rate": 3.987039429385565e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 22535936,
      "step": 11890
    },
    {
      "epoch": 1.9405334856024146,
      "grad_norm": 0.2837674021720886,
      "learning_rate": 3.986250125022277e-05,
      "loss": 0.0919,
      "num_input_tokens_seen": 22544800,
      "step": 11895
    },
    {
      "epoch": 1.941349212823232,
      "grad_norm": 0.4879368543624878,
      "learning_rate": 3.985460591460544e-05,
      "loss": 0.1179,
      "num_input_tokens_seen": 22554368,
      "step": 11900
    },
    {
      "epoch": 1.9421649400440493,
      "grad_norm": 0.05600414797663689,
      "learning_rate": 3.984670828822118e-05,
      "loss": 0.1434,
      "num_input_tokens_seen": 22562864,
      "step": 11905
    },
    {
      "epoch": 1.9429806672648666,
      "grad_norm": 1.8275933265686035,
      "learning_rate": 3.983880837228794e-05,
      "loss": 0.2723,
      "num_input_tokens_seen": 22573152,
      "step": 11910
    },
    {
      "epoch": 1.943796394485684,
      "grad_norm": 0.1928066611289978,
      "learning_rate": 3.983090616802396e-05,
      "loss": 0.0638,
      "num_input_tokens_seen": 22581728,
      "step": 11915
    },
    {
      "epoch": 1.9446121217065013,
      "grad_norm": 0.785148024559021,
      "learning_rate": 3.982300167664788e-05,
      "loss": 0.187,
      "num_input_tokens_seen": 22590688,
      "step": 11920
    },
    {
      "epoch": 1.9454278489273187,
      "grad_norm": 0.31870129704475403,
      "learning_rate": 3.981509489937868e-05,
      "loss": 0.2099,
      "num_input_tokens_seen": 22599808,
      "step": 11925
    },
    {
      "epoch": 1.946243576148136,
      "grad_norm": 0.2221892774105072,
      "learning_rate": 3.9807185837435643e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 22610128,
      "step": 11930
    },
    {
      "epoch": 1.9470593033689534,
      "grad_norm": 0.5833849310874939,
      "learning_rate": 3.9799274492038484e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 22618944,
      "step": 11935
    },
    {
      "epoch": 1.9478750305897707,
      "grad_norm": 0.15810216963291168,
      "learning_rate": 3.979136086440722e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 22628912,
      "step": 11940
    },
    {
      "epoch": 1.948690757810588,
      "grad_norm": 0.6166408658027649,
      "learning_rate": 3.9783444955762226e-05,
      "loss": 0.2222,
      "num_input_tokens_seen": 22638880,
      "step": 11945
    },
    {
      "epoch": 1.9495064850314054,
      "grad_norm": 0.46060946583747864,
      "learning_rate": 3.977552676732424e-05,
      "loss": 0.1306,
      "num_input_tokens_seen": 22648576,
      "step": 11950
    },
    {
      "epoch": 1.9503222122522228,
      "grad_norm": 1.0414763689041138,
      "learning_rate": 3.976760630031435e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 22658112,
      "step": 11955
    },
    {
      "epoch": 1.9511379394730404,
      "grad_norm": 0.40730196237564087,
      "learning_rate": 3.975968355595398e-05,
      "loss": 0.2808,
      "num_input_tokens_seen": 22666288,
      "step": 11960
    },
    {
      "epoch": 1.9519536666938575,
      "grad_norm": 0.19495777785778046,
      "learning_rate": 3.9751758535464935e-05,
      "loss": 0.0524,
      "num_input_tokens_seen": 22676720,
      "step": 11965
    },
    {
      "epoch": 1.952769393914675,
      "grad_norm": 0.6042168736457825,
      "learning_rate": 3.9743831240069326e-05,
      "loss": 0.0583,
      "num_input_tokens_seen": 22687424,
      "step": 11970
    },
    {
      "epoch": 1.9535851211354922,
      "grad_norm": 0.37556421756744385,
      "learning_rate": 3.9735901670989675e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 22697616,
      "step": 11975
    },
    {
      "epoch": 1.9544008483563098,
      "grad_norm": 0.6659266352653503,
      "learning_rate": 3.97279698294488e-05,
      "loss": 0.2005,
      "num_input_tokens_seen": 22706688,
      "step": 11980
    },
    {
      "epoch": 1.955216575577127,
      "grad_norm": 0.10067359358072281,
      "learning_rate": 3.9720035716669876e-05,
      "loss": 0.0967,
      "num_input_tokens_seen": 22716016,
      "step": 11985
    },
    {
      "epoch": 1.9560323027979445,
      "grad_norm": 0.24269096553325653,
      "learning_rate": 3.9712099333876474e-05,
      "loss": 0.0346,
      "num_input_tokens_seen": 22725936,
      "step": 11990
    },
    {
      "epoch": 1.9568480300187616,
      "grad_norm": 1.4274972677230835,
      "learning_rate": 3.9704160682292475e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 22735760,
      "step": 11995
    },
    {
      "epoch": 1.9576637572395792,
      "grad_norm": 1.1088314056396484,
      "learning_rate": 3.9696219763142106e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 22746000,
      "step": 12000
    },
    {
      "epoch": 1.9576637572395792,
      "eval_loss": 0.15141622722148895,
      "eval_runtime": 133.296,
      "eval_samples_per_second": 20.443,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 22746000,
      "step": 12000
    },
    {
      "epoch": 1.9584794844603963,
      "grad_norm": 0.6800643801689148,
      "learning_rate": 3.968827657764997e-05,
      "loss": 0.0942,
      "num_input_tokens_seen": 22754928,
      "step": 12005
    },
    {
      "epoch": 1.959295211681214,
      "grad_norm": 2.019308567047119,
      "learning_rate": 3.9680331127041e-05,
      "loss": 0.2874,
      "num_input_tokens_seen": 22763280,
      "step": 12010
    },
    {
      "epoch": 1.960110938902031,
      "grad_norm": 0.059951767325401306,
      "learning_rate": 3.9672383412540495e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 22772368,
      "step": 12015
    },
    {
      "epoch": 1.9609266661228486,
      "grad_norm": 0.12220216542482376,
      "learning_rate": 3.966443343537407e-05,
      "loss": 0.3258,
      "num_input_tokens_seen": 22781296,
      "step": 12020
    },
    {
      "epoch": 1.9617423933436657,
      "grad_norm": 0.41677185893058777,
      "learning_rate": 3.965648119676772e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 22791264,
      "step": 12025
    },
    {
      "epoch": 1.9625581205644833,
      "grad_norm": 2.234656572341919,
      "learning_rate": 3.96485266979478e-05,
      "loss": 0.2244,
      "num_input_tokens_seen": 22801136,
      "step": 12030
    },
    {
      "epoch": 1.9633738477853004,
      "grad_norm": 0.42844611406326294,
      "learning_rate": 3.9640569940140974e-05,
      "loss": 0.4248,
      "num_input_tokens_seen": 22811792,
      "step": 12035
    },
    {
      "epoch": 1.964189575006118,
      "grad_norm": 0.5006226301193237,
      "learning_rate": 3.963261092457428e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 22821616,
      "step": 12040
    },
    {
      "epoch": 1.9650053022269351,
      "grad_norm": 0.23480698466300964,
      "learning_rate": 3.962464965247509e-05,
      "loss": 0.0788,
      "num_input_tokens_seen": 22831968,
      "step": 12045
    },
    {
      "epoch": 1.9658210294477527,
      "grad_norm": 0.8284492492675781,
      "learning_rate": 3.9616686125071135e-05,
      "loss": 0.0627,
      "num_input_tokens_seen": 22841040,
      "step": 12050
    },
    {
      "epoch": 1.96663675666857,
      "grad_norm": 0.7856388688087463,
      "learning_rate": 3.9608720343590506e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 22849232,
      "step": 12055
    },
    {
      "epoch": 1.9674524838893874,
      "grad_norm": 0.4097345769405365,
      "learning_rate": 3.960075230926161e-05,
      "loss": 0.0934,
      "num_input_tokens_seen": 22859632,
      "step": 12060
    },
    {
      "epoch": 1.9682682111102048,
      "grad_norm": 0.40390732884407043,
      "learning_rate": 3.959278202331322e-05,
      "loss": 0.0772,
      "num_input_tokens_seen": 22868944,
      "step": 12065
    },
    {
      "epoch": 1.9690839383310221,
      "grad_norm": 0.5782273411750793,
      "learning_rate": 3.958480948697446e-05,
      "loss": 0.1113,
      "num_input_tokens_seen": 22878784,
      "step": 12070
    },
    {
      "epoch": 1.9698996655518395,
      "grad_norm": 0.1302112191915512,
      "learning_rate": 3.95768347014748e-05,
      "loss": 0.0823,
      "num_input_tokens_seen": 22888544,
      "step": 12075
    },
    {
      "epoch": 1.9707153927726568,
      "grad_norm": 0.2894367277622223,
      "learning_rate": 3.956885766804404e-05,
      "loss": 0.0537,
      "num_input_tokens_seen": 22897840,
      "step": 12080
    },
    {
      "epoch": 1.9715311199934742,
      "grad_norm": 0.22163352370262146,
      "learning_rate": 3.956087838791235e-05,
      "loss": 0.0833,
      "num_input_tokens_seen": 22907472,
      "step": 12085
    },
    {
      "epoch": 1.9723468472142915,
      "grad_norm": 2.068437337875366,
      "learning_rate": 3.955289686231022e-05,
      "loss": 0.2341,
      "num_input_tokens_seen": 22916800,
      "step": 12090
    },
    {
      "epoch": 1.973162574435109,
      "grad_norm": 1.7170093059539795,
      "learning_rate": 3.9544913092468504e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 22925840,
      "step": 12095
    },
    {
      "epoch": 1.9739783016559262,
      "grad_norm": 0.7267276048660278,
      "learning_rate": 3.9536927079618425e-05,
      "loss": 0.0524,
      "num_input_tokens_seen": 22935680,
      "step": 12100
    },
    {
      "epoch": 1.9747940288767436,
      "grad_norm": 0.14404815435409546,
      "learning_rate": 3.9528938824991494e-05,
      "loss": 0.2106,
      "num_input_tokens_seen": 22944144,
      "step": 12105
    },
    {
      "epoch": 1.975609756097561,
      "grad_norm": 0.32211461663246155,
      "learning_rate": 3.952094832981962e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 22954400,
      "step": 12110
    },
    {
      "epoch": 1.9764254833183783,
      "grad_norm": 0.17270123958587646,
      "learning_rate": 3.951295559533503e-05,
      "loss": 0.2316,
      "num_input_tokens_seen": 22963984,
      "step": 12115
    },
    {
      "epoch": 1.9772412105391957,
      "grad_norm": 0.6528754234313965,
      "learning_rate": 3.95049606227703e-05,
      "loss": 0.1828,
      "num_input_tokens_seen": 22972672,
      "step": 12120
    },
    {
      "epoch": 1.978056937760013,
      "grad_norm": 1.4836868047714233,
      "learning_rate": 3.949696341335838e-05,
      "loss": 0.3275,
      "num_input_tokens_seen": 22982464,
      "step": 12125
    },
    {
      "epoch": 1.9788726649808304,
      "grad_norm": 2.190645933151245,
      "learning_rate": 3.9488963968332503e-05,
      "loss": 0.1836,
      "num_input_tokens_seen": 22991936,
      "step": 12130
    },
    {
      "epoch": 1.9796883922016477,
      "grad_norm": 0.1550142765045166,
      "learning_rate": 3.948096228892631e-05,
      "loss": 0.2076,
      "num_input_tokens_seen": 23002128,
      "step": 12135
    },
    {
      "epoch": 1.980504119422465,
      "grad_norm": 0.5563657283782959,
      "learning_rate": 3.947295837637375e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 23011920,
      "step": 12140
    },
    {
      "epoch": 1.9813198466432826,
      "grad_norm": 1.1334388256072998,
      "learning_rate": 3.9464952231909135e-05,
      "loss": 0.0847,
      "num_input_tokens_seen": 23021216,
      "step": 12145
    },
    {
      "epoch": 1.9821355738640998,
      "grad_norm": 0.3843059539794922,
      "learning_rate": 3.945694385676711e-05,
      "loss": 0.1014,
      "num_input_tokens_seen": 23029616,
      "step": 12150
    },
    {
      "epoch": 1.9829513010849174,
      "grad_norm": 1.270372986793518,
      "learning_rate": 3.944893325218265e-05,
      "loss": 0.3306,
      "num_input_tokens_seen": 23038304,
      "step": 12155
    },
    {
      "epoch": 1.9837670283057345,
      "grad_norm": 0.1844525933265686,
      "learning_rate": 3.944092041939112e-05,
      "loss": 0.0776,
      "num_input_tokens_seen": 23047536,
      "step": 12160
    },
    {
      "epoch": 1.984582755526552,
      "grad_norm": 1.6386975049972534,
      "learning_rate": 3.943290535962818e-05,
      "loss": 0.1754,
      "num_input_tokens_seen": 23057456,
      "step": 12165
    },
    {
      "epoch": 1.9853984827473692,
      "grad_norm": 0.7835429310798645,
      "learning_rate": 3.942488807412985e-05,
      "loss": 0.1146,
      "num_input_tokens_seen": 23066496,
      "step": 12170
    },
    {
      "epoch": 1.9862142099681868,
      "grad_norm": 1.1459020376205444,
      "learning_rate": 3.941686856413251e-05,
      "loss": 0.2758,
      "num_input_tokens_seen": 23076256,
      "step": 12175
    },
    {
      "epoch": 1.987029937189004,
      "grad_norm": 0.7891173958778381,
      "learning_rate": 3.9408846830872874e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 23084976,
      "step": 12180
    },
    {
      "epoch": 1.9878456644098215,
      "grad_norm": 0.6193313002586365,
      "learning_rate": 3.940082287558798e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 23094816,
      "step": 12185
    },
    {
      "epoch": 1.9886613916306386,
      "grad_norm": 0.28931164741516113,
      "learning_rate": 3.939279669951522e-05,
      "loss": 0.096,
      "num_input_tokens_seen": 23103552,
      "step": 12190
    },
    {
      "epoch": 1.9894771188514562,
      "grad_norm": 1.3236078023910522,
      "learning_rate": 3.938476830389234e-05,
      "loss": 0.2546,
      "num_input_tokens_seen": 23113184,
      "step": 12195
    },
    {
      "epoch": 1.9902928460722733,
      "grad_norm": 3.1173746585845947,
      "learning_rate": 3.937673768995742e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 23122688,
      "step": 12200
    },
    {
      "epoch": 1.9902928460722733,
      "eval_loss": 0.14552147686481476,
      "eval_runtime": 133.3783,
      "eval_samples_per_second": 20.431,
      "eval_steps_per_second": 10.219,
      "num_input_tokens_seen": 23122688,
      "step": 12200
    },
    {
      "epoch": 1.9911085732930909,
      "grad_norm": 2.430396795272827,
      "learning_rate": 3.936870485894888e-05,
      "loss": 0.2162,
      "num_input_tokens_seen": 23131520,
      "step": 12205
    },
    {
      "epoch": 1.991924300513908,
      "grad_norm": 0.06866256147623062,
      "learning_rate": 3.9360669812105475e-05,
      "loss": 0.0648,
      "num_input_tokens_seen": 23141248,
      "step": 12210
    },
    {
      "epoch": 1.9927400277347256,
      "grad_norm": 0.5172948241233826,
      "learning_rate": 3.9352632550666325e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 23150080,
      "step": 12215
    },
    {
      "epoch": 1.9935557549555427,
      "grad_norm": 0.6864575147628784,
      "learning_rate": 3.9344593075870866e-05,
      "loss": 0.1335,
      "num_input_tokens_seen": 23159936,
      "step": 12220
    },
    {
      "epoch": 1.9943714821763603,
      "grad_norm": 0.4149850308895111,
      "learning_rate": 3.933655138895889e-05,
      "loss": 0.1139,
      "num_input_tokens_seen": 23168464,
      "step": 12225
    },
    {
      "epoch": 1.9951872093971774,
      "grad_norm": 0.3074190020561218,
      "learning_rate": 3.932850749117053e-05,
      "loss": 0.1201,
      "num_input_tokens_seen": 23177600,
      "step": 12230
    },
    {
      "epoch": 1.996002936617995,
      "grad_norm": 0.41044825315475464,
      "learning_rate": 3.932046138374624e-05,
      "loss": 0.0739,
      "num_input_tokens_seen": 23185872,
      "step": 12235
    },
    {
      "epoch": 1.9968186638388123,
      "grad_norm": 0.8182842135429382,
      "learning_rate": 3.9312413067926854e-05,
      "loss": 0.117,
      "num_input_tokens_seen": 23194848,
      "step": 12240
    },
    {
      "epoch": 1.9976343910596297,
      "grad_norm": 0.8925697207450867,
      "learning_rate": 3.9304362544953506e-05,
      "loss": 0.0794,
      "num_input_tokens_seen": 23203952,
      "step": 12245
    },
    {
      "epoch": 1.998450118280447,
      "grad_norm": 1.1667866706848145,
      "learning_rate": 3.929630981606769e-05,
      "loss": 0.1068,
      "num_input_tokens_seen": 23213760,
      "step": 12250
    },
    {
      "epoch": 1.9992658455012644,
      "grad_norm": 0.83303302526474,
      "learning_rate": 3.928825488251124e-05,
      "loss": 0.1088,
      "num_input_tokens_seen": 23224064,
      "step": 12255
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9931384325027466,
      "learning_rate": 3.9280197745526344e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 23232704,
      "step": 12260
    },
    {
      "epoch": 2.0008157272208176,
      "grad_norm": 1.6320682764053345,
      "learning_rate": 3.9272138406355495e-05,
      "loss": 0.2106,
      "num_input_tokens_seen": 23241536,
      "step": 12265
    },
    {
      "epoch": 2.0016314544416347,
      "grad_norm": 1.3561309576034546,
      "learning_rate": 3.926407686624154e-05,
      "loss": 0.2316,
      "num_input_tokens_seen": 23249984,
      "step": 12270
    },
    {
      "epoch": 2.0024471816624523,
      "grad_norm": 0.34414154291152954,
      "learning_rate": 3.9256013126427684e-05,
      "loss": 0.0847,
      "num_input_tokens_seen": 23260144,
      "step": 12275
    },
    {
      "epoch": 2.0032629088832694,
      "grad_norm": 0.6067934632301331,
      "learning_rate": 3.9247947188157455e-05,
      "loss": 0.0961,
      "num_input_tokens_seen": 23268880,
      "step": 12280
    },
    {
      "epoch": 2.004078636104087,
      "grad_norm": 0.09318374842405319,
      "learning_rate": 3.9239879052674715e-05,
      "loss": 0.0345,
      "num_input_tokens_seen": 23277056,
      "step": 12285
    },
    {
      "epoch": 2.004894363324904,
      "grad_norm": 0.3354227840900421,
      "learning_rate": 3.9231808721223673e-05,
      "loss": 0.2185,
      "num_input_tokens_seen": 23286144,
      "step": 12290
    },
    {
      "epoch": 2.0057100905457217,
      "grad_norm": 0.10533009469509125,
      "learning_rate": 3.9223736195048886e-05,
      "loss": 0.0978,
      "num_input_tokens_seen": 23294688,
      "step": 12295
    },
    {
      "epoch": 2.006525817766539,
      "grad_norm": 0.14456360042095184,
      "learning_rate": 3.921566147539523e-05,
      "loss": 0.1534,
      "num_input_tokens_seen": 23304672,
      "step": 12300
    },
    {
      "epoch": 2.0073415449873564,
      "grad_norm": 0.1092853844165802,
      "learning_rate": 3.920758456350792e-05,
      "loss": 0.1001,
      "num_input_tokens_seen": 23314544,
      "step": 12305
    },
    {
      "epoch": 2.0081572722081735,
      "grad_norm": 0.0611654557287693,
      "learning_rate": 3.919950546063253e-05,
      "loss": 0.0578,
      "num_input_tokens_seen": 23324848,
      "step": 12310
    },
    {
      "epoch": 2.008972999428991,
      "grad_norm": 0.9641533493995667,
      "learning_rate": 3.919142416801496e-05,
      "loss": 0.4528,
      "num_input_tokens_seen": 23333216,
      "step": 12315
    },
    {
      "epoch": 2.0097887266498082,
      "grad_norm": 0.3231406509876251,
      "learning_rate": 3.918334068690144e-05,
      "loss": 0.1038,
      "num_input_tokens_seen": 23343200,
      "step": 12320
    },
    {
      "epoch": 2.010604453870626,
      "grad_norm": 0.2532288432121277,
      "learning_rate": 3.917525501853855e-05,
      "loss": 0.1102,
      "num_input_tokens_seen": 23352736,
      "step": 12325
    },
    {
      "epoch": 2.011420181091443,
      "grad_norm": 0.2003534436225891,
      "learning_rate": 3.916716716417319e-05,
      "loss": 0.1275,
      "num_input_tokens_seen": 23362416,
      "step": 12330
    },
    {
      "epoch": 2.0122359083122605,
      "grad_norm": 1.7709068059921265,
      "learning_rate": 3.915907712505263e-05,
      "loss": 0.2142,
      "num_input_tokens_seen": 23370656,
      "step": 12335
    },
    {
      "epoch": 2.0130516355330776,
      "grad_norm": 1.4860738515853882,
      "learning_rate": 3.915098490242444e-05,
      "loss": 0.3198,
      "num_input_tokens_seen": 23380288,
      "step": 12340
    },
    {
      "epoch": 2.013867362753895,
      "grad_norm": 0.785889744758606,
      "learning_rate": 3.914289049753654e-05,
      "loss": 0.1064,
      "num_input_tokens_seen": 23390272,
      "step": 12345
    },
    {
      "epoch": 2.0146830899747123,
      "grad_norm": 1.2955818176269531,
      "learning_rate": 3.913479391163719e-05,
      "loss": 0.2592,
      "num_input_tokens_seen": 23399744,
      "step": 12350
    },
    {
      "epoch": 2.01549881719553,
      "grad_norm": 0.24101385474205017,
      "learning_rate": 3.9126695145975e-05,
      "loss": 0.1225,
      "num_input_tokens_seen": 23409616,
      "step": 12355
    },
    {
      "epoch": 2.016314544416347,
      "grad_norm": 1.2334636449813843,
      "learning_rate": 3.911859420179889e-05,
      "loss": 0.3396,
      "num_input_tokens_seen": 23418064,
      "step": 12360
    },
    {
      "epoch": 2.0171302716371646,
      "grad_norm": 0.3055287301540375,
      "learning_rate": 3.911049108035813e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 23428160,
      "step": 12365
    },
    {
      "epoch": 2.0179459988579818,
      "grad_norm": 0.5176775455474854,
      "learning_rate": 3.910238578290232e-05,
      "loss": 0.0771,
      "num_input_tokens_seen": 23437760,
      "step": 12370
    },
    {
      "epoch": 2.0187617260787993,
      "grad_norm": 0.523918867111206,
      "learning_rate": 3.90942783106814e-05,
      "loss": 0.2122,
      "num_input_tokens_seen": 23445584,
      "step": 12375
    },
    {
      "epoch": 2.0195774532996165,
      "grad_norm": 0.08426351100206375,
      "learning_rate": 3.908616866494564e-05,
      "loss": 0.0954,
      "num_input_tokens_seen": 23454704,
      "step": 12380
    },
    {
      "epoch": 2.020393180520434,
      "grad_norm": 0.87041175365448,
      "learning_rate": 3.907805684694566e-05,
      "loss": 0.2023,
      "num_input_tokens_seen": 23464720,
      "step": 12385
    },
    {
      "epoch": 2.021208907741251,
      "grad_norm": 1.675553798675537,
      "learning_rate": 3.90699428579324e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 23473888,
      "step": 12390
    },
    {
      "epoch": 2.0220246349620687,
      "grad_norm": 1.4229949712753296,
      "learning_rate": 3.906182669915713e-05,
      "loss": 0.1974,
      "num_input_tokens_seen": 23483536,
      "step": 12395
    },
    {
      "epoch": 2.022840362182886,
      "grad_norm": 0.5397219061851501,
      "learning_rate": 3.9053708371871476e-05,
      "loss": 0.092,
      "num_input_tokens_seen": 23494112,
      "step": 12400
    },
    {
      "epoch": 2.022840362182886,
      "eval_loss": 0.14734505116939545,
      "eval_runtime": 133.258,
      "eval_samples_per_second": 20.449,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 23494112,
      "step": 12400
    },
    {
      "epoch": 2.0236560894037035,
      "grad_norm": 1.823425531387329,
      "learning_rate": 3.904558787732738e-05,
      "loss": 0.0759,
      "num_input_tokens_seen": 23503232,
      "step": 12405
    },
    {
      "epoch": 2.0244718166245206,
      "grad_norm": 0.249756321310997,
      "learning_rate": 3.9037465216777135e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 23513472,
      "step": 12410
    },
    {
      "epoch": 2.025287543845338,
      "grad_norm": 0.1325598806142807,
      "learning_rate": 3.902934039147334e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 23522640,
      "step": 12415
    },
    {
      "epoch": 2.0261032710661553,
      "grad_norm": 0.2095559984445572,
      "learning_rate": 3.902121340266894e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 23532784,
      "step": 12420
    },
    {
      "epoch": 2.026918998286973,
      "grad_norm": 0.6447151303291321,
      "learning_rate": 3.9013084251617246e-05,
      "loss": 0.3105,
      "num_input_tokens_seen": 23541408,
      "step": 12425
    },
    {
      "epoch": 2.02773472550779,
      "grad_norm": 0.8178647756576538,
      "learning_rate": 3.9004952939571865e-05,
      "loss": 0.0682,
      "num_input_tokens_seen": 23551872,
      "step": 12430
    },
    {
      "epoch": 2.0285504527286076,
      "grad_norm": 0.5207479000091553,
      "learning_rate": 3.899681946778673e-05,
      "loss": 0.2006,
      "num_input_tokens_seen": 23559936,
      "step": 12435
    },
    {
      "epoch": 2.0293661799494247,
      "grad_norm": 0.4977983832359314,
      "learning_rate": 3.898868383751615e-05,
      "loss": 0.1049,
      "num_input_tokens_seen": 23570208,
      "step": 12440
    },
    {
      "epoch": 2.0301819071702423,
      "grad_norm": 0.9541086554527283,
      "learning_rate": 3.8980546050014724e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 23580000,
      "step": 12445
    },
    {
      "epoch": 2.03099763439106,
      "grad_norm": 0.31939271092414856,
      "learning_rate": 3.897240610653741e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 23589824,
      "step": 12450
    },
    {
      "epoch": 2.031813361611877,
      "grad_norm": 1.1031460762023926,
      "learning_rate": 3.896426400833948e-05,
      "loss": 0.4198,
      "num_input_tokens_seen": 23599408,
      "step": 12455
    },
    {
      "epoch": 2.0326290888326946,
      "grad_norm": 0.15975350141525269,
      "learning_rate": 3.895611975667656e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 23608720,
      "step": 12460
    },
    {
      "epoch": 2.0334448160535117,
      "grad_norm": 0.3549121618270874,
      "learning_rate": 3.8947973352804584e-05,
      "loss": 0.105,
      "num_input_tokens_seen": 23619232,
      "step": 12465
    },
    {
      "epoch": 2.0342605432743293,
      "grad_norm": 0.07029084861278534,
      "learning_rate": 3.893982479797984e-05,
      "loss": 0.2667,
      "num_input_tokens_seen": 23628864,
      "step": 12470
    },
    {
      "epoch": 2.0350762704951464,
      "grad_norm": 0.4454396367073059,
      "learning_rate": 3.8931674093458926e-05,
      "loss": 0.0878,
      "num_input_tokens_seen": 23638768,
      "step": 12475
    },
    {
      "epoch": 2.035891997715964,
      "grad_norm": 0.27649539709091187,
      "learning_rate": 3.89235212404988e-05,
      "loss": 0.0913,
      "num_input_tokens_seen": 23649056,
      "step": 12480
    },
    {
      "epoch": 2.036707724936781,
      "grad_norm": 0.08364949375391006,
      "learning_rate": 3.891536624035672e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 23658096,
      "step": 12485
    },
    {
      "epoch": 2.0375234521575987,
      "grad_norm": 0.20229433476924896,
      "learning_rate": 3.8907209094290295e-05,
      "loss": 0.0813,
      "num_input_tokens_seen": 23668096,
      "step": 12490
    },
    {
      "epoch": 2.038339179378416,
      "grad_norm": 0.38800951838493347,
      "learning_rate": 3.8899049803557466e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 23676832,
      "step": 12495
    },
    {
      "epoch": 2.0391549065992334,
      "grad_norm": 0.4184378385543823,
      "learning_rate": 3.889088836941648e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 23685328,
      "step": 12500
    },
    {
      "epoch": 2.0399706338200505,
      "grad_norm": 0.3800109922885895,
      "learning_rate": 3.8882724793125946e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 23695936,
      "step": 12505
    },
    {
      "epoch": 2.040786361040868,
      "grad_norm": 0.6817774772644043,
      "learning_rate": 3.8874559075944794e-05,
      "loss": 0.0769,
      "num_input_tokens_seen": 23706288,
      "step": 12510
    },
    {
      "epoch": 2.041602088261685,
      "grad_norm": 0.3373008966445923,
      "learning_rate": 3.886639121913227e-05,
      "loss": 0.1123,
      "num_input_tokens_seen": 23715744,
      "step": 12515
    },
    {
      "epoch": 2.042417815482503,
      "grad_norm": 0.9248743653297424,
      "learning_rate": 3.885822122394797e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 23725968,
      "step": 12520
    },
    {
      "epoch": 2.04323354270332,
      "grad_norm": 1.205683708190918,
      "learning_rate": 3.8850049091651794e-05,
      "loss": 0.1119,
      "num_input_tokens_seen": 23735664,
      "step": 12525
    },
    {
      "epoch": 2.0440492699241375,
      "grad_norm": 0.3264557123184204,
      "learning_rate": 3.8841874823504e-05,
      "loss": 0.2041,
      "num_input_tokens_seen": 23745520,
      "step": 12530
    },
    {
      "epoch": 2.0448649971449546,
      "grad_norm": 0.27525630593299866,
      "learning_rate": 3.8833698420765157e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 23754720,
      "step": 12535
    },
    {
      "epoch": 2.045680724365772,
      "grad_norm": 0.7969487309455872,
      "learning_rate": 3.882551988469618e-05,
      "loss": 0.1064,
      "num_input_tokens_seen": 23764432,
      "step": 12540
    },
    {
      "epoch": 2.0464964515865893,
      "grad_norm": 0.7032173275947571,
      "learning_rate": 3.881733921655829e-05,
      "loss": 0.0963,
      "num_input_tokens_seen": 23773728,
      "step": 12545
    },
    {
      "epoch": 2.047312178807407,
      "grad_norm": 0.16612198948860168,
      "learning_rate": 3.8809156417613054e-05,
      "loss": 0.0454,
      "num_input_tokens_seen": 23782736,
      "step": 12550
    },
    {
      "epoch": 2.048127906028224,
      "grad_norm": 0.38835644721984863,
      "learning_rate": 3.8800971489122364e-05,
      "loss": 0.0661,
      "num_input_tokens_seen": 23791520,
      "step": 12555
    },
    {
      "epoch": 2.0489436332490416,
      "grad_norm": 0.4835468530654907,
      "learning_rate": 3.8792784432348434e-05,
      "loss": 0.067,
      "num_input_tokens_seen": 23800608,
      "step": 12560
    },
    {
      "epoch": 2.0497593604698587,
      "grad_norm": 0.11131532490253448,
      "learning_rate": 3.878459524855381e-05,
      "loss": 0.1773,
      "num_input_tokens_seen": 23810384,
      "step": 12565
    },
    {
      "epoch": 2.0505750876906763,
      "grad_norm": 1.3866009712219238,
      "learning_rate": 3.8776403939001384e-05,
      "loss": 0.1302,
      "num_input_tokens_seen": 23819856,
      "step": 12570
    },
    {
      "epoch": 2.0513908149114934,
      "grad_norm": 0.8341147303581238,
      "learning_rate": 3.876821050495433e-05,
      "loss": 0.0961,
      "num_input_tokens_seen": 23828912,
      "step": 12575
    },
    {
      "epoch": 2.052206542132311,
      "grad_norm": 0.21488183736801147,
      "learning_rate": 3.87600149476762e-05,
      "loss": 0.2142,
      "num_input_tokens_seen": 23839168,
      "step": 12580
    },
    {
      "epoch": 2.053022269353128,
      "grad_norm": 1.197043776512146,
      "learning_rate": 3.8751817268430843e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 23848752,
      "step": 12585
    },
    {
      "epoch": 2.0538379965739457,
      "grad_norm": 0.8305917382240295,
      "learning_rate": 3.8743617468482464e-05,
      "loss": 0.122,
      "num_input_tokens_seen": 23857120,
      "step": 12590
    },
    {
      "epoch": 2.054653723794763,
      "grad_norm": 0.23664328455924988,
      "learning_rate": 3.8735415549095535e-05,
      "loss": 0.0556,
      "num_input_tokens_seen": 23866592,
      "step": 12595
    },
    {
      "epoch": 2.0554694510155804,
      "grad_norm": 0.7725763916969299,
      "learning_rate": 3.8727211511534934e-05,
      "loss": 0.2195,
      "num_input_tokens_seen": 23876160,
      "step": 12600
    },
    {
      "epoch": 2.0554694510155804,
      "eval_loss": 0.14916467666625977,
      "eval_runtime": 133.2502,
      "eval_samples_per_second": 20.45,
      "eval_steps_per_second": 10.229,
      "num_input_tokens_seen": 23876160,
      "step": 12600
    },
    {
      "epoch": 2.0562851782363976,
      "grad_norm": 0.5169873833656311,
      "learning_rate": 3.8719005357065804e-05,
      "loss": 0.1836,
      "num_input_tokens_seen": 23885344,
      "step": 12605
    },
    {
      "epoch": 2.057100905457215,
      "grad_norm": 0.4601498544216156,
      "learning_rate": 3.8710797086953645e-05,
      "loss": 0.2121,
      "num_input_tokens_seen": 23895024,
      "step": 12610
    },
    {
      "epoch": 2.0579166326780323,
      "grad_norm": 0.3520947992801666,
      "learning_rate": 3.870258670246427e-05,
      "loss": 0.1807,
      "num_input_tokens_seen": 23905376,
      "step": 12615
    },
    {
      "epoch": 2.05873235989885,
      "grad_norm": 0.3551803231239319,
      "learning_rate": 3.869437420486384e-05,
      "loss": 0.0819,
      "num_input_tokens_seen": 23914432,
      "step": 12620
    },
    {
      "epoch": 2.059548087119667,
      "grad_norm": 1.1043329238891602,
      "learning_rate": 3.8686159595418805e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 23922560,
      "step": 12625
    },
    {
      "epoch": 2.0603638143404845,
      "grad_norm": 1.172982931137085,
      "learning_rate": 3.867794287539597e-05,
      "loss": 0.0918,
      "num_input_tokens_seen": 23933008,
      "step": 12630
    },
    {
      "epoch": 2.0611795415613017,
      "grad_norm": 0.38045549392700195,
      "learning_rate": 3.866972404606245e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 23942080,
      "step": 12635
    },
    {
      "epoch": 2.0619952687821193,
      "grad_norm": 0.5184593796730042,
      "learning_rate": 3.866150310868571e-05,
      "loss": 0.0581,
      "num_input_tokens_seen": 23952608,
      "step": 12640
    },
    {
      "epoch": 2.062810996002937,
      "grad_norm": 0.23829449713230133,
      "learning_rate": 3.8653280064533506e-05,
      "loss": 0.0877,
      "num_input_tokens_seen": 23962848,
      "step": 12645
    },
    {
      "epoch": 2.063626723223754,
      "grad_norm": 0.4420577585697174,
      "learning_rate": 3.864505491487394e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 23972592,
      "step": 12650
    },
    {
      "epoch": 2.0644424504445715,
      "grad_norm": 0.4057670831680298,
      "learning_rate": 3.8636827660975414e-05,
      "loss": 0.0455,
      "num_input_tokens_seen": 23981776,
      "step": 12655
    },
    {
      "epoch": 2.0652581776653887,
      "grad_norm": 0.8946774005889893,
      "learning_rate": 3.862859830410671e-05,
      "loss": 0.1525,
      "num_input_tokens_seen": 23991904,
      "step": 12660
    },
    {
      "epoch": 2.0660739048862062,
      "grad_norm": 0.6711838841438293,
      "learning_rate": 3.862036684553688e-05,
      "loss": 0.2346,
      "num_input_tokens_seen": 24002896,
      "step": 12665
    },
    {
      "epoch": 2.0668896321070234,
      "grad_norm": 0.047777336090803146,
      "learning_rate": 3.8612133286535314e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 24012224,
      "step": 12670
    },
    {
      "epoch": 2.067705359327841,
      "grad_norm": 0.8970775008201599,
      "learning_rate": 3.860389762837173e-05,
      "loss": 0.2437,
      "num_input_tokens_seen": 24020672,
      "step": 12675
    },
    {
      "epoch": 2.068521086548658,
      "grad_norm": 0.3496270775794983,
      "learning_rate": 3.859565987231618e-05,
      "loss": 0.0429,
      "num_input_tokens_seen": 24029024,
      "step": 12680
    },
    {
      "epoch": 2.0693368137694756,
      "grad_norm": 0.5887122750282288,
      "learning_rate": 3.858742001963902e-05,
      "loss": 0.0992,
      "num_input_tokens_seen": 24039568,
      "step": 12685
    },
    {
      "epoch": 2.0701525409902928,
      "grad_norm": 0.14262554049491882,
      "learning_rate": 3.857917807161094e-05,
      "loss": 0.0461,
      "num_input_tokens_seen": 24049088,
      "step": 12690
    },
    {
      "epoch": 2.0709682682111104,
      "grad_norm": 0.13017761707305908,
      "learning_rate": 3.857093402950296e-05,
      "loss": 0.0635,
      "num_input_tokens_seen": 24058928,
      "step": 12695
    },
    {
      "epoch": 2.0717839954319275,
      "grad_norm": 0.47921308875083923,
      "learning_rate": 3.8562687894586414e-05,
      "loss": 0.2322,
      "num_input_tokens_seen": 24068896,
      "step": 12700
    },
    {
      "epoch": 2.072599722652745,
      "grad_norm": 1.915520429611206,
      "learning_rate": 3.8554439668132946e-05,
      "loss": 0.242,
      "num_input_tokens_seen": 24078992,
      "step": 12705
    },
    {
      "epoch": 2.073415449873562,
      "grad_norm": 0.16005180776119232,
      "learning_rate": 3.854618935141455e-05,
      "loss": 0.0984,
      "num_input_tokens_seen": 24088864,
      "step": 12710
    },
    {
      "epoch": 2.0742311770943798,
      "grad_norm": 0.3520808219909668,
      "learning_rate": 3.8537936945703525e-05,
      "loss": 0.106,
      "num_input_tokens_seen": 24099056,
      "step": 12715
    },
    {
      "epoch": 2.075046904315197,
      "grad_norm": 2.0996320247650146,
      "learning_rate": 3.852968245227249e-05,
      "loss": 0.1978,
      "num_input_tokens_seen": 24108608,
      "step": 12720
    },
    {
      "epoch": 2.0758626315360145,
      "grad_norm": 0.8159396052360535,
      "learning_rate": 3.85214258723944e-05,
      "loss": 0.0996,
      "num_input_tokens_seen": 24117696,
      "step": 12725
    },
    {
      "epoch": 2.0766783587568316,
      "grad_norm": 0.7704966068267822,
      "learning_rate": 3.8513167207342524e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 24128272,
      "step": 12730
    },
    {
      "epoch": 2.077494085977649,
      "grad_norm": 0.18332770466804504,
      "learning_rate": 3.850490645839044e-05,
      "loss": 0.1567,
      "num_input_tokens_seen": 24137264,
      "step": 12735
    },
    {
      "epoch": 2.0783098131984663,
      "grad_norm": 0.8095189332962036,
      "learning_rate": 3.849664362681207e-05,
      "loss": 0.1097,
      "num_input_tokens_seen": 24145808,
      "step": 12740
    },
    {
      "epoch": 2.079125540419284,
      "grad_norm": 0.9668994545936584,
      "learning_rate": 3.848837871388165e-05,
      "loss": 0.2193,
      "num_input_tokens_seen": 24155888,
      "step": 12745
    },
    {
      "epoch": 2.079941267640101,
      "grad_norm": 0.690744161605835,
      "learning_rate": 3.848011172087371e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 24166784,
      "step": 12750
    },
    {
      "epoch": 2.0807569948609186,
      "grad_norm": 0.22344368696212769,
      "learning_rate": 3.847184264906315e-05,
      "loss": 0.0464,
      "num_input_tokens_seen": 24176176,
      "step": 12755
    },
    {
      "epoch": 2.0815727220817357,
      "grad_norm": 0.32777875661849976,
      "learning_rate": 3.846357149972516e-05,
      "loss": 0.0614,
      "num_input_tokens_seen": 24186112,
      "step": 12760
    },
    {
      "epoch": 2.0823884493025533,
      "grad_norm": 1.98373544216156,
      "learning_rate": 3.8455298274135246e-05,
      "loss": 0.2173,
      "num_input_tokens_seen": 24195344,
      "step": 12765
    },
    {
      "epoch": 2.0832041765233704,
      "grad_norm": 0.974223792552948,
      "learning_rate": 3.8447022973569254e-05,
      "loss": 0.1744,
      "num_input_tokens_seen": 24204528,
      "step": 12770
    },
    {
      "epoch": 2.084019903744188,
      "grad_norm": 0.18070819973945618,
      "learning_rate": 3.843874559930332e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 24214192,
      "step": 12775
    },
    {
      "epoch": 2.084835630965005,
      "grad_norm": 1.16152822971344,
      "learning_rate": 3.843046615261394e-05,
      "loss": 0.1724,
      "num_input_tokens_seen": 24224000,
      "step": 12780
    },
    {
      "epoch": 2.0856513581858227,
      "grad_norm": 0.5585319399833679,
      "learning_rate": 3.842218463477791e-05,
      "loss": 0.1909,
      "num_input_tokens_seen": 24233680,
      "step": 12785
    },
    {
      "epoch": 2.08646708540664,
      "grad_norm": 0.25848349928855896,
      "learning_rate": 3.841390104707233e-05,
      "loss": 0.2034,
      "num_input_tokens_seen": 24243680,
      "step": 12790
    },
    {
      "epoch": 2.0872828126274574,
      "grad_norm": 0.2814403772354126,
      "learning_rate": 3.8405615390774643e-05,
      "loss": 0.0794,
      "num_input_tokens_seen": 24253344,
      "step": 12795
    },
    {
      "epoch": 2.0880985398482745,
      "grad_norm": 0.9028207659721375,
      "learning_rate": 3.839732766716259e-05,
      "loss": 0.1655,
      "num_input_tokens_seen": 24261904,
      "step": 12800
    },
    {
      "epoch": 2.0880985398482745,
      "eval_loss": 0.15000221133232117,
      "eval_runtime": 133.3328,
      "eval_samples_per_second": 20.438,
      "eval_steps_per_second": 10.223,
      "num_input_tokens_seen": 24261904,
      "step": 12800
    },
    {
      "epoch": 2.088914267069092,
      "grad_norm": 0.3321162760257721,
      "learning_rate": 3.838903787751425e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 24270480,
      "step": 12805
    },
    {
      "epoch": 2.0897299942899092,
      "grad_norm": 1.75950288772583,
      "learning_rate": 3.838074602310802e-05,
      "loss": 0.1907,
      "num_input_tokens_seen": 24280288,
      "step": 12810
    },
    {
      "epoch": 2.090545721510727,
      "grad_norm": 0.09997115284204483,
      "learning_rate": 3.837245210522258e-05,
      "loss": 0.0187,
      "num_input_tokens_seen": 24289568,
      "step": 12815
    },
    {
      "epoch": 2.0913614487315444,
      "grad_norm": 1.0562739372253418,
      "learning_rate": 3.8364156125136996e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 24297888,
      "step": 12820
    },
    {
      "epoch": 2.0921771759523615,
      "grad_norm": 1.3004971742630005,
      "learning_rate": 3.835585808413059e-05,
      "loss": 0.1636,
      "num_input_tokens_seen": 24307168,
      "step": 12825
    },
    {
      "epoch": 2.092992903173179,
      "grad_norm": 1.4847761392593384,
      "learning_rate": 3.8347557983483024e-05,
      "loss": 0.086,
      "num_input_tokens_seen": 24317920,
      "step": 12830
    },
    {
      "epoch": 2.0938086303939962,
      "grad_norm": 0.25550681352615356,
      "learning_rate": 3.833925582447428e-05,
      "loss": 0.2484,
      "num_input_tokens_seen": 24327056,
      "step": 12835
    },
    {
      "epoch": 2.094624357614814,
      "grad_norm": 0.12472521513700485,
      "learning_rate": 3.8330951608384656e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 24336480,
      "step": 12840
    },
    {
      "epoch": 2.095440084835631,
      "grad_norm": 0.7019777894020081,
      "learning_rate": 3.832264533649477e-05,
      "loss": 0.1061,
      "num_input_tokens_seen": 24344560,
      "step": 12845
    },
    {
      "epoch": 2.0962558120564485,
      "grad_norm": 0.4205452799797058,
      "learning_rate": 3.8314337010085555e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 24354144,
      "step": 12850
    },
    {
      "epoch": 2.0970715392772656,
      "grad_norm": 0.739622175693512,
      "learning_rate": 3.830602663043824e-05,
      "loss": 0.1169,
      "num_input_tokens_seen": 24363904,
      "step": 12855
    },
    {
      "epoch": 2.097887266498083,
      "grad_norm": 0.14626221358776093,
      "learning_rate": 3.8297714198834414e-05,
      "loss": 0.0672,
      "num_input_tokens_seen": 24373680,
      "step": 12860
    },
    {
      "epoch": 2.0987029937189003,
      "grad_norm": 0.14593300223350525,
      "learning_rate": 3.828939971655595e-05,
      "loss": 0.1224,
      "num_input_tokens_seen": 24382048,
      "step": 12865
    },
    {
      "epoch": 2.099518720939718,
      "grad_norm": 0.1163533478975296,
      "learning_rate": 3.828108318488505e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 24392448,
      "step": 12870
    },
    {
      "epoch": 2.100334448160535,
      "grad_norm": 0.6173587441444397,
      "learning_rate": 3.8272764605104216e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 24402336,
      "step": 12875
    },
    {
      "epoch": 2.1011501753813526,
      "grad_norm": 0.5591193437576294,
      "learning_rate": 3.826444397849628e-05,
      "loss": 0.102,
      "num_input_tokens_seen": 24412064,
      "step": 12880
    },
    {
      "epoch": 2.1019659026021698,
      "grad_norm": 0.25799205899238586,
      "learning_rate": 3.825612130634439e-05,
      "loss": 0.0676,
      "num_input_tokens_seen": 24422528,
      "step": 12885
    },
    {
      "epoch": 2.1027816298229873,
      "grad_norm": 0.8034988641738892,
      "learning_rate": 3.824779658993202e-05,
      "loss": 0.0798,
      "num_input_tokens_seen": 24432192,
      "step": 12890
    },
    {
      "epoch": 2.1035973570438045,
      "grad_norm": 1.1411973237991333,
      "learning_rate": 3.823946983054292e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 24441936,
      "step": 12895
    },
    {
      "epoch": 2.104413084264622,
      "grad_norm": 0.281070351600647,
      "learning_rate": 3.82311410294612e-05,
      "loss": 0.0887,
      "num_input_tokens_seen": 24451504,
      "step": 12900
    },
    {
      "epoch": 2.105228811485439,
      "grad_norm": 1.347337007522583,
      "learning_rate": 3.822281018797127e-05,
      "loss": 0.0959,
      "num_input_tokens_seen": 24461392,
      "step": 12905
    },
    {
      "epoch": 2.1060445387062567,
      "grad_norm": 1.8458410501480103,
      "learning_rate": 3.821447730735783e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 24470496,
      "step": 12910
    },
    {
      "epoch": 2.106860265927074,
      "grad_norm": 1.6284563541412354,
      "learning_rate": 3.820614238890592e-05,
      "loss": 0.2655,
      "num_input_tokens_seen": 24481024,
      "step": 12915
    },
    {
      "epoch": 2.1076759931478914,
      "grad_norm": 0.3476162850856781,
      "learning_rate": 3.819780543390091e-05,
      "loss": 0.055,
      "num_input_tokens_seen": 24491424,
      "step": 12920
    },
    {
      "epoch": 2.1084917203687086,
      "grad_norm": 0.11284426599740982,
      "learning_rate": 3.818946644362844e-05,
      "loss": 0.1078,
      "num_input_tokens_seen": 24499936,
      "step": 12925
    },
    {
      "epoch": 2.109307447589526,
      "grad_norm": 0.20991742610931396,
      "learning_rate": 3.81811254193745e-05,
      "loss": 0.0305,
      "num_input_tokens_seen": 24510368,
      "step": 12930
    },
    {
      "epoch": 2.1101231748103433,
      "grad_norm": 0.5054029822349548,
      "learning_rate": 3.8172782362425366e-05,
      "loss": 0.3074,
      "num_input_tokens_seen": 24520144,
      "step": 12935
    },
    {
      "epoch": 2.110938902031161,
      "grad_norm": 0.46986138820648193,
      "learning_rate": 3.816443727406765e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 24529392,
      "step": 12940
    },
    {
      "epoch": 2.111754629251978,
      "grad_norm": 0.6901544332504272,
      "learning_rate": 3.815609015558829e-05,
      "loss": 0.0713,
      "num_input_tokens_seen": 24538976,
      "step": 12945
    },
    {
      "epoch": 2.1125703564727956,
      "grad_norm": 1.5201892852783203,
      "learning_rate": 3.814774100827448e-05,
      "loss": 0.3287,
      "num_input_tokens_seen": 24548240,
      "step": 12950
    },
    {
      "epoch": 2.1133860836936127,
      "grad_norm": 0.28573107719421387,
      "learning_rate": 3.813938983341379e-05,
      "loss": 0.0759,
      "num_input_tokens_seen": 24557584,
      "step": 12955
    },
    {
      "epoch": 2.1142018109144303,
      "grad_norm": 0.08489274978637695,
      "learning_rate": 3.813103663229407e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 24565808,
      "step": 12960
    },
    {
      "epoch": 2.1150175381352474,
      "grad_norm": 0.5237123370170593,
      "learning_rate": 3.812268140620349e-05,
      "loss": 0.098,
      "num_input_tokens_seen": 24575968,
      "step": 12965
    },
    {
      "epoch": 2.115833265356065,
      "grad_norm": 0.4465875029563904,
      "learning_rate": 3.811432415643051e-05,
      "loss": 0.0423,
      "num_input_tokens_seen": 24584800,
      "step": 12970
    },
    {
      "epoch": 2.116648992576882,
      "grad_norm": 1.2331349849700928,
      "learning_rate": 3.8105964884263954e-05,
      "loss": 0.2079,
      "num_input_tokens_seen": 24595968,
      "step": 12975
    },
    {
      "epoch": 2.1174647197976997,
      "grad_norm": 0.44598209857940674,
      "learning_rate": 3.809760359099291e-05,
      "loss": 0.2033,
      "num_input_tokens_seen": 24605168,
      "step": 12980
    },
    {
      "epoch": 2.118280447018517,
      "grad_norm": 0.03675892949104309,
      "learning_rate": 3.8089240277906804e-05,
      "loss": 0.0351,
      "num_input_tokens_seen": 24615216,
      "step": 12985
    },
    {
      "epoch": 2.1190961742393344,
      "grad_norm": 0.2389165312051773,
      "learning_rate": 3.808087494629535e-05,
      "loss": 0.1055,
      "num_input_tokens_seen": 24624848,
      "step": 12990
    },
    {
      "epoch": 2.1199119014601515,
      "grad_norm": 2.097132444381714,
      "learning_rate": 3.8072507597448595e-05,
      "loss": 0.1771,
      "num_input_tokens_seen": 24634128,
      "step": 12995
    },
    {
      "epoch": 2.120727628680969,
      "grad_norm": 1.4525368213653564,
      "learning_rate": 3.806413823265689e-05,
      "loss": 0.122,
      "num_input_tokens_seen": 24643776,
      "step": 13000
    },
    {
      "epoch": 2.120727628680969,
      "eval_loss": 0.15104860067367554,
      "eval_runtime": 133.3391,
      "eval_samples_per_second": 20.437,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 24643776,
      "step": 13000
    },
    {
      "epoch": 2.121543355901786,
      "grad_norm": 0.8096743226051331,
      "learning_rate": 3.805576685321089e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 24654064,
      "step": 13005
    },
    {
      "epoch": 2.122359083122604,
      "grad_norm": 1.337815523147583,
      "learning_rate": 3.804739346040158e-05,
      "loss": 0.1737,
      "num_input_tokens_seen": 24662112,
      "step": 13010
    },
    {
      "epoch": 2.1231748103434214,
      "grad_norm": 0.41173991560935974,
      "learning_rate": 3.8039018055520234e-05,
      "loss": 0.333,
      "num_input_tokens_seen": 24670944,
      "step": 13015
    },
    {
      "epoch": 2.1239905375642385,
      "grad_norm": 1.1019448041915894,
      "learning_rate": 3.803064063985844e-05,
      "loss": 0.1951,
      "num_input_tokens_seen": 24680608,
      "step": 13020
    },
    {
      "epoch": 2.124806264785056,
      "grad_norm": 0.5167211294174194,
      "learning_rate": 3.802226121470811e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 24690848,
      "step": 13025
    },
    {
      "epoch": 2.125621992005873,
      "grad_norm": 1.2394444942474365,
      "learning_rate": 3.801387978136145e-05,
      "loss": 0.2273,
      "num_input_tokens_seen": 24700656,
      "step": 13030
    },
    {
      "epoch": 2.126437719226691,
      "grad_norm": 0.8859125375747681,
      "learning_rate": 3.800549634111099e-05,
      "loss": 0.1012,
      "num_input_tokens_seen": 24710576,
      "step": 13035
    },
    {
      "epoch": 2.127253446447508,
      "grad_norm": 0.3816307783126831,
      "learning_rate": 3.799711089524955e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 24719168,
      "step": 13040
    },
    {
      "epoch": 2.1280691736683255,
      "grad_norm": 1.2952128648757935,
      "learning_rate": 3.7988723445070285e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 24728064,
      "step": 13045
    },
    {
      "epoch": 2.1288849008891426,
      "grad_norm": 0.7286532521247864,
      "learning_rate": 3.798033399186663e-05,
      "loss": 0.2155,
      "num_input_tokens_seen": 24738480,
      "step": 13050
    },
    {
      "epoch": 2.12970062810996,
      "grad_norm": 0.274900883436203,
      "learning_rate": 3.797194253693237e-05,
      "loss": 0.05,
      "num_input_tokens_seen": 24747840,
      "step": 13055
    },
    {
      "epoch": 2.1305163553307773,
      "grad_norm": 0.11377889662981033,
      "learning_rate": 3.796354908156153e-05,
      "loss": 0.1018,
      "num_input_tokens_seen": 24756752,
      "step": 13060
    },
    {
      "epoch": 2.131332082551595,
      "grad_norm": 0.6132127642631531,
      "learning_rate": 3.795515362704853e-05,
      "loss": 0.3935,
      "num_input_tokens_seen": 24766000,
      "step": 13065
    },
    {
      "epoch": 2.132147809772412,
      "grad_norm": 0.0737021341919899,
      "learning_rate": 3.794675617468803e-05,
      "loss": 0.1075,
      "num_input_tokens_seen": 24775728,
      "step": 13070
    },
    {
      "epoch": 2.1329635369932296,
      "grad_norm": 0.23649334907531738,
      "learning_rate": 3.793835672577503e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 24786176,
      "step": 13075
    },
    {
      "epoch": 2.1337792642140467,
      "grad_norm": 0.5383009314537048,
      "learning_rate": 3.7929955281604826e-05,
      "loss": 0.1811,
      "num_input_tokens_seen": 24794384,
      "step": 13080
    },
    {
      "epoch": 2.1345949914348643,
      "grad_norm": 0.5012834072113037,
      "learning_rate": 3.7921551843473036e-05,
      "loss": 0.0771,
      "num_input_tokens_seen": 24803008,
      "step": 13085
    },
    {
      "epoch": 2.1354107186556814,
      "grad_norm": 0.577995240688324,
      "learning_rate": 3.791314641267557e-05,
      "loss": 0.1838,
      "num_input_tokens_seen": 24813168,
      "step": 13090
    },
    {
      "epoch": 2.136226445876499,
      "grad_norm": 0.667745053768158,
      "learning_rate": 3.790473899050864e-05,
      "loss": 0.1876,
      "num_input_tokens_seen": 24821792,
      "step": 13095
    },
    {
      "epoch": 2.137042173097316,
      "grad_norm": 0.25428685545921326,
      "learning_rate": 3.7896329578268794e-05,
      "loss": 0.0718,
      "num_input_tokens_seen": 24831984,
      "step": 13100
    },
    {
      "epoch": 2.1378579003181337,
      "grad_norm": 1.077193260192871,
      "learning_rate": 3.7887918177252855e-05,
      "loss": 0.0797,
      "num_input_tokens_seen": 24841296,
      "step": 13105
    },
    {
      "epoch": 2.138673627538951,
      "grad_norm": 1.2379651069641113,
      "learning_rate": 3.787950478875798e-05,
      "loss": 0.2135,
      "num_input_tokens_seen": 24852208,
      "step": 13110
    },
    {
      "epoch": 2.1394893547597684,
      "grad_norm": 1.3670318126678467,
      "learning_rate": 3.787108941408162e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 24861264,
      "step": 13115
    },
    {
      "epoch": 2.1403050819805856,
      "grad_norm": 0.18435461819171906,
      "learning_rate": 3.786267205452151e-05,
      "loss": 0.0411,
      "num_input_tokens_seen": 24869856,
      "step": 13120
    },
    {
      "epoch": 2.141120809201403,
      "grad_norm": 0.10851617157459259,
      "learning_rate": 3.785425271137573e-05,
      "loss": 0.0492,
      "num_input_tokens_seen": 24878880,
      "step": 13125
    },
    {
      "epoch": 2.1419365364222203,
      "grad_norm": 1.159319281578064,
      "learning_rate": 3.7845831385942655e-05,
      "loss": 0.2141,
      "num_input_tokens_seen": 24889568,
      "step": 13130
    },
    {
      "epoch": 2.142752263643038,
      "grad_norm": 0.2230066955089569,
      "learning_rate": 3.7837408079520944e-05,
      "loss": 0.2109,
      "num_input_tokens_seen": 24898224,
      "step": 13135
    },
    {
      "epoch": 2.143567990863855,
      "grad_norm": 0.14697062969207764,
      "learning_rate": 3.782898279340957e-05,
      "loss": 0.1906,
      "num_input_tokens_seen": 24907024,
      "step": 13140
    },
    {
      "epoch": 2.1443837180846725,
      "grad_norm": 0.4862424433231354,
      "learning_rate": 3.782055552890784e-05,
      "loss": 0.0627,
      "num_input_tokens_seen": 24916480,
      "step": 13145
    },
    {
      "epoch": 2.1451994453054897,
      "grad_norm": 0.23166286945343018,
      "learning_rate": 3.781212628731534e-05,
      "loss": 0.0858,
      "num_input_tokens_seen": 24925328,
      "step": 13150
    },
    {
      "epoch": 2.1460151725263072,
      "grad_norm": 0.8454722762107849,
      "learning_rate": 3.7803695069931946e-05,
      "loss": 0.1951,
      "num_input_tokens_seen": 24935248,
      "step": 13155
    },
    {
      "epoch": 2.1468308997471244,
      "grad_norm": 0.8461867570877075,
      "learning_rate": 3.779526187805789e-05,
      "loss": 0.238,
      "num_input_tokens_seen": 24944272,
      "step": 13160
    },
    {
      "epoch": 2.147646626967942,
      "grad_norm": 1.3518034219741821,
      "learning_rate": 3.778682671299364e-05,
      "loss": 0.2406,
      "num_input_tokens_seen": 24953616,
      "step": 13165
    },
    {
      "epoch": 2.148462354188759,
      "grad_norm": 0.5143248438835144,
      "learning_rate": 3.777838957604003e-05,
      "loss": 0.2376,
      "num_input_tokens_seen": 24962352,
      "step": 13170
    },
    {
      "epoch": 2.1492780814095767,
      "grad_norm": 1.4846885204315186,
      "learning_rate": 3.776995046849816e-05,
      "loss": 0.1866,
      "num_input_tokens_seen": 24971504,
      "step": 13175
    },
    {
      "epoch": 2.150093808630394,
      "grad_norm": 0.5620781779289246,
      "learning_rate": 3.776150939166945e-05,
      "loss": 0.1876,
      "num_input_tokens_seen": 24980976,
      "step": 13180
    },
    {
      "epoch": 2.1509095358512114,
      "grad_norm": 0.05396336689591408,
      "learning_rate": 3.775306634685562e-05,
      "loss": 0.1704,
      "num_input_tokens_seen": 24992096,
      "step": 13185
    },
    {
      "epoch": 2.151725263072029,
      "grad_norm": 0.6514242887496948,
      "learning_rate": 3.7744621335358696e-05,
      "loss": 0.2344,
      "num_input_tokens_seen": 25001408,
      "step": 13190
    },
    {
      "epoch": 2.152540990292846,
      "grad_norm": 1.1528143882751465,
      "learning_rate": 3.7736174358481e-05,
      "loss": 0.1247,
      "num_input_tokens_seen": 25011264,
      "step": 13195
    },
    {
      "epoch": 2.153356717513663,
      "grad_norm": 0.17083725333213806,
      "learning_rate": 3.7727725417525175e-05,
      "loss": 0.0303,
      "num_input_tokens_seen": 25020496,
      "step": 13200
    },
    {
      "epoch": 2.153356717513663,
      "eval_loss": 0.14681769907474518,
      "eval_runtime": 133.3204,
      "eval_samples_per_second": 20.439,
      "eval_steps_per_second": 10.223,
      "num_input_tokens_seen": 25020496,
      "step": 13200
    },
    {
      "epoch": 2.1541724447344808,
      "grad_norm": 0.932531476020813,
      "learning_rate": 3.771927451379414e-05,
      "loss": 0.102,
      "num_input_tokens_seen": 25031200,
      "step": 13205
    },
    {
      "epoch": 2.1549881719552983,
      "grad_norm": 0.13545961678028107,
      "learning_rate": 3.7710821648591135e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 25039680,
      "step": 13210
    },
    {
      "epoch": 2.1558038991761155,
      "grad_norm": 0.5360726118087769,
      "learning_rate": 3.7702366823219694e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 25048544,
      "step": 13215
    },
    {
      "epoch": 2.156619626396933,
      "grad_norm": 0.03951575607061386,
      "learning_rate": 3.769391003898366e-05,
      "loss": 0.1114,
      "num_input_tokens_seen": 25057408,
      "step": 13220
    },
    {
      "epoch": 2.15743535361775,
      "grad_norm": 0.5895017385482788,
      "learning_rate": 3.768545129718718e-05,
      "loss": 0.0901,
      "num_input_tokens_seen": 25068272,
      "step": 13225
    },
    {
      "epoch": 2.1582510808385678,
      "grad_norm": 1.7424765825271606,
      "learning_rate": 3.7676990599134686e-05,
      "loss": 0.1214,
      "num_input_tokens_seen": 25078032,
      "step": 13230
    },
    {
      "epoch": 2.159066808059385,
      "grad_norm": 2.1025187969207764,
      "learning_rate": 3.766852794613095e-05,
      "loss": 0.2672,
      "num_input_tokens_seen": 25087440,
      "step": 13235
    },
    {
      "epoch": 2.1598825352802025,
      "grad_norm": 0.4110572040081024,
      "learning_rate": 3.766006333948099e-05,
      "loss": 0.3726,
      "num_input_tokens_seen": 25096208,
      "step": 13240
    },
    {
      "epoch": 2.1606982625010196,
      "grad_norm": 0.5737015604972839,
      "learning_rate": 3.765159678049017e-05,
      "loss": 0.2002,
      "num_input_tokens_seen": 25106544,
      "step": 13245
    },
    {
      "epoch": 2.161513989721837,
      "grad_norm": 0.1011875793337822,
      "learning_rate": 3.7643128270464134e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 25116816,
      "step": 13250
    },
    {
      "epoch": 2.1623297169426543,
      "grad_norm": 0.4474388659000397,
      "learning_rate": 3.763465781070884e-05,
      "loss": 0.0922,
      "num_input_tokens_seen": 25125888,
      "step": 13255
    },
    {
      "epoch": 2.163145444163472,
      "grad_norm": 0.5625931620597839,
      "learning_rate": 3.762618540253052e-05,
      "loss": 0.2299,
      "num_input_tokens_seen": 25135520,
      "step": 13260
    },
    {
      "epoch": 2.163961171384289,
      "grad_norm": 0.3054049015045166,
      "learning_rate": 3.761771104723576e-05,
      "loss": 0.0667,
      "num_input_tokens_seen": 25144416,
      "step": 13265
    },
    {
      "epoch": 2.1647768986051066,
      "grad_norm": 0.2723044455051422,
      "learning_rate": 3.7609234746131386e-05,
      "loss": 0.1881,
      "num_input_tokens_seen": 25154032,
      "step": 13270
    },
    {
      "epoch": 2.1655926258259237,
      "grad_norm": 2.1287951469421387,
      "learning_rate": 3.7600756500524556e-05,
      "loss": 0.1991,
      "num_input_tokens_seen": 25163024,
      "step": 13275
    },
    {
      "epoch": 2.1664083530467413,
      "grad_norm": 1.0394567251205444,
      "learning_rate": 3.759227631172271e-05,
      "loss": 0.0959,
      "num_input_tokens_seen": 25171664,
      "step": 13280
    },
    {
      "epoch": 2.1672240802675584,
      "grad_norm": 1.1880689859390259,
      "learning_rate": 3.758379418103363e-05,
      "loss": 0.1936,
      "num_input_tokens_seen": 25180960,
      "step": 13285
    },
    {
      "epoch": 2.168039807488376,
      "grad_norm": 1.0492358207702637,
      "learning_rate": 3.757531010976534e-05,
      "loss": 0.2228,
      "num_input_tokens_seen": 25190064,
      "step": 13290
    },
    {
      "epoch": 2.168855534709193,
      "grad_norm": 1.5590232610702515,
      "learning_rate": 3.75668240992262e-05,
      "loss": 0.1996,
      "num_input_tokens_seen": 25199872,
      "step": 13295
    },
    {
      "epoch": 2.1696712619300107,
      "grad_norm": 0.20506535470485687,
      "learning_rate": 3.7558336150724865e-05,
      "loss": 0.0737,
      "num_input_tokens_seen": 25209680,
      "step": 13300
    },
    {
      "epoch": 2.170486989150828,
      "grad_norm": 1.1963785886764526,
      "learning_rate": 3.754984626557028e-05,
      "loss": 0.1201,
      "num_input_tokens_seen": 25219024,
      "step": 13305
    },
    {
      "epoch": 2.1713027163716454,
      "grad_norm": 0.5090208053588867,
      "learning_rate": 3.754135444507168e-05,
      "loss": 0.1023,
      "num_input_tokens_seen": 25229136,
      "step": 13310
    },
    {
      "epoch": 2.1721184435924625,
      "grad_norm": 0.26222550868988037,
      "learning_rate": 3.753286069053863e-05,
      "loss": 0.0914,
      "num_input_tokens_seen": 25238576,
      "step": 13315
    },
    {
      "epoch": 2.17293417081328,
      "grad_norm": 1.058225393295288,
      "learning_rate": 3.7524365003280945e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 25247568,
      "step": 13320
    },
    {
      "epoch": 2.1737498980340972,
      "grad_norm": 1.1456215381622314,
      "learning_rate": 3.75158673846088e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 25254896,
      "step": 13325
    },
    {
      "epoch": 2.174565625254915,
      "grad_norm": 0.20018868148326874,
      "learning_rate": 3.750736783583262e-05,
      "loss": 0.0541,
      "num_input_tokens_seen": 25263216,
      "step": 13330
    },
    {
      "epoch": 2.175381352475732,
      "grad_norm": 0.03368765115737915,
      "learning_rate": 3.7498866358263144e-05,
      "loss": 0.1012,
      "num_input_tokens_seen": 25273072,
      "step": 13335
    },
    {
      "epoch": 2.1761970796965495,
      "grad_norm": 0.3003321588039398,
      "learning_rate": 3.74903629532114e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 25282544,
      "step": 13340
    },
    {
      "epoch": 2.1770128069173666,
      "grad_norm": 0.17274591326713562,
      "learning_rate": 3.748185762198873e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 25291136,
      "step": 13345
    },
    {
      "epoch": 2.1778285341381842,
      "grad_norm": 1.3319238424301147,
      "learning_rate": 3.747335036590676e-05,
      "loss": 0.1006,
      "num_input_tokens_seen": 25299984,
      "step": 13350
    },
    {
      "epoch": 2.1786442613590014,
      "grad_norm": 1.318505883216858,
      "learning_rate": 3.7464841186277405e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 25308896,
      "step": 13355
    },
    {
      "epoch": 2.179459988579819,
      "grad_norm": 1.309870719909668,
      "learning_rate": 3.7456330084412896e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 25317872,
      "step": 13360
    },
    {
      "epoch": 2.180275715800636,
      "grad_norm": 0.4181438386440277,
      "learning_rate": 3.744781706162576e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 25326112,
      "step": 13365
    },
    {
      "epoch": 2.1810914430214536,
      "grad_norm": 0.4524760842323303,
      "learning_rate": 3.743930211922879e-05,
      "loss": 0.3045,
      "num_input_tokens_seen": 25336624,
      "step": 13370
    },
    {
      "epoch": 2.1819071702422708,
      "grad_norm": 0.3053187131881714,
      "learning_rate": 3.743078525853513e-05,
      "loss": 0.1141,
      "num_input_tokens_seen": 25346720,
      "step": 13375
    },
    {
      "epoch": 2.1827228974630883,
      "grad_norm": 0.8193637132644653,
      "learning_rate": 3.7422266480858154e-05,
      "loss": 0.0658,
      "num_input_tokens_seen": 25355392,
      "step": 13380
    },
    {
      "epoch": 2.183538624683906,
      "grad_norm": 0.5557044148445129,
      "learning_rate": 3.741374578751158e-05,
      "loss": 0.19,
      "num_input_tokens_seen": 25364736,
      "step": 13385
    },
    {
      "epoch": 2.184354351904723,
      "grad_norm": 1.4275974035263062,
      "learning_rate": 3.740522317980941e-05,
      "loss": 0.1174,
      "num_input_tokens_seen": 25372192,
      "step": 13390
    },
    {
      "epoch": 2.1851700791255406,
      "grad_norm": 1.428638219833374,
      "learning_rate": 3.739669865906593e-05,
      "loss": 0.1418,
      "num_input_tokens_seen": 25382064,
      "step": 13395
    },
    {
      "epoch": 2.1859858063463578,
      "grad_norm": 0.06451517343521118,
      "learning_rate": 3.738817222659573e-05,
      "loss": 0.1037,
      "num_input_tokens_seen": 25391072,
      "step": 13400
    },
    {
      "epoch": 2.1859858063463578,
      "eval_loss": 0.14517082273960114,
      "eval_runtime": 133.3415,
      "eval_samples_per_second": 20.436,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 25391072,
      "step": 13400
    },
    {
      "epoch": 2.1868015335671753,
      "grad_norm": 0.4965091049671173,
      "learning_rate": 3.73796438837137e-05,
      "loss": 0.2426,
      "num_input_tokens_seen": 25400688,
      "step": 13405
    },
    {
      "epoch": 2.1876172607879925,
      "grad_norm": 0.2081955224275589,
      "learning_rate": 3.7371113631735e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 25410768,
      "step": 13410
    },
    {
      "epoch": 2.18843298800881,
      "grad_norm": 0.5181936621665955,
      "learning_rate": 3.736258147197512e-05,
      "loss": 0.067,
      "num_input_tokens_seen": 25420576,
      "step": 13415
    },
    {
      "epoch": 2.189248715229627,
      "grad_norm": 0.8011373281478882,
      "learning_rate": 3.735404740574981e-05,
      "loss": 0.1253,
      "num_input_tokens_seen": 25428688,
      "step": 13420
    },
    {
      "epoch": 2.1900644424504447,
      "grad_norm": 0.7960135340690613,
      "learning_rate": 3.7345511434375145e-05,
      "loss": 0.3062,
      "num_input_tokens_seen": 25438112,
      "step": 13425
    },
    {
      "epoch": 2.190880169671262,
      "grad_norm": 0.3516843914985657,
      "learning_rate": 3.733697355916748e-05,
      "loss": 0.1184,
      "num_input_tokens_seen": 25447024,
      "step": 13430
    },
    {
      "epoch": 2.1916958968920794,
      "grad_norm": 0.5937644243240356,
      "learning_rate": 3.732843378144345e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 25455824,
      "step": 13435
    },
    {
      "epoch": 2.1925116241128966,
      "grad_norm": 0.6947759389877319,
      "learning_rate": 3.7319892102519995e-05,
      "loss": 0.1773,
      "num_input_tokens_seen": 25464688,
      "step": 13440
    },
    {
      "epoch": 2.193327351333714,
      "grad_norm": 0.4027089774608612,
      "learning_rate": 3.731134852371436e-05,
      "loss": 0.1663,
      "num_input_tokens_seen": 25474480,
      "step": 13445
    },
    {
      "epoch": 2.1941430785545313,
      "grad_norm": 1.277462124824524,
      "learning_rate": 3.730280304634408e-05,
      "loss": 0.2428,
      "num_input_tokens_seen": 25484384,
      "step": 13450
    },
    {
      "epoch": 2.194958805775349,
      "grad_norm": 0.9289950132369995,
      "learning_rate": 3.729425567172696e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 25495104,
      "step": 13455
    },
    {
      "epoch": 2.195774532996166,
      "grad_norm": 0.557313084602356,
      "learning_rate": 3.728570640118111e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 25504800,
      "step": 13460
    },
    {
      "epoch": 2.1965902602169836,
      "grad_norm": 0.449586421251297,
      "learning_rate": 3.727715523602494e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 25513760,
      "step": 13465
    },
    {
      "epoch": 2.1974059874378007,
      "grad_norm": 0.11417495459318161,
      "learning_rate": 3.726860217757715e-05,
      "loss": 0.0471,
      "num_input_tokens_seen": 25523808,
      "step": 13470
    },
    {
      "epoch": 2.1982217146586183,
      "grad_norm": 1.4303004741668701,
      "learning_rate": 3.726004722715673e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 25531920,
      "step": 13475
    },
    {
      "epoch": 2.1990374418794354,
      "grad_norm": 0.10504009574651718,
      "learning_rate": 3.725149038608296e-05,
      "loss": 0.0584,
      "num_input_tokens_seen": 25541696,
      "step": 13480
    },
    {
      "epoch": 2.199853169100253,
      "grad_norm": 0.3883955180644989,
      "learning_rate": 3.7242931655675404e-05,
      "loss": 0.0905,
      "num_input_tokens_seen": 25551184,
      "step": 13485
    },
    {
      "epoch": 2.20066889632107,
      "grad_norm": 1.5016086101531982,
      "learning_rate": 3.7234371037253937e-05,
      "loss": 0.1748,
      "num_input_tokens_seen": 25559840,
      "step": 13490
    },
    {
      "epoch": 2.2014846235418877,
      "grad_norm": 0.7107875943183899,
      "learning_rate": 3.7225808532138705e-05,
      "loss": 0.209,
      "num_input_tokens_seen": 25568448,
      "step": 13495
    },
    {
      "epoch": 2.202300350762705,
      "grad_norm": 1.0991337299346924,
      "learning_rate": 3.721724414165016e-05,
      "loss": 0.2067,
      "num_input_tokens_seen": 25578352,
      "step": 13500
    },
    {
      "epoch": 2.2031160779835224,
      "grad_norm": 0.4757579267024994,
      "learning_rate": 3.720867786710904e-05,
      "loss": 0.1512,
      "num_input_tokens_seen": 25587360,
      "step": 13505
    },
    {
      "epoch": 2.2039318052043395,
      "grad_norm": 0.23639096319675446,
      "learning_rate": 3.7200109709836366e-05,
      "loss": 0.059,
      "num_input_tokens_seen": 25596112,
      "step": 13510
    },
    {
      "epoch": 2.204747532425157,
      "grad_norm": 0.6382490396499634,
      "learning_rate": 3.7191539671153465e-05,
      "loss": 0.1022,
      "num_input_tokens_seen": 25606336,
      "step": 13515
    },
    {
      "epoch": 2.205563259645974,
      "grad_norm": 1.578553557395935,
      "learning_rate": 3.718296775238193e-05,
      "loss": 0.1864,
      "num_input_tokens_seen": 25615248,
      "step": 13520
    },
    {
      "epoch": 2.206378986866792,
      "grad_norm": 0.20534121990203857,
      "learning_rate": 3.7174393954843675e-05,
      "loss": 0.0998,
      "num_input_tokens_seen": 25624064,
      "step": 13525
    },
    {
      "epoch": 2.207194714087609,
      "grad_norm": 0.5584843754768372,
      "learning_rate": 3.716581827986087e-05,
      "loss": 0.0483,
      "num_input_tokens_seen": 25633472,
      "step": 13530
    },
    {
      "epoch": 2.2080104413084265,
      "grad_norm": 0.19583763182163239,
      "learning_rate": 3.7157240728756004e-05,
      "loss": 0.2357,
      "num_input_tokens_seen": 25642240,
      "step": 13535
    },
    {
      "epoch": 2.2088261685292436,
      "grad_norm": 1.4709134101867676,
      "learning_rate": 3.714866130285184e-05,
      "loss": 0.1271,
      "num_input_tokens_seen": 25651680,
      "step": 13540
    },
    {
      "epoch": 2.209641895750061,
      "grad_norm": 0.6087249517440796,
      "learning_rate": 3.714008000347143e-05,
      "loss": 0.1773,
      "num_input_tokens_seen": 25660512,
      "step": 13545
    },
    {
      "epoch": 2.2104576229708783,
      "grad_norm": 2.5353994369506836,
      "learning_rate": 3.7131496831938126e-05,
      "loss": 0.4403,
      "num_input_tokens_seen": 25670144,
      "step": 13550
    },
    {
      "epoch": 2.211273350191696,
      "grad_norm": 2.220362901687622,
      "learning_rate": 3.7122911789575565e-05,
      "loss": 0.1746,
      "num_input_tokens_seen": 25678352,
      "step": 13555
    },
    {
      "epoch": 2.2120890774125135,
      "grad_norm": 0.48734021186828613,
      "learning_rate": 3.711432487770765e-05,
      "loss": 0.1048,
      "num_input_tokens_seen": 25688112,
      "step": 13560
    },
    {
      "epoch": 2.2129048046333306,
      "grad_norm": 0.3495413661003113,
      "learning_rate": 3.710573609765861e-05,
      "loss": 0.1067,
      "num_input_tokens_seen": 25697680,
      "step": 13565
    },
    {
      "epoch": 2.2137205318541477,
      "grad_norm": 0.3346834182739258,
      "learning_rate": 3.709714545075292e-05,
      "loss": 0.2286,
      "num_input_tokens_seen": 25706496,
      "step": 13570
    },
    {
      "epoch": 2.2145362590749653,
      "grad_norm": 1.2942469120025635,
      "learning_rate": 3.708855293831538e-05,
      "loss": 0.1057,
      "num_input_tokens_seen": 25715888,
      "step": 13575
    },
    {
      "epoch": 2.215351986295783,
      "grad_norm": 0.22794321179389954,
      "learning_rate": 3.707995856167107e-05,
      "loss": 0.1253,
      "num_input_tokens_seen": 25724448,
      "step": 13580
    },
    {
      "epoch": 2.2161677135166,
      "grad_norm": 0.2487674206495285,
      "learning_rate": 3.707136232214534e-05,
      "loss": 0.0939,
      "num_input_tokens_seen": 25733952,
      "step": 13585
    },
    {
      "epoch": 2.2169834407374176,
      "grad_norm": 0.7541723847389221,
      "learning_rate": 3.7062764221063844e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 25743152,
      "step": 13590
    },
    {
      "epoch": 2.2177991679582347,
      "grad_norm": 0.10922252386808395,
      "learning_rate": 3.705416425975252e-05,
      "loss": 0.0664,
      "num_input_tokens_seen": 25752784,
      "step": 13595
    },
    {
      "epoch": 2.2186148951790523,
      "grad_norm": 0.13461346924304962,
      "learning_rate": 3.704556243953758e-05,
      "loss": 0.0869,
      "num_input_tokens_seen": 25762416,
      "step": 13600
    },
    {
      "epoch": 2.2186148951790523,
      "eval_loss": 0.14976383745670319,
      "eval_runtime": 133.2572,
      "eval_samples_per_second": 20.449,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 25762416,
      "step": 13600
    },
    {
      "epoch": 2.2194306223998694,
      "grad_norm": 0.23535776138305664,
      "learning_rate": 3.7036958761745535e-05,
      "loss": 0.1086,
      "num_input_tokens_seen": 25770752,
      "step": 13605
    },
    {
      "epoch": 2.220246349620687,
      "grad_norm": 0.37730610370635986,
      "learning_rate": 3.702835322770318e-05,
      "loss": 0.063,
      "num_input_tokens_seen": 25779728,
      "step": 13610
    },
    {
      "epoch": 2.221062076841504,
      "grad_norm": 0.2914361357688904,
      "learning_rate": 3.701974583873761e-05,
      "loss": 0.0725,
      "num_input_tokens_seen": 25788496,
      "step": 13615
    },
    {
      "epoch": 2.2218778040623217,
      "grad_norm": 1.794702410697937,
      "learning_rate": 3.701113659617618e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 25799280,
      "step": 13620
    },
    {
      "epoch": 2.222693531283139,
      "grad_norm": 0.14908650517463684,
      "learning_rate": 3.7002525501346535e-05,
      "loss": 0.0474,
      "num_input_tokens_seen": 25808208,
      "step": 13625
    },
    {
      "epoch": 2.2235092585039564,
      "grad_norm": 0.7580570578575134,
      "learning_rate": 3.699391255557664e-05,
      "loss": 0.088,
      "num_input_tokens_seen": 25818000,
      "step": 13630
    },
    {
      "epoch": 2.2243249857247736,
      "grad_norm": 1.376882791519165,
      "learning_rate": 3.69852977601947e-05,
      "loss": 0.1558,
      "num_input_tokens_seen": 25827200,
      "step": 13635
    },
    {
      "epoch": 2.225140712945591,
      "grad_norm": 1.2425175905227661,
      "learning_rate": 3.697668111652922e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 25836144,
      "step": 13640
    },
    {
      "epoch": 2.2259564401664083,
      "grad_norm": 0.13669025897979736,
      "learning_rate": 3.6968062625909005e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 25845696,
      "step": 13645
    },
    {
      "epoch": 2.226772167387226,
      "grad_norm": 0.6685404777526855,
      "learning_rate": 3.6959442289663135e-05,
      "loss": 0.0849,
      "num_input_tokens_seen": 25855408,
      "step": 13650
    },
    {
      "epoch": 2.227587894608043,
      "grad_norm": 0.518927276134491,
      "learning_rate": 3.695082010912098e-05,
      "loss": 0.1037,
      "num_input_tokens_seen": 25865456,
      "step": 13655
    },
    {
      "epoch": 2.2284036218288605,
      "grad_norm": 0.5278394818305969,
      "learning_rate": 3.694219608561217e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 25875088,
      "step": 13660
    },
    {
      "epoch": 2.2292193490496777,
      "grad_norm": 0.20291171967983246,
      "learning_rate": 3.693357022046665e-05,
      "loss": 0.0483,
      "num_input_tokens_seen": 25884288,
      "step": 13665
    },
    {
      "epoch": 2.2300350762704952,
      "grad_norm": 0.9787400364875793,
      "learning_rate": 3.6924942515014644e-05,
      "loss": 0.1777,
      "num_input_tokens_seen": 25895072,
      "step": 13670
    },
    {
      "epoch": 2.2308508034913124,
      "grad_norm": 1.340881586074829,
      "learning_rate": 3.691631297058664e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 25904480,
      "step": 13675
    },
    {
      "epoch": 2.23166653071213,
      "grad_norm": 0.27622953057289124,
      "learning_rate": 3.6907681588513424e-05,
      "loss": 0.0344,
      "num_input_tokens_seen": 25913648,
      "step": 13680
    },
    {
      "epoch": 2.232482257932947,
      "grad_norm": 0.3411341905593872,
      "learning_rate": 3.689904837012606e-05,
      "loss": 0.2668,
      "num_input_tokens_seen": 25923968,
      "step": 13685
    },
    {
      "epoch": 2.2332979851537647,
      "grad_norm": 0.300667941570282,
      "learning_rate": 3.689041331675591e-05,
      "loss": 0.0562,
      "num_input_tokens_seen": 25934256,
      "step": 13690
    },
    {
      "epoch": 2.234113712374582,
      "grad_norm": 0.8948947191238403,
      "learning_rate": 3.688177642973461e-05,
      "loss": 0.5041,
      "num_input_tokens_seen": 25943888,
      "step": 13695
    },
    {
      "epoch": 2.2349294395953994,
      "grad_norm": 1.0000871419906616,
      "learning_rate": 3.687313771039406e-05,
      "loss": 0.0982,
      "num_input_tokens_seen": 25953808,
      "step": 13700
    },
    {
      "epoch": 2.2357451668162165,
      "grad_norm": 1.0851593017578125,
      "learning_rate": 3.686449716006647e-05,
      "loss": 0.2211,
      "num_input_tokens_seen": 25962752,
      "step": 13705
    },
    {
      "epoch": 2.236560894037034,
      "grad_norm": 0.20606446266174316,
      "learning_rate": 3.685585478008432e-05,
      "loss": 0.123,
      "num_input_tokens_seen": 25972288,
      "step": 13710
    },
    {
      "epoch": 2.237376621257851,
      "grad_norm": 0.25374624133110046,
      "learning_rate": 3.6847210571780364e-05,
      "loss": 0.12,
      "num_input_tokens_seen": 25981504,
      "step": 13715
    },
    {
      "epoch": 2.2381923484786688,
      "grad_norm": 2.4190444946289062,
      "learning_rate": 3.683856453648767e-05,
      "loss": 0.1954,
      "num_input_tokens_seen": 25989984,
      "step": 13720
    },
    {
      "epoch": 2.239008075699486,
      "grad_norm": 0.09228308498859406,
      "learning_rate": 3.682991667553954e-05,
      "loss": 0.0772,
      "num_input_tokens_seen": 25999920,
      "step": 13725
    },
    {
      "epoch": 2.2398238029203035,
      "grad_norm": 1.1736942529678345,
      "learning_rate": 3.6821266990269606e-05,
      "loss": 0.3073,
      "num_input_tokens_seen": 26008528,
      "step": 13730
    },
    {
      "epoch": 2.2406395301411206,
      "grad_norm": 0.26634854078292847,
      "learning_rate": 3.681261548201174e-05,
      "loss": 0.1219,
      "num_input_tokens_seen": 26018800,
      "step": 13735
    },
    {
      "epoch": 2.241455257361938,
      "grad_norm": 0.8922498226165771,
      "learning_rate": 3.6803962152100125e-05,
      "loss": 0.0951,
      "num_input_tokens_seen": 26027072,
      "step": 13740
    },
    {
      "epoch": 2.2422709845827553,
      "grad_norm": 0.5835890173912048,
      "learning_rate": 3.67953070018692e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 26037584,
      "step": 13745
    },
    {
      "epoch": 2.243086711803573,
      "grad_norm": 0.2970099151134491,
      "learning_rate": 3.678665003265371e-05,
      "loss": 0.0871,
      "num_input_tokens_seen": 26047440,
      "step": 13750
    },
    {
      "epoch": 2.2439024390243905,
      "grad_norm": 2.0164318084716797,
      "learning_rate": 3.677799124578867e-05,
      "loss": 0.2783,
      "num_input_tokens_seen": 26056896,
      "step": 13755
    },
    {
      "epoch": 2.2447181662452076,
      "grad_norm": 0.250767320394516,
      "learning_rate": 3.676933064260937e-05,
      "loss": 0.0253,
      "num_input_tokens_seen": 26067184,
      "step": 13760
    },
    {
      "epoch": 2.245533893466025,
      "grad_norm": 0.9932929873466492,
      "learning_rate": 3.6760668224451365e-05,
      "loss": 0.2345,
      "num_input_tokens_seen": 26075952,
      "step": 13765
    },
    {
      "epoch": 2.2463496206868423,
      "grad_norm": 0.4965848922729492,
      "learning_rate": 3.675200399265054e-05,
      "loss": 0.1059,
      "num_input_tokens_seen": 26084656,
      "step": 13770
    },
    {
      "epoch": 2.24716534790766,
      "grad_norm": 0.5446244478225708,
      "learning_rate": 3.6743337948543014e-05,
      "loss": 0.0618,
      "num_input_tokens_seen": 26094496,
      "step": 13775
    },
    {
      "epoch": 2.247981075128477,
      "grad_norm": 0.0797429010272026,
      "learning_rate": 3.6734670093465204e-05,
      "loss": 0.0491,
      "num_input_tokens_seen": 26104112,
      "step": 13780
    },
    {
      "epoch": 2.2487968023492946,
      "grad_norm": 0.20067888498306274,
      "learning_rate": 3.672600042875379e-05,
      "loss": 0.2771,
      "num_input_tokens_seen": 26112032,
      "step": 13785
    },
    {
      "epoch": 2.2496125295701117,
      "grad_norm": 1.066284418106079,
      "learning_rate": 3.671732895574575e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 26120720,
      "step": 13790
    },
    {
      "epoch": 2.2504282567909293,
      "grad_norm": 1.373171329498291,
      "learning_rate": 3.670865567577834e-05,
      "loss": 0.2014,
      "num_input_tokens_seen": 26129328,
      "step": 13795
    },
    {
      "epoch": 2.2512439840117464,
      "grad_norm": 0.6231725811958313,
      "learning_rate": 3.669998059018909e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 26139456,
      "step": 13800
    },
    {
      "epoch": 2.2512439840117464,
      "eval_loss": 0.14745649695396423,
      "eval_runtime": 133.2881,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 26139456,
      "step": 13800
    },
    {
      "epoch": 2.252059711232564,
      "grad_norm": 1.2746206521987915,
      "learning_rate": 3.6691303700315796e-05,
      "loss": 0.221,
      "num_input_tokens_seen": 26147936,
      "step": 13805
    },
    {
      "epoch": 2.252875438453381,
      "grad_norm": 0.47394275665283203,
      "learning_rate": 3.668262500749655e-05,
      "loss": 0.0982,
      "num_input_tokens_seen": 26156432,
      "step": 13810
    },
    {
      "epoch": 2.2536911656741987,
      "grad_norm": 1.162774920463562,
      "learning_rate": 3.667394451306971e-05,
      "loss": 0.2538,
      "num_input_tokens_seen": 26165632,
      "step": 13815
    },
    {
      "epoch": 2.254506892895016,
      "grad_norm": 0.4863773584365845,
      "learning_rate": 3.666526221837393e-05,
      "loss": 0.0831,
      "num_input_tokens_seen": 26175168,
      "step": 13820
    },
    {
      "epoch": 2.2553226201158334,
      "grad_norm": 0.3678531348705292,
      "learning_rate": 3.665657812474812e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 26184832,
      "step": 13825
    },
    {
      "epoch": 2.2561383473366505,
      "grad_norm": 0.5475358366966248,
      "learning_rate": 3.664789223353147e-05,
      "loss": 0.0963,
      "num_input_tokens_seen": 26193616,
      "step": 13830
    },
    {
      "epoch": 2.256954074557468,
      "grad_norm": 0.24156861007213593,
      "learning_rate": 3.663920454606347e-05,
      "loss": 0.2237,
      "num_input_tokens_seen": 26204208,
      "step": 13835
    },
    {
      "epoch": 2.2577698017782852,
      "grad_norm": 0.9669285416603088,
      "learning_rate": 3.6630515063683856e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 26212176,
      "step": 13840
    },
    {
      "epoch": 2.258585528999103,
      "grad_norm": 0.061852164566516876,
      "learning_rate": 3.662182378773267e-05,
      "loss": 0.1977,
      "num_input_tokens_seen": 26222064,
      "step": 13845
    },
    {
      "epoch": 2.25940125621992,
      "grad_norm": 1.0422923564910889,
      "learning_rate": 3.66131307195502e-05,
      "loss": 0.1985,
      "num_input_tokens_seen": 26231392,
      "step": 13850
    },
    {
      "epoch": 2.2602169834407375,
      "grad_norm": 1.1268640756607056,
      "learning_rate": 3.6604435860477034e-05,
      "loss": 0.1146,
      "num_input_tokens_seen": 26240224,
      "step": 13855
    },
    {
      "epoch": 2.2610327106615546,
      "grad_norm": 0.6707919836044312,
      "learning_rate": 3.6595739211854025e-05,
      "loss": 0.0899,
      "num_input_tokens_seen": 26251248,
      "step": 13860
    },
    {
      "epoch": 2.261848437882372,
      "grad_norm": 0.4249420166015625,
      "learning_rate": 3.658704077502231e-05,
      "loss": 0.0883,
      "num_input_tokens_seen": 26259344,
      "step": 13865
    },
    {
      "epoch": 2.2626641651031894,
      "grad_norm": 1.1765098571777344,
      "learning_rate": 3.65783405513233e-05,
      "loss": 0.0736,
      "num_input_tokens_seen": 26268288,
      "step": 13870
    },
    {
      "epoch": 2.263479892324007,
      "grad_norm": 1.949059009552002,
      "learning_rate": 3.656963854209867e-05,
      "loss": 0.2928,
      "num_input_tokens_seen": 26277488,
      "step": 13875
    },
    {
      "epoch": 2.264295619544824,
      "grad_norm": 1.3047930002212524,
      "learning_rate": 3.656093474869038e-05,
      "loss": 0.1993,
      "num_input_tokens_seen": 26286976,
      "step": 13880
    },
    {
      "epoch": 2.2651113467656416,
      "grad_norm": 1.4323053359985352,
      "learning_rate": 3.655222917244068e-05,
      "loss": 0.1248,
      "num_input_tokens_seen": 26295984,
      "step": 13885
    },
    {
      "epoch": 2.2659270739864588,
      "grad_norm": 0.22467023134231567,
      "learning_rate": 3.6543521814692054e-05,
      "loss": 0.1114,
      "num_input_tokens_seen": 26304832,
      "step": 13890
    },
    {
      "epoch": 2.2667428012072763,
      "grad_norm": 1.6484431028366089,
      "learning_rate": 3.653481267678731e-05,
      "loss": 0.1895,
      "num_input_tokens_seen": 26313808,
      "step": 13895
    },
    {
      "epoch": 2.2675585284280935,
      "grad_norm": 0.18252551555633545,
      "learning_rate": 3.652610176006949e-05,
      "loss": 0.0778,
      "num_input_tokens_seen": 26322816,
      "step": 13900
    },
    {
      "epoch": 2.268374255648911,
      "grad_norm": 0.5190219879150391,
      "learning_rate": 3.6517389065881925e-05,
      "loss": 0.1103,
      "num_input_tokens_seen": 26331792,
      "step": 13905
    },
    {
      "epoch": 2.269189982869728,
      "grad_norm": 0.2508098781108856,
      "learning_rate": 3.650867459556824e-05,
      "loss": 0.2164,
      "num_input_tokens_seen": 26340944,
      "step": 13910
    },
    {
      "epoch": 2.2700057100905457,
      "grad_norm": 0.09321420639753342,
      "learning_rate": 3.64999583504723e-05,
      "loss": 0.3295,
      "num_input_tokens_seen": 26349216,
      "step": 13915
    },
    {
      "epoch": 2.270821437311363,
      "grad_norm": 0.949495255947113,
      "learning_rate": 3.649124033193827e-05,
      "loss": 0.0946,
      "num_input_tokens_seen": 26358624,
      "step": 13920
    },
    {
      "epoch": 2.2716371645321805,
      "grad_norm": 0.4738242030143738,
      "learning_rate": 3.648252054131057e-05,
      "loss": 0.0816,
      "num_input_tokens_seen": 26366928,
      "step": 13925
    },
    {
      "epoch": 2.272452891752998,
      "grad_norm": 0.1735081821680069,
      "learning_rate": 3.647379897993391e-05,
      "loss": 0.018,
      "num_input_tokens_seen": 26376192,
      "step": 13930
    },
    {
      "epoch": 2.273268618973815,
      "grad_norm": 0.786553144454956,
      "learning_rate": 3.646507564915325e-05,
      "loss": 0.2721,
      "num_input_tokens_seen": 26384768,
      "step": 13935
    },
    {
      "epoch": 2.2740843461946323,
      "grad_norm": 0.9405083656311035,
      "learning_rate": 3.645635055031385e-05,
      "loss": 0.1065,
      "num_input_tokens_seen": 26394848,
      "step": 13940
    },
    {
      "epoch": 2.27490007341545,
      "grad_norm": 0.4674779772758484,
      "learning_rate": 3.6447623684761224e-05,
      "loss": 0.1429,
      "num_input_tokens_seen": 26403936,
      "step": 13945
    },
    {
      "epoch": 2.2757158006362674,
      "grad_norm": 1.0882647037506104,
      "learning_rate": 3.643889505384117e-05,
      "loss": 0.087,
      "num_input_tokens_seen": 26414064,
      "step": 13950
    },
    {
      "epoch": 2.2765315278570846,
      "grad_norm": 0.8892838358879089,
      "learning_rate": 3.6430164658899744e-05,
      "loss": 0.1184,
      "num_input_tokens_seen": 26424224,
      "step": 13955
    },
    {
      "epoch": 2.2773472550779017,
      "grad_norm": 0.2943135201931,
      "learning_rate": 3.642143250128329e-05,
      "loss": 0.0467,
      "num_input_tokens_seen": 26434256,
      "step": 13960
    },
    {
      "epoch": 2.2781629822987193,
      "grad_norm": 0.7713640928268433,
      "learning_rate": 3.641269858233841e-05,
      "loss": 0.0946,
      "num_input_tokens_seen": 26444336,
      "step": 13965
    },
    {
      "epoch": 2.278978709519537,
      "grad_norm": 0.27982404828071594,
      "learning_rate": 3.640396290341199e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 26453952,
      "step": 13970
    },
    {
      "epoch": 2.279794436740354,
      "grad_norm": 0.5646967887878418,
      "learning_rate": 3.639522546585118e-05,
      "loss": 0.0846,
      "num_input_tokens_seen": 26463024,
      "step": 13975
    },
    {
      "epoch": 2.2806101639611716,
      "grad_norm": 0.8609535694122314,
      "learning_rate": 3.6386486271003404e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 26472128,
      "step": 13980
    },
    {
      "epoch": 2.2814258911819887,
      "grad_norm": 2.040665626525879,
      "learning_rate": 3.6377745320216346e-05,
      "loss": 0.3072,
      "num_input_tokens_seen": 26481312,
      "step": 13985
    },
    {
      "epoch": 2.2822416184028063,
      "grad_norm": 1.7408207654953003,
      "learning_rate": 3.636900261483798e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 26491104,
      "step": 13990
    },
    {
      "epoch": 2.2830573456236234,
      "grad_norm": 0.3286711871623993,
      "learning_rate": 3.636025815621654e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 26501584,
      "step": 13995
    },
    {
      "epoch": 2.283873072844441,
      "grad_norm": 0.6769561171531677,
      "learning_rate": 3.635151194570054e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 26511344,
      "step": 14000
    },
    {
      "epoch": 2.283873072844441,
      "eval_loss": 0.1471024453639984,
      "eval_runtime": 133.1277,
      "eval_samples_per_second": 20.469,
      "eval_steps_per_second": 10.238,
      "num_input_tokens_seen": 26511344,
      "step": 14000
    },
    {
      "epoch": 2.284688800065258,
      "grad_norm": 1.1334868669509888,
      "learning_rate": 3.634276398463873e-05,
      "loss": 0.1063,
      "num_input_tokens_seen": 26521392,
      "step": 14005
    },
    {
      "epoch": 2.2855045272860757,
      "grad_norm": 0.33257588744163513,
      "learning_rate": 3.633401427438018e-05,
      "loss": 0.0606,
      "num_input_tokens_seen": 26530896,
      "step": 14010
    },
    {
      "epoch": 2.286320254506893,
      "grad_norm": 0.34742599725723267,
      "learning_rate": 3.63252628162742e-05,
      "loss": 0.0783,
      "num_input_tokens_seen": 26539744,
      "step": 14015
    },
    {
      "epoch": 2.2871359817277104,
      "grad_norm": 0.20564277470111847,
      "learning_rate": 3.6316509611670364e-05,
      "loss": 0.1174,
      "num_input_tokens_seen": 26549824,
      "step": 14020
    },
    {
      "epoch": 2.2879517089485275,
      "grad_norm": 0.5728995203971863,
      "learning_rate": 3.630775466191854e-05,
      "loss": 0.1856,
      "num_input_tokens_seen": 26558320,
      "step": 14025
    },
    {
      "epoch": 2.288767436169345,
      "grad_norm": 0.5011151432991028,
      "learning_rate": 3.629899796836884e-05,
      "loss": 0.2123,
      "num_input_tokens_seen": 26567568,
      "step": 14030
    },
    {
      "epoch": 2.289583163390162,
      "grad_norm": 0.4148302674293518,
      "learning_rate": 3.6290239532371666e-05,
      "loss": 0.0838,
      "num_input_tokens_seen": 26577360,
      "step": 14035
    },
    {
      "epoch": 2.29039889061098,
      "grad_norm": 0.37017807364463806,
      "learning_rate": 3.628147935527767e-05,
      "loss": 0.0463,
      "num_input_tokens_seen": 26586592,
      "step": 14040
    },
    {
      "epoch": 2.291214617831797,
      "grad_norm": 0.21686522662639618,
      "learning_rate": 3.627271743843779e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 26595024,
      "step": 14045
    },
    {
      "epoch": 2.2920303450526145,
      "grad_norm": 0.5985280871391296,
      "learning_rate": 3.626395378320321e-05,
      "loss": 0.0572,
      "num_input_tokens_seen": 26604864,
      "step": 14050
    },
    {
      "epoch": 2.2928460722734316,
      "grad_norm": 0.11188516020774841,
      "learning_rate": 3.625518839092541e-05,
      "loss": 0.1695,
      "num_input_tokens_seen": 26613984,
      "step": 14055
    },
    {
      "epoch": 2.293661799494249,
      "grad_norm": 1.1824723482131958,
      "learning_rate": 3.624642126295612e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 26624128,
      "step": 14060
    },
    {
      "epoch": 2.2944775267150663,
      "grad_norm": 0.9373087286949158,
      "learning_rate": 3.6237652400647345e-05,
      "loss": 0.0729,
      "num_input_tokens_seen": 26633312,
      "step": 14065
    },
    {
      "epoch": 2.295293253935884,
      "grad_norm": 0.14584718644618988,
      "learning_rate": 3.622888180535134e-05,
      "loss": 0.069,
      "num_input_tokens_seen": 26642416,
      "step": 14070
    },
    {
      "epoch": 2.296108981156701,
      "grad_norm": 0.6828784346580505,
      "learning_rate": 3.6220109478420655e-05,
      "loss": 0.0629,
      "num_input_tokens_seen": 26652048,
      "step": 14075
    },
    {
      "epoch": 2.2969247083775186,
      "grad_norm": 0.5722581148147583,
      "learning_rate": 3.6211335421208084e-05,
      "loss": 0.1241,
      "num_input_tokens_seen": 26661904,
      "step": 14080
    },
    {
      "epoch": 2.2977404355983357,
      "grad_norm": 0.7344954609870911,
      "learning_rate": 3.62025596350667e-05,
      "loss": 0.0697,
      "num_input_tokens_seen": 26671392,
      "step": 14085
    },
    {
      "epoch": 2.2985561628191533,
      "grad_norm": 1.40638267993927,
      "learning_rate": 3.619378212134984e-05,
      "loss": 0.2429,
      "num_input_tokens_seen": 26681248,
      "step": 14090
    },
    {
      "epoch": 2.2993718900399704,
      "grad_norm": 1.3588850498199463,
      "learning_rate": 3.618500288141111e-05,
      "loss": 0.2627,
      "num_input_tokens_seen": 26691344,
      "step": 14095
    },
    {
      "epoch": 2.300187617260788,
      "grad_norm": 1.1647800207138062,
      "learning_rate": 3.617622191660438e-05,
      "loss": 0.1761,
      "num_input_tokens_seen": 26700432,
      "step": 14100
    },
    {
      "epoch": 2.3010033444816056,
      "grad_norm": 0.5976670980453491,
      "learning_rate": 3.616743922828377e-05,
      "loss": 0.0956,
      "num_input_tokens_seen": 26710112,
      "step": 14105
    },
    {
      "epoch": 2.3018190717024227,
      "grad_norm": 0.3004637062549591,
      "learning_rate": 3.615865481780371e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 26719040,
      "step": 14110
    },
    {
      "epoch": 2.30263479892324,
      "grad_norm": 0.3049620985984802,
      "learning_rate": 3.614986868651883e-05,
      "loss": 0.092,
      "num_input_tokens_seen": 26728640,
      "step": 14115
    },
    {
      "epoch": 2.3034505261440574,
      "grad_norm": 0.6904109120368958,
      "learning_rate": 3.614108083578409e-05,
      "loss": 0.0935,
      "num_input_tokens_seen": 26738720,
      "step": 14120
    },
    {
      "epoch": 2.304266253364875,
      "grad_norm": 0.9055516719818115,
      "learning_rate": 3.613229126695467e-05,
      "loss": 0.1738,
      "num_input_tokens_seen": 26747296,
      "step": 14125
    },
    {
      "epoch": 2.305081980585692,
      "grad_norm": 2.0668866634368896,
      "learning_rate": 3.612349998138605e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 26756368,
      "step": 14130
    },
    {
      "epoch": 2.3058977078065093,
      "grad_norm": 0.7807207703590393,
      "learning_rate": 3.6114706980433946e-05,
      "loss": 0.1886,
      "num_input_tokens_seen": 26765984,
      "step": 14135
    },
    {
      "epoch": 2.306713435027327,
      "grad_norm": 0.5971055030822754,
      "learning_rate": 3.610591226545435e-05,
      "loss": 0.1346,
      "num_input_tokens_seen": 26774768,
      "step": 14140
    },
    {
      "epoch": 2.3075291622481444,
      "grad_norm": 0.5072062015533447,
      "learning_rate": 3.6097115837803505e-05,
      "loss": 0.0713,
      "num_input_tokens_seen": 26784464,
      "step": 14145
    },
    {
      "epoch": 2.3083448894689615,
      "grad_norm": 0.27296483516693115,
      "learning_rate": 3.608831769883795e-05,
      "loss": 0.0358,
      "num_input_tokens_seen": 26794848,
      "step": 14150
    },
    {
      "epoch": 2.309160616689779,
      "grad_norm": 1.429369568824768,
      "learning_rate": 3.607951784991446e-05,
      "loss": 0.3421,
      "num_input_tokens_seen": 26803872,
      "step": 14155
    },
    {
      "epoch": 2.3099763439105963,
      "grad_norm": 0.5121952295303345,
      "learning_rate": 3.6070716292390085e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 26812544,
      "step": 14160
    },
    {
      "epoch": 2.310792071131414,
      "grad_norm": 0.059322819113731384,
      "learning_rate": 3.606191302762213e-05,
      "loss": 0.037,
      "num_input_tokens_seen": 26822608,
      "step": 14165
    },
    {
      "epoch": 2.311607798352231,
      "grad_norm": 0.34325703978538513,
      "learning_rate": 3.605310805696818e-05,
      "loss": 0.1007,
      "num_input_tokens_seen": 26831680,
      "step": 14170
    },
    {
      "epoch": 2.3124235255730485,
      "grad_norm": 2.0933101177215576,
      "learning_rate": 3.6044301381786067e-05,
      "loss": 0.1997,
      "num_input_tokens_seen": 26841264,
      "step": 14175
    },
    {
      "epoch": 2.3132392527938657,
      "grad_norm": 1.2871379852294922,
      "learning_rate": 3.6035493003433883e-05,
      "loss": 0.2177,
      "num_input_tokens_seen": 26852464,
      "step": 14180
    },
    {
      "epoch": 2.3140549800146832,
      "grad_norm": 0.47548404335975647,
      "learning_rate": 3.6026682923269994e-05,
      "loss": 0.1085,
      "num_input_tokens_seen": 26861536,
      "step": 14185
    },
    {
      "epoch": 2.3148707072355004,
      "grad_norm": 0.32365718483924866,
      "learning_rate": 3.6017871142653034e-05,
      "loss": 0.1165,
      "num_input_tokens_seen": 26871712,
      "step": 14190
    },
    {
      "epoch": 2.315686434456318,
      "grad_norm": 0.3578319847583771,
      "learning_rate": 3.600905766294189e-05,
      "loss": 0.056,
      "num_input_tokens_seen": 26882208,
      "step": 14195
    },
    {
      "epoch": 2.316502161677135,
      "grad_norm": 0.23740577697753906,
      "learning_rate": 3.60002424854957e-05,
      "loss": 0.097,
      "num_input_tokens_seen": 26891616,
      "step": 14200
    },
    {
      "epoch": 2.316502161677135,
      "eval_loss": 0.15101030468940735,
      "eval_runtime": 133.2941,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 26891616,
      "step": 14200
    },
    {
      "epoch": 2.3173178888979526,
      "grad_norm": 0.37190207839012146,
      "learning_rate": 3.5991425611673876e-05,
      "loss": 0.2045,
      "num_input_tokens_seen": 26902032,
      "step": 14205
    },
    {
      "epoch": 2.31813361611877,
      "grad_norm": 1.459755301475525,
      "learning_rate": 3.5982607042836105e-05,
      "loss": 0.2618,
      "num_input_tokens_seen": 26911664,
      "step": 14210
    },
    {
      "epoch": 2.3189493433395874,
      "grad_norm": 0.6110014915466309,
      "learning_rate": 3.597378678034231e-05,
      "loss": 0.1027,
      "num_input_tokens_seen": 26922320,
      "step": 14215
    },
    {
      "epoch": 2.3197650705604045,
      "grad_norm": 1.2597829103469849,
      "learning_rate": 3.596496482555269e-05,
      "loss": 0.1915,
      "num_input_tokens_seen": 26932640,
      "step": 14220
    },
    {
      "epoch": 2.320580797781222,
      "grad_norm": 0.5957769155502319,
      "learning_rate": 3.595614117982769e-05,
      "loss": 0.0632,
      "num_input_tokens_seen": 26942608,
      "step": 14225
    },
    {
      "epoch": 2.321396525002039,
      "grad_norm": 0.3873403072357178,
      "learning_rate": 3.594731584452805e-05,
      "loss": 0.2053,
      "num_input_tokens_seen": 26951264,
      "step": 14230
    },
    {
      "epoch": 2.3222122522228568,
      "grad_norm": 2.114093780517578,
      "learning_rate": 3.593848882101472e-05,
      "loss": 0.0795,
      "num_input_tokens_seen": 26960944,
      "step": 14235
    },
    {
      "epoch": 2.323027979443674,
      "grad_norm": 1.487625002861023,
      "learning_rate": 3.592966011064896e-05,
      "loss": 0.2175,
      "num_input_tokens_seen": 26969968,
      "step": 14240
    },
    {
      "epoch": 2.3238437066644915,
      "grad_norm": 1.1550779342651367,
      "learning_rate": 3.592082971479226e-05,
      "loss": 0.1838,
      "num_input_tokens_seen": 26979168,
      "step": 14245
    },
    {
      "epoch": 2.3246594338853086,
      "grad_norm": 0.8559977412223816,
      "learning_rate": 3.5911997634806385e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 26989680,
      "step": 14250
    },
    {
      "epoch": 2.325475161106126,
      "grad_norm": 1.4780409336090088,
      "learning_rate": 3.5903163872053336e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 26999824,
      "step": 14255
    },
    {
      "epoch": 2.3262908883269433,
      "grad_norm": 0.2530288100242615,
      "learning_rate": 3.58943284278954e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 27009216,
      "step": 14260
    },
    {
      "epoch": 2.327106615547761,
      "grad_norm": 0.9050304293632507,
      "learning_rate": 3.588549130369512e-05,
      "loss": 0.1028,
      "num_input_tokens_seen": 27016384,
      "step": 14265
    },
    {
      "epoch": 2.327922342768578,
      "grad_norm": 0.041228096932172775,
      "learning_rate": 3.5876652500815274e-05,
      "loss": 0.0637,
      "num_input_tokens_seen": 27026384,
      "step": 14270
    },
    {
      "epoch": 2.3287380699893956,
      "grad_norm": 1.0547832250595093,
      "learning_rate": 3.586781202061894e-05,
      "loss": 0.1809,
      "num_input_tokens_seen": 27035504,
      "step": 14275
    },
    {
      "epoch": 2.3295537972102127,
      "grad_norm": 0.49608150124549866,
      "learning_rate": 3.585896986446942e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 27045472,
      "step": 14280
    },
    {
      "epoch": 2.3303695244310303,
      "grad_norm": 0.7223042249679565,
      "learning_rate": 3.585012603373028e-05,
      "loss": 0.0908,
      "num_input_tokens_seen": 27054992,
      "step": 14285
    },
    {
      "epoch": 2.3311852516518474,
      "grad_norm": 0.44981691241264343,
      "learning_rate": 3.584128052976535e-05,
      "loss": 0.0424,
      "num_input_tokens_seen": 27065344,
      "step": 14290
    },
    {
      "epoch": 2.332000978872665,
      "grad_norm": 0.24312815070152283,
      "learning_rate": 3.5832433353938724e-05,
      "loss": 0.062,
      "num_input_tokens_seen": 27076160,
      "step": 14295
    },
    {
      "epoch": 2.3328167060934826,
      "grad_norm": 1.7305256128311157,
      "learning_rate": 3.5823584507614746e-05,
      "loss": 0.1945,
      "num_input_tokens_seen": 27084464,
      "step": 14300
    },
    {
      "epoch": 2.3336324333142997,
      "grad_norm": 0.24269910156726837,
      "learning_rate": 3.581473399215802e-05,
      "loss": 0.0856,
      "num_input_tokens_seen": 27093264,
      "step": 14305
    },
    {
      "epoch": 2.334448160535117,
      "grad_norm": 0.2519865930080414,
      "learning_rate": 3.580588180893341e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 27102208,
      "step": 14310
    },
    {
      "epoch": 2.3352638877559344,
      "grad_norm": 0.3831743001937866,
      "learning_rate": 3.579702795930602e-05,
      "loss": 0.0631,
      "num_input_tokens_seen": 27110592,
      "step": 14315
    },
    {
      "epoch": 2.336079614976752,
      "grad_norm": 0.1816895455121994,
      "learning_rate": 3.578817244464125e-05,
      "loss": 0.1126,
      "num_input_tokens_seen": 27120256,
      "step": 14320
    },
    {
      "epoch": 2.336895342197569,
      "grad_norm": 1.0048636198043823,
      "learning_rate": 3.577931526630471e-05,
      "loss": 0.2157,
      "num_input_tokens_seen": 27130080,
      "step": 14325
    },
    {
      "epoch": 2.3377110694183862,
      "grad_norm": 0.3992343842983246,
      "learning_rate": 3.577045642566229e-05,
      "loss": 0.0828,
      "num_input_tokens_seen": 27138816,
      "step": 14330
    },
    {
      "epoch": 2.338526796639204,
      "grad_norm": 0.03503255918622017,
      "learning_rate": 3.576159592408014e-05,
      "loss": 0.0448,
      "num_input_tokens_seen": 27148832,
      "step": 14335
    },
    {
      "epoch": 2.3393425238600214,
      "grad_norm": 0.139938622713089,
      "learning_rate": 3.575273376292466e-05,
      "loss": 0.0817,
      "num_input_tokens_seen": 27158304,
      "step": 14340
    },
    {
      "epoch": 2.3401582510808385,
      "grad_norm": 0.7618364095687866,
      "learning_rate": 3.574386994356251e-05,
      "loss": 0.2416,
      "num_input_tokens_seen": 27167712,
      "step": 14345
    },
    {
      "epoch": 2.340973978301656,
      "grad_norm": 0.0666903406381607,
      "learning_rate": 3.573500446736059e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 27176672,
      "step": 14350
    },
    {
      "epoch": 2.3417897055224732,
      "grad_norm": 1.0236196517944336,
      "learning_rate": 3.5726137335686094e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 27185648,
      "step": 14355
    },
    {
      "epoch": 2.342605432743291,
      "grad_norm": 1.8524394035339355,
      "learning_rate": 3.571726854990642e-05,
      "loss": 0.1778,
      "num_input_tokens_seen": 27196192,
      "step": 14360
    },
    {
      "epoch": 2.343421159964108,
      "grad_norm": 0.13411113619804382,
      "learning_rate": 3.570839811138925e-05,
      "loss": 0.2589,
      "num_input_tokens_seen": 27207152,
      "step": 14365
    },
    {
      "epoch": 2.3442368871849255,
      "grad_norm": 0.9901127219200134,
      "learning_rate": 3.569952602150252e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 27216752,
      "step": 14370
    },
    {
      "epoch": 2.3450526144057426,
      "grad_norm": 0.15658341348171234,
      "learning_rate": 3.569065228161442e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 27226752,
      "step": 14375
    },
    {
      "epoch": 2.34586834162656,
      "grad_norm": 0.11141233146190643,
      "learning_rate": 3.5681776893093395e-05,
      "loss": 0.0618,
      "num_input_tokens_seen": 27237488,
      "step": 14380
    },
    {
      "epoch": 2.3466840688473773,
      "grad_norm": 0.7814478874206543,
      "learning_rate": 3.5672899857308134e-05,
      "loss": 0.0972,
      "num_input_tokens_seen": 27247008,
      "step": 14385
    },
    {
      "epoch": 2.347499796068195,
      "grad_norm": 0.11458011716604233,
      "learning_rate": 3.566402117562759e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 27257408,
      "step": 14390
    },
    {
      "epoch": 2.348315523289012,
      "grad_norm": 2.076300621032715,
      "learning_rate": 3.565514084942097e-05,
      "loss": 0.183,
      "num_input_tokens_seen": 27266480,
      "step": 14395
    },
    {
      "epoch": 2.3491312505098296,
      "grad_norm": 1.4477626085281372,
      "learning_rate": 3.564625888005773e-05,
      "loss": 0.176,
      "num_input_tokens_seen": 27274960,
      "step": 14400
    },
    {
      "epoch": 2.3491312505098296,
      "eval_loss": 0.1493006944656372,
      "eval_runtime": 133.2948,
      "eval_samples_per_second": 20.443,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 27274960,
      "step": 14400
    },
    {
      "epoch": 2.3499469777306468,
      "grad_norm": 0.04457894712686539,
      "learning_rate": 3.563737526890759e-05,
      "loss": 0.0509,
      "num_input_tokens_seen": 27284032,
      "step": 14405
    },
    {
      "epoch": 2.3507627049514643,
      "grad_norm": 0.5299391746520996,
      "learning_rate": 3.562849001734049e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 27292848,
      "step": 14410
    },
    {
      "epoch": 2.3515784321722815,
      "grad_norm": 0.15896619856357574,
      "learning_rate": 3.561960312672667e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 27302336,
      "step": 14415
    },
    {
      "epoch": 2.352394159393099,
      "grad_norm": 0.7957951426506042,
      "learning_rate": 3.5610714598436596e-05,
      "loss": 0.0527,
      "num_input_tokens_seen": 27312384,
      "step": 14420
    },
    {
      "epoch": 2.353209886613916,
      "grad_norm": 0.12224186956882477,
      "learning_rate": 3.5601824433840986e-05,
      "loss": 0.0804,
      "num_input_tokens_seen": 27321360,
      "step": 14425
    },
    {
      "epoch": 2.3540256138347337,
      "grad_norm": 0.91254061460495,
      "learning_rate": 3.559293263431082e-05,
      "loss": 0.2372,
      "num_input_tokens_seen": 27331584,
      "step": 14430
    },
    {
      "epoch": 2.354841341055551,
      "grad_norm": 0.34561190009117126,
      "learning_rate": 3.558403920121732e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 27340208,
      "step": 14435
    },
    {
      "epoch": 2.3556570682763684,
      "grad_norm": 1.5531648397445679,
      "learning_rate": 3.557514413593197e-05,
      "loss": 0.1152,
      "num_input_tokens_seen": 27350544,
      "step": 14440
    },
    {
      "epoch": 2.3564727954971856,
      "grad_norm": 0.4340711236000061,
      "learning_rate": 3.55662474398265e-05,
      "loss": 0.0732,
      "num_input_tokens_seen": 27359680,
      "step": 14445
    },
    {
      "epoch": 2.357288522718003,
      "grad_norm": 1.5580966472625732,
      "learning_rate": 3.555734911427288e-05,
      "loss": 0.3004,
      "num_input_tokens_seen": 27367760,
      "step": 14450
    },
    {
      "epoch": 2.3581042499388203,
      "grad_norm": 1.5527336597442627,
      "learning_rate": 3.5548449160643363e-05,
      "loss": 0.1991,
      "num_input_tokens_seen": 27378272,
      "step": 14455
    },
    {
      "epoch": 2.358919977159638,
      "grad_norm": 0.18229132890701294,
      "learning_rate": 3.553954758031043e-05,
      "loss": 0.0556,
      "num_input_tokens_seen": 27388880,
      "step": 14460
    },
    {
      "epoch": 2.359735704380455,
      "grad_norm": 0.30971747636795044,
      "learning_rate": 3.5530644374646815e-05,
      "loss": 0.0871,
      "num_input_tokens_seen": 27398656,
      "step": 14465
    },
    {
      "epoch": 2.3605514316012726,
      "grad_norm": 1.1232928037643433,
      "learning_rate": 3.552173954502549e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 27407968,
      "step": 14470
    },
    {
      "epoch": 2.36136715882209,
      "grad_norm": 0.2527836561203003,
      "learning_rate": 3.55128330928197e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 27416544,
      "step": 14475
    },
    {
      "epoch": 2.3621828860429073,
      "grad_norm": 0.14198337495326996,
      "learning_rate": 3.550392501940294e-05,
      "loss": 0.269,
      "num_input_tokens_seen": 27426608,
      "step": 14480
    },
    {
      "epoch": 2.3629986132637244,
      "grad_norm": 0.12551409006118774,
      "learning_rate": 3.5495015326148945e-05,
      "loss": 0.2038,
      "num_input_tokens_seen": 27436768,
      "step": 14485
    },
    {
      "epoch": 2.363814340484542,
      "grad_norm": 0.3598979711532593,
      "learning_rate": 3.548610401443169e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 27444896,
      "step": 14490
    },
    {
      "epoch": 2.3646300677053596,
      "grad_norm": 0.8140979409217834,
      "learning_rate": 3.547719108562543e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 27454768,
      "step": 14495
    },
    {
      "epoch": 2.3654457949261767,
      "grad_norm": 1.2222771644592285,
      "learning_rate": 3.546827654110464e-05,
      "loss": 0.2502,
      "num_input_tokens_seen": 27463584,
      "step": 14500
    },
    {
      "epoch": 2.366261522146994,
      "grad_norm": 2.4669125080108643,
      "learning_rate": 3.545936038224405e-05,
      "loss": 0.0926,
      "num_input_tokens_seen": 27472304,
      "step": 14505
    },
    {
      "epoch": 2.3670772493678114,
      "grad_norm": 0.47131338715553284,
      "learning_rate": 3.545044261041864e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 27481120,
      "step": 14510
    },
    {
      "epoch": 2.367892976588629,
      "grad_norm": 0.022779524326324463,
      "learning_rate": 3.5441523227003657e-05,
      "loss": 0.1379,
      "num_input_tokens_seen": 27491008,
      "step": 14515
    },
    {
      "epoch": 2.368708703809446,
      "grad_norm": 0.5376589894294739,
      "learning_rate": 3.543260223337459e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 27500752,
      "step": 14520
    },
    {
      "epoch": 2.3695244310302637,
      "grad_norm": 1.5013726949691772,
      "learning_rate": 3.542367963090714e-05,
      "loss": 0.3526,
      "num_input_tokens_seen": 27509504,
      "step": 14525
    },
    {
      "epoch": 2.370340158251081,
      "grad_norm": 0.14662086963653564,
      "learning_rate": 3.5414755420977295e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 27518032,
      "step": 14530
    },
    {
      "epoch": 2.3711558854718984,
      "grad_norm": 0.6707424521446228,
      "learning_rate": 3.54058296049613e-05,
      "loss": 0.0894,
      "num_input_tokens_seen": 27526112,
      "step": 14535
    },
    {
      "epoch": 2.3719716126927155,
      "grad_norm": 0.9236326813697815,
      "learning_rate": 3.53969021842356e-05,
      "loss": 0.1739,
      "num_input_tokens_seen": 27535488,
      "step": 14540
    },
    {
      "epoch": 2.372787339913533,
      "grad_norm": 0.36850571632385254,
      "learning_rate": 3.5387973160176926e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 27544784,
      "step": 14545
    },
    {
      "epoch": 2.37360306713435,
      "grad_norm": 1.266146183013916,
      "learning_rate": 3.537904253416224e-05,
      "loss": 0.2192,
      "num_input_tokens_seen": 27554480,
      "step": 14550
    },
    {
      "epoch": 2.374418794355168,
      "grad_norm": 0.5463860034942627,
      "learning_rate": 3.537011030756878e-05,
      "loss": 0.0613,
      "num_input_tokens_seen": 27563728,
      "step": 14555
    },
    {
      "epoch": 2.375234521575985,
      "grad_norm": 0.8694894909858704,
      "learning_rate": 3.536117648177399e-05,
      "loss": 0.0833,
      "num_input_tokens_seen": 27573152,
      "step": 14560
    },
    {
      "epoch": 2.3760502487968025,
      "grad_norm": 0.5450307130813599,
      "learning_rate": 3.535224105815558e-05,
      "loss": 0.0443,
      "num_input_tokens_seen": 27583072,
      "step": 14565
    },
    {
      "epoch": 2.3768659760176196,
      "grad_norm": 0.44234633445739746,
      "learning_rate": 3.5343304038091494e-05,
      "loss": 0.0776,
      "num_input_tokens_seen": 27591984,
      "step": 14570
    },
    {
      "epoch": 2.377681703238437,
      "grad_norm": 0.25964629650115967,
      "learning_rate": 3.5334365422959955e-05,
      "loss": 0.2057,
      "num_input_tokens_seen": 27602096,
      "step": 14575
    },
    {
      "epoch": 2.3784974304592543,
      "grad_norm": 0.03010065108537674,
      "learning_rate": 3.5325425214139396e-05,
      "loss": 0.1925,
      "num_input_tokens_seen": 27611744,
      "step": 14580
    },
    {
      "epoch": 2.379313157680072,
      "grad_norm": 0.343801349401474,
      "learning_rate": 3.531648341300851e-05,
      "loss": 0.1949,
      "num_input_tokens_seen": 27622000,
      "step": 14585
    },
    {
      "epoch": 2.380128884900889,
      "grad_norm": 0.21994028985500336,
      "learning_rate": 3.530754002094623e-05,
      "loss": 0.0961,
      "num_input_tokens_seen": 27632112,
      "step": 14590
    },
    {
      "epoch": 2.3809446121217066,
      "grad_norm": 0.592134416103363,
      "learning_rate": 3.529859503933175e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 27642256,
      "step": 14595
    },
    {
      "epoch": 2.3817603393425237,
      "grad_norm": 1.280501365661621,
      "learning_rate": 3.52896484695445e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 27652224,
      "step": 14600
    },
    {
      "epoch": 2.3817603393425237,
      "eval_loss": 0.14540241658687592,
      "eval_runtime": 133.3696,
      "eval_samples_per_second": 20.432,
      "eval_steps_per_second": 10.22,
      "num_input_tokens_seen": 27652224,
      "step": 14600
    },
    {
      "epoch": 2.3825760665633413,
      "grad_norm": 0.027464285492897034,
      "learning_rate": 3.528070031296414e-05,
      "loss": 0.0791,
      "num_input_tokens_seen": 27661616,
      "step": 14605
    },
    {
      "epoch": 2.3833917937841584,
      "grad_norm": 0.03678230568766594,
      "learning_rate": 3.5271750570970605e-05,
      "loss": 0.0442,
      "num_input_tokens_seen": 27670304,
      "step": 14610
    },
    {
      "epoch": 2.384207521004976,
      "grad_norm": 0.3407311737537384,
      "learning_rate": 3.526279924494405e-05,
      "loss": 0.0325,
      "num_input_tokens_seen": 27678784,
      "step": 14615
    },
    {
      "epoch": 2.385023248225793,
      "grad_norm": 0.693748950958252,
      "learning_rate": 3.5253846336264874e-05,
      "loss": 0.0884,
      "num_input_tokens_seen": 27688064,
      "step": 14620
    },
    {
      "epoch": 2.3858389754466107,
      "grad_norm": 0.3126915395259857,
      "learning_rate": 3.5244891846313736e-05,
      "loss": 0.1321,
      "num_input_tokens_seen": 27697904,
      "step": 14625
    },
    {
      "epoch": 2.386654702667428,
      "grad_norm": 0.040208008140325546,
      "learning_rate": 3.5235935776471527e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 27708528,
      "step": 14630
    },
    {
      "epoch": 2.3874704298882454,
      "grad_norm": 0.2544792592525482,
      "learning_rate": 3.522697812811939e-05,
      "loss": 0.2425,
      "num_input_tokens_seen": 27717552,
      "step": 14635
    },
    {
      "epoch": 2.3882861571090626,
      "grad_norm": 0.4351957440376282,
      "learning_rate": 3.521801890263871e-05,
      "loss": 0.2864,
      "num_input_tokens_seen": 27727088,
      "step": 14640
    },
    {
      "epoch": 2.38910188432988,
      "grad_norm": 0.45363426208496094,
      "learning_rate": 3.5209058101411114e-05,
      "loss": 0.0358,
      "num_input_tokens_seen": 27736800,
      "step": 14645
    },
    {
      "epoch": 2.3899176115506973,
      "grad_norm": 0.5730546116828918,
      "learning_rate": 3.520009572581845e-05,
      "loss": 0.2083,
      "num_input_tokens_seen": 27746608,
      "step": 14650
    },
    {
      "epoch": 2.390733338771515,
      "grad_norm": 1.8479868173599243,
      "learning_rate": 3.519113177724285e-05,
      "loss": 0.045,
      "num_input_tokens_seen": 27756464,
      "step": 14655
    },
    {
      "epoch": 2.391549065992332,
      "grad_norm": 1.0961627960205078,
      "learning_rate": 3.5182166257066656e-05,
      "loss": 0.1528,
      "num_input_tokens_seen": 27766688,
      "step": 14660
    },
    {
      "epoch": 2.3923647932131495,
      "grad_norm": 0.7594425082206726,
      "learning_rate": 3.517319916667247e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 27775792,
      "step": 14665
    },
    {
      "epoch": 2.393180520433967,
      "grad_norm": 0.17462123930454254,
      "learning_rate": 3.516423050744313e-05,
      "loss": 0.0695,
      "num_input_tokens_seen": 27786720,
      "step": 14670
    },
    {
      "epoch": 2.3939962476547842,
      "grad_norm": 0.9223204851150513,
      "learning_rate": 3.5155260280761704e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 27796832,
      "step": 14675
    },
    {
      "epoch": 2.3948119748756014,
      "grad_norm": 0.4028564691543579,
      "learning_rate": 3.514628848801154e-05,
      "loss": 0.2579,
      "num_input_tokens_seen": 27805584,
      "step": 14680
    },
    {
      "epoch": 2.395627702096419,
      "grad_norm": 2.2711212635040283,
      "learning_rate": 3.5137315130576174e-05,
      "loss": 0.2281,
      "num_input_tokens_seen": 27816208,
      "step": 14685
    },
    {
      "epoch": 2.3964434293172365,
      "grad_norm": 0.8827929496765137,
      "learning_rate": 3.512834020983942e-05,
      "loss": 0.2628,
      "num_input_tokens_seen": 27825536,
      "step": 14690
    },
    {
      "epoch": 2.3972591565380537,
      "grad_norm": 0.16584837436676025,
      "learning_rate": 3.5119363727185334e-05,
      "loss": 0.2282,
      "num_input_tokens_seen": 27835488,
      "step": 14695
    },
    {
      "epoch": 2.398074883758871,
      "grad_norm": 0.4740711450576782,
      "learning_rate": 3.511038568399819e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 27844368,
      "step": 14700
    },
    {
      "epoch": 2.3988906109796884,
      "grad_norm": 0.15900878608226776,
      "learning_rate": 3.510140608166251e-05,
      "loss": 0.0898,
      "num_input_tokens_seen": 27853104,
      "step": 14705
    },
    {
      "epoch": 2.399706338200506,
      "grad_norm": 0.3075067400932312,
      "learning_rate": 3.509242492156308e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 27863120,
      "step": 14710
    },
    {
      "epoch": 2.400522065421323,
      "grad_norm": 1.2981226444244385,
      "learning_rate": 3.5083442205084896e-05,
      "loss": 0.1689,
      "num_input_tokens_seen": 27872560,
      "step": 14715
    },
    {
      "epoch": 2.4013377926421406,
      "grad_norm": 0.11563107371330261,
      "learning_rate": 3.507445793361321e-05,
      "loss": 0.0643,
      "num_input_tokens_seen": 27881728,
      "step": 14720
    },
    {
      "epoch": 2.4021535198629578,
      "grad_norm": 0.3651931583881378,
      "learning_rate": 3.5065472108533505e-05,
      "loss": 0.0826,
      "num_input_tokens_seen": 27891664,
      "step": 14725
    },
    {
      "epoch": 2.4029692470837754,
      "grad_norm": 0.5993765592575073,
      "learning_rate": 3.5056484731231504e-05,
      "loss": 0.0874,
      "num_input_tokens_seen": 27901504,
      "step": 14730
    },
    {
      "epoch": 2.4037849743045925,
      "grad_norm": 0.19572813808918,
      "learning_rate": 3.504749580309319e-05,
      "loss": 0.0912,
      "num_input_tokens_seen": 27910720,
      "step": 14735
    },
    {
      "epoch": 2.40460070152541,
      "grad_norm": 0.33224889636039734,
      "learning_rate": 3.5038505325504753e-05,
      "loss": 0.119,
      "num_input_tokens_seen": 27920288,
      "step": 14740
    },
    {
      "epoch": 2.405416428746227,
      "grad_norm": 0.5318416953086853,
      "learning_rate": 3.502951329985264e-05,
      "loss": 0.1176,
      "num_input_tokens_seen": 27929312,
      "step": 14745
    },
    {
      "epoch": 2.4062321559670448,
      "grad_norm": 0.2602984309196472,
      "learning_rate": 3.502051972752354e-05,
      "loss": 0.0545,
      "num_input_tokens_seen": 27938656,
      "step": 14750
    },
    {
      "epoch": 2.407047883187862,
      "grad_norm": 0.3040507733821869,
      "learning_rate": 3.5011524609904374e-05,
      "loss": 0.1204,
      "num_input_tokens_seen": 27949184,
      "step": 14755
    },
    {
      "epoch": 2.4078636104086795,
      "grad_norm": 0.34726619720458984,
      "learning_rate": 3.50025279483823e-05,
      "loss": 0.0813,
      "num_input_tokens_seen": 27957632,
      "step": 14760
    },
    {
      "epoch": 2.4086793376294966,
      "grad_norm": 0.38397008180618286,
      "learning_rate": 3.499352974434472e-05,
      "loss": 0.2442,
      "num_input_tokens_seen": 27966976,
      "step": 14765
    },
    {
      "epoch": 2.409495064850314,
      "grad_norm": 0.40166857838630676,
      "learning_rate": 3.498452999917926e-05,
      "loss": 0.1889,
      "num_input_tokens_seen": 27975856,
      "step": 14770
    },
    {
      "epoch": 2.4103107920711313,
      "grad_norm": 0.4476436376571655,
      "learning_rate": 3.4975528714273795e-05,
      "loss": 0.0725,
      "num_input_tokens_seen": 27985600,
      "step": 14775
    },
    {
      "epoch": 2.411126519291949,
      "grad_norm": 1.2755331993103027,
      "learning_rate": 3.4966525891016454e-05,
      "loss": 0.2138,
      "num_input_tokens_seen": 27996160,
      "step": 14780
    },
    {
      "epoch": 2.411942246512766,
      "grad_norm": 1.8744397163391113,
      "learning_rate": 3.495752153079557e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 28006240,
      "step": 14785
    },
    {
      "epoch": 2.4127579737335836,
      "grad_norm": 0.6431491374969482,
      "learning_rate": 3.494851563499974e-05,
      "loss": 0.1071,
      "num_input_tokens_seen": 28014912,
      "step": 14790
    },
    {
      "epoch": 2.4135737009544007,
      "grad_norm": 0.869003415107727,
      "learning_rate": 3.493950820501777e-05,
      "loss": 0.0663,
      "num_input_tokens_seen": 28023472,
      "step": 14795
    },
    {
      "epoch": 2.4143894281752183,
      "grad_norm": 1.893977165222168,
      "learning_rate": 3.493049924223872e-05,
      "loss": 0.2243,
      "num_input_tokens_seen": 28033168,
      "step": 14800
    },
    {
      "epoch": 2.4143894281752183,
      "eval_loss": 0.14566288888454437,
      "eval_runtime": 133.3256,
      "eval_samples_per_second": 20.439,
      "eval_steps_per_second": 10.223,
      "num_input_tokens_seen": 28033168,
      "step": 14800
    },
    {
      "epoch": 2.4152051553960354,
      "grad_norm": 0.037970010191202164,
      "learning_rate": 3.49214887480519e-05,
      "loss": 0.3283,
      "num_input_tokens_seen": 28042560,
      "step": 14805
    },
    {
      "epoch": 2.416020882616853,
      "grad_norm": 0.8617724776268005,
      "learning_rate": 3.4912476723846834e-05,
      "loss": 0.2404,
      "num_input_tokens_seen": 28051488,
      "step": 14810
    },
    {
      "epoch": 2.41683660983767,
      "grad_norm": 0.370486319065094,
      "learning_rate": 3.490346317101328e-05,
      "loss": 0.1124,
      "num_input_tokens_seen": 28061072,
      "step": 14815
    },
    {
      "epoch": 2.4176523370584877,
      "grad_norm": 0.6741062998771667,
      "learning_rate": 3.4894448090941266e-05,
      "loss": 0.1143,
      "num_input_tokens_seen": 28070256,
      "step": 14820
    },
    {
      "epoch": 2.418468064279305,
      "grad_norm": 0.0711764320731163,
      "learning_rate": 3.488543148502101e-05,
      "loss": 0.0296,
      "num_input_tokens_seen": 28080064,
      "step": 14825
    },
    {
      "epoch": 2.4192837915001224,
      "grad_norm": 0.90719074010849,
      "learning_rate": 3.487641335464299e-05,
      "loss": 0.0826,
      "num_input_tokens_seen": 28087712,
      "step": 14830
    },
    {
      "epoch": 2.4200995187209395,
      "grad_norm": 0.061175744980573654,
      "learning_rate": 3.4867393701197914e-05,
      "loss": 0.0796,
      "num_input_tokens_seen": 28098160,
      "step": 14835
    },
    {
      "epoch": 2.420915245941757,
      "grad_norm": 0.4908466935157776,
      "learning_rate": 3.485837252607673e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 28107920,
      "step": 14840
    },
    {
      "epoch": 2.4217309731625742,
      "grad_norm": 0.7161101698875427,
      "learning_rate": 3.4849349830670615e-05,
      "loss": 0.2087,
      "num_input_tokens_seen": 28117312,
      "step": 14845
    },
    {
      "epoch": 2.422546700383392,
      "grad_norm": 0.4376928508281708,
      "learning_rate": 3.4840325616370976e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 28126192,
      "step": 14850
    },
    {
      "epoch": 2.423362427604209,
      "grad_norm": 1.0744678974151611,
      "learning_rate": 3.483129988456947e-05,
      "loss": 0.2149,
      "num_input_tokens_seen": 28135040,
      "step": 14855
    },
    {
      "epoch": 2.4241781548250265,
      "grad_norm": 0.8785051703453064,
      "learning_rate": 3.482227263665797e-05,
      "loss": 0.0861,
      "num_input_tokens_seen": 28143488,
      "step": 14860
    },
    {
      "epoch": 2.424993882045844,
      "grad_norm": 0.41906124353408813,
      "learning_rate": 3.48132438740286e-05,
      "loss": 0.2425,
      "num_input_tokens_seen": 28154304,
      "step": 14865
    },
    {
      "epoch": 2.4258096092666612,
      "grad_norm": 0.45430266857147217,
      "learning_rate": 3.48042135980737e-05,
      "loss": 0.0856,
      "num_input_tokens_seen": 28163632,
      "step": 14870
    },
    {
      "epoch": 2.4266253364874784,
      "grad_norm": 2.2270472049713135,
      "learning_rate": 3.479518181018586e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 28173248,
      "step": 14875
    },
    {
      "epoch": 2.427441063708296,
      "grad_norm": 1.0457675457000732,
      "learning_rate": 3.4786148511757886e-05,
      "loss": 0.0842,
      "num_input_tokens_seen": 28183600,
      "step": 14880
    },
    {
      "epoch": 2.4282567909291135,
      "grad_norm": 0.5003524422645569,
      "learning_rate": 3.477711370418284e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 28194672,
      "step": 14885
    },
    {
      "epoch": 2.4290725181499306,
      "grad_norm": 0.6108632683753967,
      "learning_rate": 3.476807738885399e-05,
      "loss": 0.0405,
      "num_input_tokens_seen": 28204288,
      "step": 14890
    },
    {
      "epoch": 2.429888245370748,
      "grad_norm": 1.1629434823989868,
      "learning_rate": 3.475903956716485e-05,
      "loss": 0.2375,
      "num_input_tokens_seen": 28213392,
      "step": 14895
    },
    {
      "epoch": 2.4307039725915653,
      "grad_norm": 0.28323566913604736,
      "learning_rate": 3.475000024050917e-05,
      "loss": 0.2829,
      "num_input_tokens_seen": 28222464,
      "step": 14900
    },
    {
      "epoch": 2.431519699812383,
      "grad_norm": 0.5120511651039124,
      "learning_rate": 3.4740959410280926e-05,
      "loss": 0.2593,
      "num_input_tokens_seen": 28231440,
      "step": 14905
    },
    {
      "epoch": 2.4323354270332,
      "grad_norm": 0.1383836418390274,
      "learning_rate": 3.4731917077874324e-05,
      "loss": 0.2189,
      "num_input_tokens_seen": 28242736,
      "step": 14910
    },
    {
      "epoch": 2.4331511542540176,
      "grad_norm": 0.09822043776512146,
      "learning_rate": 3.4722873244683816e-05,
      "loss": 0.2321,
      "num_input_tokens_seen": 28252880,
      "step": 14915
    },
    {
      "epoch": 2.4339668814748348,
      "grad_norm": 0.2946453392505646,
      "learning_rate": 3.4713827912104065e-05,
      "loss": 0.0591,
      "num_input_tokens_seen": 28262464,
      "step": 14920
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 0.23088596761226654,
      "learning_rate": 3.470478108152998e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 28271424,
      "step": 14925
    },
    {
      "epoch": 2.4355983359164695,
      "grad_norm": 0.17221558094024658,
      "learning_rate": 3.4695732754356695e-05,
      "loss": 0.0424,
      "num_input_tokens_seen": 28281456,
      "step": 14930
    },
    {
      "epoch": 2.436414063137287,
      "grad_norm": 0.7811086773872375,
      "learning_rate": 3.4686682931979576e-05,
      "loss": 0.2083,
      "num_input_tokens_seen": 28291984,
      "step": 14935
    },
    {
      "epoch": 2.437229790358104,
      "grad_norm": 0.06485319137573242,
      "learning_rate": 3.467763161579422e-05,
      "loss": 0.0749,
      "num_input_tokens_seen": 28300592,
      "step": 14940
    },
    {
      "epoch": 2.4380455175789217,
      "grad_norm": 0.1955939084291458,
      "learning_rate": 3.466857880719645e-05,
      "loss": 0.0537,
      "num_input_tokens_seen": 28309248,
      "step": 14945
    },
    {
      "epoch": 2.438861244799739,
      "grad_norm": 0.47808510065078735,
      "learning_rate": 3.465952450758233e-05,
      "loss": 0.0714,
      "num_input_tokens_seen": 28318704,
      "step": 14950
    },
    {
      "epoch": 2.4396769720205564,
      "grad_norm": 0.6653221845626831,
      "learning_rate": 3.4650468718348126e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 28328704,
      "step": 14955
    },
    {
      "epoch": 2.4404926992413736,
      "grad_norm": 0.5665784478187561,
      "learning_rate": 3.464141144089038e-05,
      "loss": 0.0836,
      "num_input_tokens_seen": 28337248,
      "step": 14960
    },
    {
      "epoch": 2.441308426462191,
      "grad_norm": 0.07314970344305038,
      "learning_rate": 3.463235267660583e-05,
      "loss": 0.0671,
      "num_input_tokens_seen": 28347040,
      "step": 14965
    },
    {
      "epoch": 2.4421241536830083,
      "grad_norm": 3.094735860824585,
      "learning_rate": 3.462329242689145e-05,
      "loss": 0.1102,
      "num_input_tokens_seen": 28356960,
      "step": 14970
    },
    {
      "epoch": 2.442939880903826,
      "grad_norm": 0.30612489581108093,
      "learning_rate": 3.461423069314444e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 28367008,
      "step": 14975
    },
    {
      "epoch": 2.443755608124643,
      "grad_norm": 0.41707393527030945,
      "learning_rate": 3.460516747676224e-05,
      "loss": 0.0486,
      "num_input_tokens_seen": 28376592,
      "step": 14980
    },
    {
      "epoch": 2.4445713353454606,
      "grad_norm": 2.0200271606445312,
      "learning_rate": 3.459610277914251e-05,
      "loss": 0.3482,
      "num_input_tokens_seen": 28386688,
      "step": 14985
    },
    {
      "epoch": 2.4453870625662777,
      "grad_norm": 0.7731131315231323,
      "learning_rate": 3.458703660168314e-05,
      "loss": 0.0749,
      "num_input_tokens_seen": 28396304,
      "step": 14990
    },
    {
      "epoch": 2.4462027897870953,
      "grad_norm": 0.16643407940864563,
      "learning_rate": 3.457796894578224e-05,
      "loss": 0.2898,
      "num_input_tokens_seen": 28405712,
      "step": 14995
    },
    {
      "epoch": 2.4470185170079124,
      "grad_norm": 0.047046199440956116,
      "learning_rate": 3.456889981283817e-05,
      "loss": 0.2481,
      "num_input_tokens_seen": 28414784,
      "step": 15000
    },
    {
      "epoch": 2.4470185170079124,
      "eval_loss": 0.14551900327205658,
      "eval_runtime": 133.2899,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 28414784,
      "step": 15000
    },
    {
      "epoch": 2.44783424422873,
      "grad_norm": 0.8871802091598511,
      "learning_rate": 3.45598292042495e-05,
      "loss": 0.2562,
      "num_input_tokens_seen": 28423024,
      "step": 15005
    },
    {
      "epoch": 2.448649971449547,
      "grad_norm": 1.1791341304779053,
      "learning_rate": 3.4550757121415035e-05,
      "loss": 0.1059,
      "num_input_tokens_seen": 28430992,
      "step": 15010
    },
    {
      "epoch": 2.4494656986703647,
      "grad_norm": 0.2891627252101898,
      "learning_rate": 3.454168356573378e-05,
      "loss": 0.0907,
      "num_input_tokens_seen": 28440432,
      "step": 15015
    },
    {
      "epoch": 2.450281425891182,
      "grad_norm": 1.143950343132019,
      "learning_rate": 3.453260853860503e-05,
      "loss": 0.0909,
      "num_input_tokens_seen": 28449248,
      "step": 15020
    },
    {
      "epoch": 2.4510971531119994,
      "grad_norm": 0.18598763644695282,
      "learning_rate": 3.452353204142824e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 28458240,
      "step": 15025
    },
    {
      "epoch": 2.4519128803328165,
      "grad_norm": 0.6014482378959656,
      "learning_rate": 3.4514454075603136e-05,
      "loss": 0.0526,
      "num_input_tokens_seen": 28467200,
      "step": 15030
    },
    {
      "epoch": 2.452728607553634,
      "grad_norm": 0.7782996892929077,
      "learning_rate": 3.450537464252964e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 28476432,
      "step": 15035
    },
    {
      "epoch": 2.4535443347744517,
      "grad_norm": 0.7581204175949097,
      "learning_rate": 3.4496293743607925e-05,
      "loss": 0.0696,
      "num_input_tokens_seen": 28485840,
      "step": 15040
    },
    {
      "epoch": 2.454360061995269,
      "grad_norm": 0.21126283705234528,
      "learning_rate": 3.448721138023838e-05,
      "loss": 0.0725,
      "num_input_tokens_seen": 28494576,
      "step": 15045
    },
    {
      "epoch": 2.455175789216086,
      "grad_norm": 0.3699236810207367,
      "learning_rate": 3.447812755382162e-05,
      "loss": 0.0779,
      "num_input_tokens_seen": 28504960,
      "step": 15050
    },
    {
      "epoch": 2.4559915164369035,
      "grad_norm": 0.5519641637802124,
      "learning_rate": 3.446904226575847e-05,
      "loss": 0.2412,
      "num_input_tokens_seen": 28515072,
      "step": 15055
    },
    {
      "epoch": 2.456807243657721,
      "grad_norm": 0.1277916133403778,
      "learning_rate": 3.445995551745002e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 28524096,
      "step": 15060
    },
    {
      "epoch": 2.457622970878538,
      "grad_norm": 1.6715666055679321,
      "learning_rate": 3.445086731029753e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 28533056,
      "step": 15065
    },
    {
      "epoch": 2.4584386980993553,
      "grad_norm": 1.3284715414047241,
      "learning_rate": 3.444177764570255e-05,
      "loss": 0.1248,
      "num_input_tokens_seen": 28543424,
      "step": 15070
    },
    {
      "epoch": 2.459254425320173,
      "grad_norm": 1.0762038230895996,
      "learning_rate": 3.44326865250668e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 28553632,
      "step": 15075
    },
    {
      "epoch": 2.4600701525409905,
      "grad_norm": 0.291343092918396,
      "learning_rate": 3.442359394979225e-05,
      "loss": 0.0705,
      "num_input_tokens_seen": 28563536,
      "step": 15080
    },
    {
      "epoch": 2.4608858797618076,
      "grad_norm": 0.8432093262672424,
      "learning_rate": 3.441449992128108e-05,
      "loss": 0.0795,
      "num_input_tokens_seen": 28572784,
      "step": 15085
    },
    {
      "epoch": 2.461701606982625,
      "grad_norm": 1.326744794845581,
      "learning_rate": 3.440540444093573e-05,
      "loss": 0.0899,
      "num_input_tokens_seen": 28581008,
      "step": 15090
    },
    {
      "epoch": 2.4625173342034423,
      "grad_norm": 0.04325085133314133,
      "learning_rate": 3.43963075101588e-05,
      "loss": 0.1055,
      "num_input_tokens_seen": 28590192,
      "step": 15095
    },
    {
      "epoch": 2.46333306142426,
      "grad_norm": 0.1935480535030365,
      "learning_rate": 3.438720913035318e-05,
      "loss": 0.0686,
      "num_input_tokens_seen": 28600016,
      "step": 15100
    },
    {
      "epoch": 2.464148788645077,
      "grad_norm": 0.2594442367553711,
      "learning_rate": 3.437810930292195e-05,
      "loss": 0.0913,
      "num_input_tokens_seen": 28609664,
      "step": 15105
    },
    {
      "epoch": 2.4649645158658946,
      "grad_norm": 0.22864115238189697,
      "learning_rate": 3.43690080292684e-05,
      "loss": 0.127,
      "num_input_tokens_seen": 28617648,
      "step": 15110
    },
    {
      "epoch": 2.4657802430867117,
      "grad_norm": 0.291552871465683,
      "learning_rate": 3.435990531079608e-05,
      "loss": 0.0673,
      "num_input_tokens_seen": 28625952,
      "step": 15115
    },
    {
      "epoch": 2.4665959703075293,
      "grad_norm": 0.3968774676322937,
      "learning_rate": 3.435080114890874e-05,
      "loss": 0.0945,
      "num_input_tokens_seen": 28636256,
      "step": 15120
    },
    {
      "epoch": 2.4674116975283464,
      "grad_norm": 0.09681129455566406,
      "learning_rate": 3.434169554501035e-05,
      "loss": 0.2424,
      "num_input_tokens_seen": 28645280,
      "step": 15125
    },
    {
      "epoch": 2.468227424749164,
      "grad_norm": 0.17665763199329376,
      "learning_rate": 3.433258850050511e-05,
      "loss": 0.0502,
      "num_input_tokens_seen": 28655312,
      "step": 15130
    },
    {
      "epoch": 2.469043151969981,
      "grad_norm": 0.7956153154373169,
      "learning_rate": 3.4323480016797446e-05,
      "loss": 0.2078,
      "num_input_tokens_seen": 28663904,
      "step": 15135
    },
    {
      "epoch": 2.4698588791907987,
      "grad_norm": 0.6386601328849792,
      "learning_rate": 3.4314370095291995e-05,
      "loss": 0.0635,
      "num_input_tokens_seen": 28673920,
      "step": 15140
    },
    {
      "epoch": 2.470674606411616,
      "grad_norm": 1.110192894935608,
      "learning_rate": 3.430525873739363e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 28683024,
      "step": 15145
    },
    {
      "epoch": 2.4714903336324334,
      "grad_norm": 2.333568811416626,
      "learning_rate": 3.429614594450743e-05,
      "loss": 0.2808,
      "num_input_tokens_seen": 28691840,
      "step": 15150
    },
    {
      "epoch": 2.4723060608532506,
      "grad_norm": 0.12748318910598755,
      "learning_rate": 3.428703171803869e-05,
      "loss": 0.1019,
      "num_input_tokens_seen": 28700000,
      "step": 15155
    },
    {
      "epoch": 2.473121788074068,
      "grad_norm": 1.4109989404678345,
      "learning_rate": 3.4277916059392964e-05,
      "loss": 0.2176,
      "num_input_tokens_seen": 28710336,
      "step": 15160
    },
    {
      "epoch": 2.4739375152948853,
      "grad_norm": 1.54707932472229,
      "learning_rate": 3.426879896997598e-05,
      "loss": 0.1049,
      "num_input_tokens_seen": 28719632,
      "step": 15165
    },
    {
      "epoch": 2.474753242515703,
      "grad_norm": 0.14016665518283844,
      "learning_rate": 3.425968045119372e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 28729008,
      "step": 15170
    },
    {
      "epoch": 2.47556896973652,
      "grad_norm": 0.13914109766483307,
      "learning_rate": 3.425056050445237e-05,
      "loss": 0.1255,
      "num_input_tokens_seen": 28738256,
      "step": 15175
    },
    {
      "epoch": 2.4763846969573375,
      "grad_norm": 0.40189918875694275,
      "learning_rate": 3.4241439131158336e-05,
      "loss": 0.198,
      "num_input_tokens_seen": 28748544,
      "step": 15180
    },
    {
      "epoch": 2.4772004241781547,
      "grad_norm": 0.578120231628418,
      "learning_rate": 3.423231633271825e-05,
      "loss": 0.0709,
      "num_input_tokens_seen": 28758448,
      "step": 15185
    },
    {
      "epoch": 2.4780161513989722,
      "grad_norm": 1.2915236949920654,
      "learning_rate": 3.4223192110538985e-05,
      "loss": 0.2021,
      "num_input_tokens_seen": 28768704,
      "step": 15190
    },
    {
      "epoch": 2.4788318786197894,
      "grad_norm": 1.7417246103286743,
      "learning_rate": 3.4214066466027575e-05,
      "loss": 0.1993,
      "num_input_tokens_seen": 28778640,
      "step": 15195
    },
    {
      "epoch": 2.479647605840607,
      "grad_norm": 0.4005047678947449,
      "learning_rate": 3.4204939400591325e-05,
      "loss": 0.2391,
      "num_input_tokens_seen": 28787168,
      "step": 15200
    },
    {
      "epoch": 2.479647605840607,
      "eval_loss": 0.15121088922023773,
      "eval_runtime": 133.3415,
      "eval_samples_per_second": 20.436,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 28787168,
      "step": 15200
    },
    {
      "epoch": 2.480463333061424,
      "grad_norm": 2.9256975650787354,
      "learning_rate": 3.419581091563775e-05,
      "loss": 0.3585,
      "num_input_tokens_seen": 28796512,
      "step": 15205
    },
    {
      "epoch": 2.4812790602822417,
      "grad_norm": 1.211462140083313,
      "learning_rate": 3.418668101257456e-05,
      "loss": 0.0979,
      "num_input_tokens_seen": 28805392,
      "step": 15210
    },
    {
      "epoch": 2.482094787503059,
      "grad_norm": 0.41505980491638184,
      "learning_rate": 3.417754969280971e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 28815024,
      "step": 15215
    },
    {
      "epoch": 2.4829105147238764,
      "grad_norm": 1.3694305419921875,
      "learning_rate": 3.416841695775137e-05,
      "loss": 0.1748,
      "num_input_tokens_seen": 28824464,
      "step": 15220
    },
    {
      "epoch": 2.4837262419446935,
      "grad_norm": 0.0765702873468399,
      "learning_rate": 3.415928280880792e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 28833968,
      "step": 15225
    },
    {
      "epoch": 2.484541969165511,
      "grad_norm": 0.06251150369644165,
      "learning_rate": 3.4150147247387965e-05,
      "loss": 0.0551,
      "num_input_tokens_seen": 28843232,
      "step": 15230
    },
    {
      "epoch": 2.4853576963863286,
      "grad_norm": 0.5310118794441223,
      "learning_rate": 3.4141010274900306e-05,
      "loss": 0.2019,
      "num_input_tokens_seen": 28851040,
      "step": 15235
    },
    {
      "epoch": 2.4861734236071458,
      "grad_norm": 0.7653997540473938,
      "learning_rate": 3.413187189275399e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 28860896,
      "step": 15240
    },
    {
      "epoch": 2.486989150827963,
      "grad_norm": 0.03781246021389961,
      "learning_rate": 3.4122732102358265e-05,
      "loss": 0.0964,
      "num_input_tokens_seen": 28869904,
      "step": 15245
    },
    {
      "epoch": 2.4878048780487805,
      "grad_norm": 0.15079480409622192,
      "learning_rate": 3.411359090512261e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 28879600,
      "step": 15250
    },
    {
      "epoch": 2.488620605269598,
      "grad_norm": 0.1477055698633194,
      "learning_rate": 3.410444830245672e-05,
      "loss": 0.0329,
      "num_input_tokens_seen": 28888944,
      "step": 15255
    },
    {
      "epoch": 2.489436332490415,
      "grad_norm": 0.4751497507095337,
      "learning_rate": 3.409530429577048e-05,
      "loss": 0.0516,
      "num_input_tokens_seen": 28898016,
      "step": 15260
    },
    {
      "epoch": 2.4902520597112328,
      "grad_norm": 0.17289353907108307,
      "learning_rate": 3.408615888647402e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 28907232,
      "step": 15265
    },
    {
      "epoch": 2.49106778693205,
      "grad_norm": 0.7725971937179565,
      "learning_rate": 3.4077012075977675e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 28916240,
      "step": 15270
    },
    {
      "epoch": 2.4918835141528675,
      "grad_norm": 1.5769081115722656,
      "learning_rate": 3.4067863865692e-05,
      "loss": 0.2277,
      "num_input_tokens_seen": 28925936,
      "step": 15275
    },
    {
      "epoch": 2.4926992413736846,
      "grad_norm": 0.1347542703151703,
      "learning_rate": 3.4058714257027755e-05,
      "loss": 0.0968,
      "num_input_tokens_seen": 28935440,
      "step": 15280
    },
    {
      "epoch": 2.493514968594502,
      "grad_norm": 0.621975302696228,
      "learning_rate": 3.404956325139594e-05,
      "loss": 0.229,
      "num_input_tokens_seen": 28946048,
      "step": 15285
    },
    {
      "epoch": 2.4943306958153193,
      "grad_norm": 1.4775617122650146,
      "learning_rate": 3.404041085020775e-05,
      "loss": 0.1002,
      "num_input_tokens_seen": 28955664,
      "step": 15290
    },
    {
      "epoch": 2.495146423036137,
      "grad_norm": 1.2553898096084595,
      "learning_rate": 3.403125705487459e-05,
      "loss": 0.1188,
      "num_input_tokens_seen": 28964704,
      "step": 15295
    },
    {
      "epoch": 2.495962150256954,
      "grad_norm": 0.35638707876205444,
      "learning_rate": 3.402210186680811e-05,
      "loss": 0.1272,
      "num_input_tokens_seen": 28974176,
      "step": 15300
    },
    {
      "epoch": 2.4967778774777716,
      "grad_norm": 0.7892554998397827,
      "learning_rate": 3.4012945287420137e-05,
      "loss": 0.1998,
      "num_input_tokens_seen": 28983184,
      "step": 15305
    },
    {
      "epoch": 2.4975936046985887,
      "grad_norm": 1.6801023483276367,
      "learning_rate": 3.400378731812274e-05,
      "loss": 0.2302,
      "num_input_tokens_seen": 28992080,
      "step": 15310
    },
    {
      "epoch": 2.4984093319194063,
      "grad_norm": 0.45703110098838806,
      "learning_rate": 3.399462796032817e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 29002368,
      "step": 15315
    },
    {
      "epoch": 2.4992250591402234,
      "grad_norm": 1.2145299911499023,
      "learning_rate": 3.3985467215448954e-05,
      "loss": 0.1008,
      "num_input_tokens_seen": 29011392,
      "step": 15320
    },
    {
      "epoch": 2.500040786361041,
      "grad_norm": 0.7345907688140869,
      "learning_rate": 3.3976305084897776e-05,
      "loss": 0.0882,
      "num_input_tokens_seen": 29020832,
      "step": 15325
    },
    {
      "epoch": 2.500856513581858,
      "grad_norm": 0.5748437643051147,
      "learning_rate": 3.3967141570087544e-05,
      "loss": 0.0459,
      "num_input_tokens_seen": 29030272,
      "step": 15330
    },
    {
      "epoch": 2.5016722408026757,
      "grad_norm": 0.07623119652271271,
      "learning_rate": 3.39579766724314e-05,
      "loss": 0.126,
      "num_input_tokens_seen": 29040800,
      "step": 15335
    },
    {
      "epoch": 2.502487968023493,
      "grad_norm": 0.08467543870210648,
      "learning_rate": 3.3948810393342677e-05,
      "loss": 0.1846,
      "num_input_tokens_seen": 29050512,
      "step": 15340
    },
    {
      "epoch": 2.5033036952443104,
      "grad_norm": 1.4438601732254028,
      "learning_rate": 3.3939642734234936e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 29060528,
      "step": 15345
    },
    {
      "epoch": 2.5041194224651275,
      "grad_norm": 0.5565962195396423,
      "learning_rate": 3.393047369652194e-05,
      "loss": 0.1962,
      "num_input_tokens_seen": 29070448,
      "step": 15350
    },
    {
      "epoch": 2.504935149685945,
      "grad_norm": 0.3362228274345398,
      "learning_rate": 3.3921303281617664e-05,
      "loss": 0.0725,
      "num_input_tokens_seen": 29079760,
      "step": 15355
    },
    {
      "epoch": 2.5057508769067622,
      "grad_norm": 0.09245380759239197,
      "learning_rate": 3.391213149093632e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 29089008,
      "step": 15360
    },
    {
      "epoch": 2.50656660412758,
      "grad_norm": 0.03126201033592224,
      "learning_rate": 3.3902958325892303e-05,
      "loss": 0.2999,
      "num_input_tokens_seen": 29099968,
      "step": 15365
    },
    {
      "epoch": 2.507382331348397,
      "grad_norm": 0.6026391386985779,
      "learning_rate": 3.389378378790023e-05,
      "loss": 0.094,
      "num_input_tokens_seen": 29108912,
      "step": 15370
    },
    {
      "epoch": 2.5081980585692145,
      "grad_norm": 1.761401891708374,
      "learning_rate": 3.388460787837493e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 29118656,
      "step": 15375
    },
    {
      "epoch": 2.5090137857900316,
      "grad_norm": 0.29005029797554016,
      "learning_rate": 3.387543059873145e-05,
      "loss": 0.0938,
      "num_input_tokens_seen": 29128560,
      "step": 15380
    },
    {
      "epoch": 2.5098295130108492,
      "grad_norm": 1.325842261314392,
      "learning_rate": 3.386625195038503e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 29138032,
      "step": 15385
    },
    {
      "epoch": 2.510645240231667,
      "grad_norm": 1.517878770828247,
      "learning_rate": 3.3857071934751136e-05,
      "loss": 0.3229,
      "num_input_tokens_seen": 29147120,
      "step": 15390
    },
    {
      "epoch": 2.511460967452484,
      "grad_norm": 0.12483509629964828,
      "learning_rate": 3.384789055324544e-05,
      "loss": 0.0761,
      "num_input_tokens_seen": 29155504,
      "step": 15395
    },
    {
      "epoch": 2.512276694673301,
      "grad_norm": 0.20009098947048187,
      "learning_rate": 3.3838707807283843e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 29164512,
      "step": 15400
    },
    {
      "epoch": 2.512276694673301,
      "eval_loss": 0.14524851739406586,
      "eval_runtime": 133.2788,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 29164512,
      "step": 15400
    },
    {
      "epoch": 2.5130924218941186,
      "grad_norm": 1.7263997793197632,
      "learning_rate": 3.382952369828243e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 29172336,
      "step": 15405
    },
    {
      "epoch": 2.513908149114936,
      "grad_norm": 0.1650303453207016,
      "learning_rate": 3.38203382276575e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 29182592,
      "step": 15410
    },
    {
      "epoch": 2.5147238763357533,
      "grad_norm": 0.09998579323291779,
      "learning_rate": 3.381115139682557e-05,
      "loss": 0.0866,
      "num_input_tokens_seen": 29191968,
      "step": 15415
    },
    {
      "epoch": 2.5155396035565705,
      "grad_norm": 0.4434095025062561,
      "learning_rate": 3.3801963207203366e-05,
      "loss": 0.0441,
      "num_input_tokens_seen": 29202944,
      "step": 15420
    },
    {
      "epoch": 2.516355330777388,
      "grad_norm": 0.6225411295890808,
      "learning_rate": 3.379277366020782e-05,
      "loss": 0.1183,
      "num_input_tokens_seen": 29213328,
      "step": 15425
    },
    {
      "epoch": 2.5171710579982056,
      "grad_norm": 0.8857307434082031,
      "learning_rate": 3.3783582757256085e-05,
      "loss": 0.274,
      "num_input_tokens_seen": 29223120,
      "step": 15430
    },
    {
      "epoch": 2.5179867852190227,
      "grad_norm": 0.6830242276191711,
      "learning_rate": 3.3774390499765504e-05,
      "loss": 0.1038,
      "num_input_tokens_seen": 29232240,
      "step": 15435
    },
    {
      "epoch": 2.51880251243984,
      "grad_norm": 2.7003095149993896,
      "learning_rate": 3.376519688915364e-05,
      "loss": 0.2701,
      "num_input_tokens_seen": 29241376,
      "step": 15440
    },
    {
      "epoch": 2.5196182396606575,
      "grad_norm": 0.2426426261663437,
      "learning_rate": 3.3756001926838273e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 29250704,
      "step": 15445
    },
    {
      "epoch": 2.520433966881475,
      "grad_norm": 0.4317224323749542,
      "learning_rate": 3.374680561423737e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 29260976,
      "step": 15450
    },
    {
      "epoch": 2.521249694102292,
      "grad_norm": 0.4048636555671692,
      "learning_rate": 3.373760795276912e-05,
      "loss": 0.1806,
      "num_input_tokens_seen": 29270576,
      "step": 15455
    },
    {
      "epoch": 2.5220654213231093,
      "grad_norm": 0.38892385363578796,
      "learning_rate": 3.372840894385192e-05,
      "loss": 0.2147,
      "num_input_tokens_seen": 29279888,
      "step": 15460
    },
    {
      "epoch": 2.522881148543927,
      "grad_norm": 0.15478220582008362,
      "learning_rate": 3.3719208588904375e-05,
      "loss": 0.1111,
      "num_input_tokens_seen": 29287936,
      "step": 15465
    },
    {
      "epoch": 2.5236968757647444,
      "grad_norm": 0.678960919380188,
      "learning_rate": 3.371000688934529e-05,
      "loss": 0.1432,
      "num_input_tokens_seen": 29297216,
      "step": 15470
    },
    {
      "epoch": 2.5245126029855616,
      "grad_norm": 0.24900804460048676,
      "learning_rate": 3.370080384659369e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 29306848,
      "step": 15475
    },
    {
      "epoch": 2.525328330206379,
      "grad_norm": 0.6888859272003174,
      "learning_rate": 3.36915994620688e-05,
      "loss": 0.1451,
      "num_input_tokens_seen": 29316368,
      "step": 15480
    },
    {
      "epoch": 2.5261440574271963,
      "grad_norm": 0.14636529982089996,
      "learning_rate": 3.3682393737190035e-05,
      "loss": 0.0324,
      "num_input_tokens_seen": 29326032,
      "step": 15485
    },
    {
      "epoch": 2.526959784648014,
      "grad_norm": 1.8196303844451904,
      "learning_rate": 3.3673186673377054e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 29335328,
      "step": 15490
    },
    {
      "epoch": 2.527775511868831,
      "grad_norm": 0.8026204109191895,
      "learning_rate": 3.366397827204969e-05,
      "loss": 0.0668,
      "num_input_tokens_seen": 29345392,
      "step": 15495
    },
    {
      "epoch": 2.5285912390896486,
      "grad_norm": 0.39217084646224976,
      "learning_rate": 3.3654768534628e-05,
      "loss": 0.1207,
      "num_input_tokens_seen": 29354816,
      "step": 15500
    },
    {
      "epoch": 2.5294069663104657,
      "grad_norm": 1.8129830360412598,
      "learning_rate": 3.3645557462532245e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 29364928,
      "step": 15505
    },
    {
      "epoch": 2.5302226935312833,
      "grad_norm": 0.5334117412567139,
      "learning_rate": 3.363634505718288e-05,
      "loss": 0.0672,
      "num_input_tokens_seen": 29374576,
      "step": 15510
    },
    {
      "epoch": 2.5310384207521004,
      "grad_norm": 0.054405760020017624,
      "learning_rate": 3.362713132000057e-05,
      "loss": 0.0345,
      "num_input_tokens_seen": 29383760,
      "step": 15515
    },
    {
      "epoch": 2.531854147972918,
      "grad_norm": 0.733742892742157,
      "learning_rate": 3.36179162524062e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 29393248,
      "step": 15520
    },
    {
      "epoch": 2.532669875193735,
      "grad_norm": 1.2746719121932983,
      "learning_rate": 3.3608699855820846e-05,
      "loss": 0.1795,
      "num_input_tokens_seen": 29402960,
      "step": 15525
    },
    {
      "epoch": 2.5334856024145527,
      "grad_norm": 1.6671290397644043,
      "learning_rate": 3.359948213166578e-05,
      "loss": 0.1787,
      "num_input_tokens_seen": 29413344,
      "step": 15530
    },
    {
      "epoch": 2.53430132963537,
      "grad_norm": 0.9036718010902405,
      "learning_rate": 3.359026308136252e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 29422432,
      "step": 15535
    },
    {
      "epoch": 2.5351170568561874,
      "grad_norm": 0.5884495377540588,
      "learning_rate": 3.358104270633272e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 29431920,
      "step": 15540
    },
    {
      "epoch": 2.5359327840770045,
      "grad_norm": 0.22110344469547272,
      "learning_rate": 3.357182100799831e-05,
      "loss": 0.0444,
      "num_input_tokens_seen": 29441152,
      "step": 15545
    },
    {
      "epoch": 2.536748511297822,
      "grad_norm": 0.41346096992492676,
      "learning_rate": 3.3562597987781384e-05,
      "loss": 0.2623,
      "num_input_tokens_seen": 29451648,
      "step": 15550
    },
    {
      "epoch": 2.537564238518639,
      "grad_norm": 0.19220033288002014,
      "learning_rate": 3.355337364710424e-05,
      "loss": 0.0894,
      "num_input_tokens_seen": 29460256,
      "step": 15555
    },
    {
      "epoch": 2.538379965739457,
      "grad_norm": 0.9868831634521484,
      "learning_rate": 3.354414798738939e-05,
      "loss": 0.2133,
      "num_input_tokens_seen": 29468496,
      "step": 15560
    },
    {
      "epoch": 2.539195692960274,
      "grad_norm": 0.5840455889701843,
      "learning_rate": 3.353492101005955e-05,
      "loss": 0.048,
      "num_input_tokens_seen": 29477888,
      "step": 15565
    },
    {
      "epoch": 2.5400114201810915,
      "grad_norm": 0.502252995967865,
      "learning_rate": 3.352569271653763e-05,
      "loss": 0.1887,
      "num_input_tokens_seen": 29488224,
      "step": 15570
    },
    {
      "epoch": 2.5408271474019086,
      "grad_norm": 0.710688591003418,
      "learning_rate": 3.351646310824675e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 29497632,
      "step": 15575
    },
    {
      "epoch": 2.541642874622726,
      "grad_norm": 0.10559165477752686,
      "learning_rate": 3.350723218661023e-05,
      "loss": 0.1277,
      "num_input_tokens_seen": 29507936,
      "step": 15580
    },
    {
      "epoch": 2.5424586018435438,
      "grad_norm": 0.3961643576622009,
      "learning_rate": 3.349799995305162e-05,
      "loss": 0.0432,
      "num_input_tokens_seen": 29518048,
      "step": 15585
    },
    {
      "epoch": 2.543274329064361,
      "grad_norm": 0.6275065541267395,
      "learning_rate": 3.348876640899461e-05,
      "loss": 0.0695,
      "num_input_tokens_seen": 29526848,
      "step": 15590
    },
    {
      "epoch": 2.544090056285178,
      "grad_norm": 0.173539400100708,
      "learning_rate": 3.3479531555863144e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 29535024,
      "step": 15595
    },
    {
      "epoch": 2.5449057835059956,
      "grad_norm": 0.1939859539270401,
      "learning_rate": 3.3470295395081344e-05,
      "loss": 0.0712,
      "num_input_tokens_seen": 29545056,
      "step": 15600
    },
    {
      "epoch": 2.5449057835059956,
      "eval_loss": 0.14583754539489746,
      "eval_runtime": 133.2747,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 29545056,
      "step": 15600
    },
    {
      "epoch": 2.545721510726813,
      "grad_norm": 0.05922416225075722,
      "learning_rate": 3.3461057928073556e-05,
      "loss": 0.1151,
      "num_input_tokens_seen": 29553664,
      "step": 15605
    },
    {
      "epoch": 2.5465372379476303,
      "grad_norm": 0.12629006803035736,
      "learning_rate": 3.345181915626431e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 29563904,
      "step": 15610
    },
    {
      "epoch": 2.5473529651684474,
      "grad_norm": 0.2445676624774933,
      "learning_rate": 3.344257908107834e-05,
      "loss": 0.2609,
      "num_input_tokens_seen": 29573968,
      "step": 15615
    },
    {
      "epoch": 2.548168692389265,
      "grad_norm": 0.03590301796793938,
      "learning_rate": 3.343333770394058e-05,
      "loss": 0.1868,
      "num_input_tokens_seen": 29582768,
      "step": 15620
    },
    {
      "epoch": 2.5489844196100826,
      "grad_norm": 0.10552951693534851,
      "learning_rate": 3.342409502627616e-05,
      "loss": 0.146,
      "num_input_tokens_seen": 29591744,
      "step": 15625
    },
    {
      "epoch": 2.5498001468308997,
      "grad_norm": 0.5393683910369873,
      "learning_rate": 3.341485104951043e-05,
      "loss": 0.1204,
      "num_input_tokens_seen": 29599984,
      "step": 15630
    },
    {
      "epoch": 2.550615874051717,
      "grad_norm": 0.05557796359062195,
      "learning_rate": 3.340560577506892e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 29607936,
      "step": 15635
    },
    {
      "epoch": 2.5514316012725344,
      "grad_norm": 0.3486917018890381,
      "learning_rate": 3.339635920437735e-05,
      "loss": 0.2811,
      "num_input_tokens_seen": 29617680,
      "step": 15640
    },
    {
      "epoch": 2.552247328493352,
      "grad_norm": 0.32958245277404785,
      "learning_rate": 3.338711133886169e-05,
      "loss": 0.0495,
      "num_input_tokens_seen": 29627216,
      "step": 15645
    },
    {
      "epoch": 2.553063055714169,
      "grad_norm": 0.13643759489059448,
      "learning_rate": 3.3377862179948064e-05,
      "loss": 0.1084,
      "num_input_tokens_seen": 29636480,
      "step": 15650
    },
    {
      "epoch": 2.5538787829349863,
      "grad_norm": 0.06615734100341797,
      "learning_rate": 3.336861172906281e-05,
      "loss": 0.0734,
      "num_input_tokens_seen": 29646400,
      "step": 15655
    },
    {
      "epoch": 2.554694510155804,
      "grad_norm": 1.5293803215026855,
      "learning_rate": 3.335935998763245e-05,
      "loss": 0.208,
      "num_input_tokens_seen": 29655856,
      "step": 15660
    },
    {
      "epoch": 2.5555102373766214,
      "grad_norm": 0.5384883284568787,
      "learning_rate": 3.3350106957083744e-05,
      "loss": 0.1007,
      "num_input_tokens_seen": 29666336,
      "step": 15665
    },
    {
      "epoch": 2.5563259645974385,
      "grad_norm": 0.7469584345817566,
      "learning_rate": 3.33408526388436e-05,
      "loss": 0.1023,
      "num_input_tokens_seen": 29675504,
      "step": 15670
    },
    {
      "epoch": 2.557141691818256,
      "grad_norm": 0.30266711115837097,
      "learning_rate": 3.3331597034339166e-05,
      "loss": 0.1945,
      "num_input_tokens_seen": 29684192,
      "step": 15675
    },
    {
      "epoch": 2.5579574190390733,
      "grad_norm": 0.9227862358093262,
      "learning_rate": 3.3322340144997764e-05,
      "loss": 0.4174,
      "num_input_tokens_seen": 29693744,
      "step": 15680
    },
    {
      "epoch": 2.558773146259891,
      "grad_norm": 0.38840439915657043,
      "learning_rate": 3.331308197224693e-05,
      "loss": 0.1004,
      "num_input_tokens_seen": 29702224,
      "step": 15685
    },
    {
      "epoch": 2.559588873480708,
      "grad_norm": 0.5897827744483948,
      "learning_rate": 3.330382251751438e-05,
      "loss": 0.081,
      "num_input_tokens_seen": 29712304,
      "step": 15690
    },
    {
      "epoch": 2.5604046007015255,
      "grad_norm": 0.654518723487854,
      "learning_rate": 3.3294561782228054e-05,
      "loss": 0.0594,
      "num_input_tokens_seen": 29721680,
      "step": 15695
    },
    {
      "epoch": 2.5612203279223427,
      "grad_norm": 0.620415985584259,
      "learning_rate": 3.328529976781607e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 29730992,
      "step": 15700
    },
    {
      "epoch": 2.5620360551431602,
      "grad_norm": 1.4521141052246094,
      "learning_rate": 3.327603647570673e-05,
      "loss": 0.2166,
      "num_input_tokens_seen": 29740576,
      "step": 15705
    },
    {
      "epoch": 2.5628517823639774,
      "grad_norm": 0.16586387157440186,
      "learning_rate": 3.326677190732857e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 29751776,
      "step": 15710
    },
    {
      "epoch": 2.563667509584795,
      "grad_norm": 0.25125041604042053,
      "learning_rate": 3.325750606411029e-05,
      "loss": 0.2176,
      "num_input_tokens_seen": 29761840,
      "step": 15715
    },
    {
      "epoch": 2.564483236805612,
      "grad_norm": 1.3916901350021362,
      "learning_rate": 3.3248238947480804e-05,
      "loss": 0.1245,
      "num_input_tokens_seen": 29770784,
      "step": 15720
    },
    {
      "epoch": 2.5652989640264297,
      "grad_norm": 0.14800886809825897,
      "learning_rate": 3.323897055886922e-05,
      "loss": 0.097,
      "num_input_tokens_seen": 29779984,
      "step": 15725
    },
    {
      "epoch": 2.566114691247247,
      "grad_norm": 0.9388455152511597,
      "learning_rate": 3.322970089970484e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 29788800,
      "step": 15730
    },
    {
      "epoch": 2.5669304184680644,
      "grad_norm": 0.19589214026927948,
      "learning_rate": 3.3220429971417165e-05,
      "loss": 0.0481,
      "num_input_tokens_seen": 29797760,
      "step": 15735
    },
    {
      "epoch": 2.5677461456888815,
      "grad_norm": 0.28389787673950195,
      "learning_rate": 3.321115777543588e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 29805552,
      "step": 15740
    },
    {
      "epoch": 2.568561872909699,
      "grad_norm": 0.19957488775253296,
      "learning_rate": 3.320188431319088e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 29814512,
      "step": 15745
    },
    {
      "epoch": 2.569377600130516,
      "grad_norm": 0.9906616806983948,
      "learning_rate": 3.319260958611224e-05,
      "loss": 0.1336,
      "num_input_tokens_seen": 29823952,
      "step": 15750
    },
    {
      "epoch": 2.5701933273513338,
      "grad_norm": 1.7435396909713745,
      "learning_rate": 3.3183333595630256e-05,
      "loss": 0.0897,
      "num_input_tokens_seen": 29832592,
      "step": 15755
    },
    {
      "epoch": 2.5710090545721513,
      "grad_norm": 0.21319551765918732,
      "learning_rate": 3.317405634317538e-05,
      "loss": 0.2152,
      "num_input_tokens_seen": 29842288,
      "step": 15760
    },
    {
      "epoch": 2.5718247817929685,
      "grad_norm": 1.2952089309692383,
      "learning_rate": 3.3164777830178315e-05,
      "loss": 0.2258,
      "num_input_tokens_seen": 29852224,
      "step": 15765
    },
    {
      "epoch": 2.5726405090137856,
      "grad_norm": 0.8773311376571655,
      "learning_rate": 3.315549805806989e-05,
      "loss": 0.221,
      "num_input_tokens_seen": 29861296,
      "step": 15770
    },
    {
      "epoch": 2.573456236234603,
      "grad_norm": 0.6692203283309937,
      "learning_rate": 3.314621702828118e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 29870480,
      "step": 15775
    },
    {
      "epoch": 2.5742719634554208,
      "grad_norm": 1.522950291633606,
      "learning_rate": 3.313693474224342e-05,
      "loss": 0.2914,
      "num_input_tokens_seen": 29880880,
      "step": 15780
    },
    {
      "epoch": 2.575087690676238,
      "grad_norm": 0.47616615891456604,
      "learning_rate": 3.312765120138809e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 29891952,
      "step": 15785
    },
    {
      "epoch": 2.575903417897055,
      "grad_norm": 0.5408231616020203,
      "learning_rate": 3.311836640714679e-05,
      "loss": 0.0648,
      "num_input_tokens_seen": 29902528,
      "step": 15790
    },
    {
      "epoch": 2.5767191451178726,
      "grad_norm": 0.08361679315567017,
      "learning_rate": 3.310908036095137e-05,
      "loss": 0.0147,
      "num_input_tokens_seen": 29911616,
      "step": 15795
    },
    {
      "epoch": 2.57753487233869,
      "grad_norm": 0.13249975442886353,
      "learning_rate": 3.309979306423386e-05,
      "loss": 0.0906,
      "num_input_tokens_seen": 29922176,
      "step": 15800
    },
    {
      "epoch": 2.57753487233869,
      "eval_loss": 0.14718957245349884,
      "eval_runtime": 133.3211,
      "eval_samples_per_second": 20.439,
      "eval_steps_per_second": 10.223,
      "num_input_tokens_seen": 29922176,
      "step": 15800
    },
    {
      "epoch": 2.5783505995595073,
      "grad_norm": 0.35347992181777954,
      "learning_rate": 3.309050451842647e-05,
      "loss": 0.107,
      "num_input_tokens_seen": 29930912,
      "step": 15805
    },
    {
      "epoch": 2.5791663267803244,
      "grad_norm": 0.3917006254196167,
      "learning_rate": 3.3081214724961604e-05,
      "loss": 0.1339,
      "num_input_tokens_seen": 29940464,
      "step": 15810
    },
    {
      "epoch": 2.579982054001142,
      "grad_norm": 0.050537753850221634,
      "learning_rate": 3.307192368527188e-05,
      "loss": 0.0897,
      "num_input_tokens_seen": 29949632,
      "step": 15815
    },
    {
      "epoch": 2.5807977812219596,
      "grad_norm": 0.1072564348578453,
      "learning_rate": 3.306263140079008e-05,
      "loss": 0.1141,
      "num_input_tokens_seen": 29959008,
      "step": 15820
    },
    {
      "epoch": 2.5816135084427767,
      "grad_norm": 0.2170894593000412,
      "learning_rate": 3.30533378729492e-05,
      "loss": 0.032,
      "num_input_tokens_seen": 29968528,
      "step": 15825
    },
    {
      "epoch": 2.582429235663594,
      "grad_norm": 0.33416271209716797,
      "learning_rate": 3.304404310318242e-05,
      "loss": 0.1038,
      "num_input_tokens_seen": 29977488,
      "step": 15830
    },
    {
      "epoch": 2.5832449628844114,
      "grad_norm": 0.6638283729553223,
      "learning_rate": 3.3034747092923105e-05,
      "loss": 0.1413,
      "num_input_tokens_seen": 29987280,
      "step": 15835
    },
    {
      "epoch": 2.584060690105229,
      "grad_norm": 0.48329612612724304,
      "learning_rate": 3.3025449843604806e-05,
      "loss": 0.0882,
      "num_input_tokens_seen": 29996896,
      "step": 15840
    },
    {
      "epoch": 2.584876417326046,
      "grad_norm": 1.636694073677063,
      "learning_rate": 3.30161513566613e-05,
      "loss": 0.1704,
      "num_input_tokens_seen": 30006496,
      "step": 15845
    },
    {
      "epoch": 2.5856921445468637,
      "grad_norm": 0.3302052915096283,
      "learning_rate": 3.3006851633526506e-05,
      "loss": 0.0671,
      "num_input_tokens_seen": 30016032,
      "step": 15850
    },
    {
      "epoch": 2.586507871767681,
      "grad_norm": 0.6387620568275452,
      "learning_rate": 3.2997550675634584e-05,
      "loss": 0.1194,
      "num_input_tokens_seen": 30025168,
      "step": 15855
    },
    {
      "epoch": 2.5873235989884984,
      "grad_norm": 0.08716876059770584,
      "learning_rate": 3.2988248484419825e-05,
      "loss": 0.2232,
      "num_input_tokens_seen": 30034128,
      "step": 15860
    },
    {
      "epoch": 2.5881393262093155,
      "grad_norm": 0.9087420701980591,
      "learning_rate": 3.2978945061316776e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 30043504,
      "step": 15865
    },
    {
      "epoch": 2.588955053430133,
      "grad_norm": 0.27966877818107605,
      "learning_rate": 3.296964040776013e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 30052992,
      "step": 15870
    },
    {
      "epoch": 2.5897707806509502,
      "grad_norm": 0.09711278975009918,
      "learning_rate": 3.296033452518478e-05,
      "loss": 0.0399,
      "num_input_tokens_seen": 30062528,
      "step": 15875
    },
    {
      "epoch": 2.590586507871768,
      "grad_norm": 0.10384722054004669,
      "learning_rate": 3.2951027415025806e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 30072688,
      "step": 15880
    },
    {
      "epoch": 2.591402235092585,
      "grad_norm": 1.4806421995162964,
      "learning_rate": 3.294171907871849e-05,
      "loss": 0.231,
      "num_input_tokens_seen": 30080752,
      "step": 15885
    },
    {
      "epoch": 2.5922179623134025,
      "grad_norm": 0.16053546965122223,
      "learning_rate": 3.293240951769828e-05,
      "loss": 0.0977,
      "num_input_tokens_seen": 30089504,
      "step": 15890
    },
    {
      "epoch": 2.5930336895342196,
      "grad_norm": 1.0824023485183716,
      "learning_rate": 3.2923098733400846e-05,
      "loss": 0.1256,
      "num_input_tokens_seen": 30098912,
      "step": 15895
    },
    {
      "epoch": 2.593849416755037,
      "grad_norm": 0.022756287828087807,
      "learning_rate": 3.291378672726202e-05,
      "loss": 0.1915,
      "num_input_tokens_seen": 30108208,
      "step": 15900
    },
    {
      "epoch": 2.5946651439758543,
      "grad_norm": 0.7329471707344055,
      "learning_rate": 3.2904473500717824e-05,
      "loss": 0.1809,
      "num_input_tokens_seen": 30118352,
      "step": 15905
    },
    {
      "epoch": 2.595480871196672,
      "grad_norm": 0.5338101983070374,
      "learning_rate": 3.289515905520449e-05,
      "loss": 0.1222,
      "num_input_tokens_seen": 30126640,
      "step": 15910
    },
    {
      "epoch": 2.596296598417489,
      "grad_norm": 0.16853919625282288,
      "learning_rate": 3.288584339215841e-05,
      "loss": 0.0763,
      "num_input_tokens_seen": 30136800,
      "step": 15915
    },
    {
      "epoch": 2.5971123256383066,
      "grad_norm": 2.7282681465148926,
      "learning_rate": 3.287652651301617e-05,
      "loss": 0.2273,
      "num_input_tokens_seen": 30146480,
      "step": 15920
    },
    {
      "epoch": 2.5979280528591238,
      "grad_norm": 0.25544342398643494,
      "learning_rate": 3.286720841921457e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 30156448,
      "step": 15925
    },
    {
      "epoch": 2.5987437800799413,
      "grad_norm": 0.07359220832586288,
      "learning_rate": 3.285788911219056e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 30166048,
      "step": 15930
    },
    {
      "epoch": 2.5995595073007585,
      "grad_norm": 0.22193016111850739,
      "learning_rate": 3.284856859338131e-05,
      "loss": 0.026,
      "num_input_tokens_seen": 30175856,
      "step": 15935
    },
    {
      "epoch": 2.600375234521576,
      "grad_norm": 1.3239517211914062,
      "learning_rate": 3.283924686422414e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 30187008,
      "step": 15940
    },
    {
      "epoch": 2.601190961742393,
      "grad_norm": 0.799494206905365,
      "learning_rate": 3.282992392615659e-05,
      "loss": 0.2658,
      "num_input_tokens_seen": 30196384,
      "step": 15945
    },
    {
      "epoch": 2.6020066889632107,
      "grad_norm": 1.0479055643081665,
      "learning_rate": 3.282059978061638e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 30206848,
      "step": 15950
    },
    {
      "epoch": 2.6028224161840283,
      "grad_norm": 0.893173336982727,
      "learning_rate": 3.28112744290414e-05,
      "loss": 0.0591,
      "num_input_tokens_seen": 30216176,
      "step": 15955
    },
    {
      "epoch": 2.6036381434048455,
      "grad_norm": 0.27260127663612366,
      "learning_rate": 3.280194787286974e-05,
      "loss": 0.1619,
      "num_input_tokens_seen": 30226720,
      "step": 15960
    },
    {
      "epoch": 2.6044538706256626,
      "grad_norm": 0.4677249789237976,
      "learning_rate": 3.2792620113539674e-05,
      "loss": 0.1036,
      "num_input_tokens_seen": 30236832,
      "step": 15965
    },
    {
      "epoch": 2.60526959784648,
      "grad_norm": 0.5543109774589539,
      "learning_rate": 3.278329115248966e-05,
      "loss": 0.062,
      "num_input_tokens_seen": 30247424,
      "step": 15970
    },
    {
      "epoch": 2.6060853250672977,
      "grad_norm": 0.11144057661294937,
      "learning_rate": 3.277396099115834e-05,
      "loss": 0.0317,
      "num_input_tokens_seen": 30256960,
      "step": 15975
    },
    {
      "epoch": 2.606901052288115,
      "grad_norm": 0.45952746272087097,
      "learning_rate": 3.276462963098454e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 30266544,
      "step": 15980
    },
    {
      "epoch": 2.607716779508932,
      "grad_norm": 0.8831455111503601,
      "learning_rate": 3.275529707340728e-05,
      "loss": 0.1756,
      "num_input_tokens_seen": 30276848,
      "step": 15985
    },
    {
      "epoch": 2.6085325067297496,
      "grad_norm": 0.10604549944400787,
      "learning_rate": 3.274596331986574e-05,
      "loss": 0.0659,
      "num_input_tokens_seen": 30285696,
      "step": 15990
    },
    {
      "epoch": 2.609348233950567,
      "grad_norm": 0.23428307473659515,
      "learning_rate": 3.273662837179932e-05,
      "loss": 0.0724,
      "num_input_tokens_seen": 30295664,
      "step": 15995
    },
    {
      "epoch": 2.6101639611713843,
      "grad_norm": 0.14351986348628998,
      "learning_rate": 3.272729223064758e-05,
      "loss": 0.0529,
      "num_input_tokens_seen": 30304336,
      "step": 16000
    },
    {
      "epoch": 2.6101639611713843,
      "eval_loss": 0.14735989272594452,
      "eval_runtime": 133.262,
      "eval_samples_per_second": 20.448,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 30304336,
      "step": 16000
    },
    {
      "epoch": 2.6109796883922014,
      "grad_norm": 0.9203829765319824,
      "learning_rate": 3.2717954897850264e-05,
      "loss": 0.2317,
      "num_input_tokens_seen": 30313696,
      "step": 16005
    },
    {
      "epoch": 2.611795415613019,
      "grad_norm": 0.42403241991996765,
      "learning_rate": 3.270861637484733e-05,
      "loss": 0.0592,
      "num_input_tokens_seen": 30323168,
      "step": 16010
    },
    {
      "epoch": 2.6126111428338366,
      "grad_norm": 0.3944023549556732,
      "learning_rate": 3.2699276663078867e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 30333152,
      "step": 16015
    },
    {
      "epoch": 2.6134268700546537,
      "grad_norm": 1.1253200769424438,
      "learning_rate": 3.268993576398519e-05,
      "loss": 0.05,
      "num_input_tokens_seen": 30344176,
      "step": 16020
    },
    {
      "epoch": 2.614242597275471,
      "grad_norm": 0.24814295768737793,
      "learning_rate": 3.268059367900678e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 30354336,
      "step": 16025
    },
    {
      "epoch": 2.6150583244962884,
      "grad_norm": 0.5919486880302429,
      "learning_rate": 3.26712504095843e-05,
      "loss": 0.0364,
      "num_input_tokens_seen": 30363552,
      "step": 16030
    },
    {
      "epoch": 2.615874051717106,
      "grad_norm": 1.1130542755126953,
      "learning_rate": 3.2661905957158615e-05,
      "loss": 0.0872,
      "num_input_tokens_seen": 30373392,
      "step": 16035
    },
    {
      "epoch": 2.616689778937923,
      "grad_norm": 1.4216208457946777,
      "learning_rate": 3.2652560323170734e-05,
      "loss": 0.2179,
      "num_input_tokens_seen": 30383408,
      "step": 16040
    },
    {
      "epoch": 2.6175055061587407,
      "grad_norm": 0.6549538969993591,
      "learning_rate": 3.264321350906189e-05,
      "loss": 0.0787,
      "num_input_tokens_seen": 30391760,
      "step": 16045
    },
    {
      "epoch": 2.618321233379558,
      "grad_norm": 0.7824116945266724,
      "learning_rate": 3.263386551627346e-05,
      "loss": 0.0614,
      "num_input_tokens_seen": 30400400,
      "step": 16050
    },
    {
      "epoch": 2.6191369606003754,
      "grad_norm": 0.06082388013601303,
      "learning_rate": 3.2624516346247055e-05,
      "loss": 0.3233,
      "num_input_tokens_seen": 30409984,
      "step": 16055
    },
    {
      "epoch": 2.6199526878211925,
      "grad_norm": 1.5762765407562256,
      "learning_rate": 3.2615166000424404e-05,
      "loss": 0.179,
      "num_input_tokens_seen": 30418224,
      "step": 16060
    },
    {
      "epoch": 2.62076841504201,
      "grad_norm": 1.2988436222076416,
      "learning_rate": 3.260581448024745e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 30427040,
      "step": 16065
    },
    {
      "epoch": 2.621584142262827,
      "grad_norm": 0.0871456116437912,
      "learning_rate": 3.2596461787158335e-05,
      "loss": 0.0553,
      "num_input_tokens_seen": 30436640,
      "step": 16070
    },
    {
      "epoch": 2.622399869483645,
      "grad_norm": 0.27804163098335266,
      "learning_rate": 3.258710792259934e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 30446288,
      "step": 16075
    },
    {
      "epoch": 2.623215596704462,
      "grad_norm": 0.31501808762550354,
      "learning_rate": 3.257775288801296e-05,
      "loss": 0.2125,
      "num_input_tokens_seen": 30456080,
      "step": 16080
    },
    {
      "epoch": 2.6240313239252795,
      "grad_norm": 0.9715969562530518,
      "learning_rate": 3.256839668484186e-05,
      "loss": 0.2196,
      "num_input_tokens_seen": 30465552,
      "step": 16085
    },
    {
      "epoch": 2.6248470511460966,
      "grad_norm": 0.10751515626907349,
      "learning_rate": 3.255903931452888e-05,
      "loss": 0.1077,
      "num_input_tokens_seen": 30475488,
      "step": 16090
    },
    {
      "epoch": 2.625662778366914,
      "grad_norm": 0.1700371354818344,
      "learning_rate": 3.2549680778517045e-05,
      "loss": 0.1068,
      "num_input_tokens_seen": 30484656,
      "step": 16095
    },
    {
      "epoch": 2.6264785055877313,
      "grad_norm": 2.9543635845184326,
      "learning_rate": 3.2540321078249556e-05,
      "loss": 0.329,
      "num_input_tokens_seen": 30494848,
      "step": 16100
    },
    {
      "epoch": 2.627294232808549,
      "grad_norm": 0.7482928037643433,
      "learning_rate": 3.2530960215169795e-05,
      "loss": 0.0814,
      "num_input_tokens_seen": 30503712,
      "step": 16105
    },
    {
      "epoch": 2.628109960029366,
      "grad_norm": 0.5995243787765503,
      "learning_rate": 3.2521598190721345e-05,
      "loss": 0.0906,
      "num_input_tokens_seen": 30512240,
      "step": 16110
    },
    {
      "epoch": 2.6289256872501836,
      "grad_norm": 0.47055426239967346,
      "learning_rate": 3.251223500634792e-05,
      "loss": 0.1065,
      "num_input_tokens_seen": 30520464,
      "step": 16115
    },
    {
      "epoch": 2.6297414144710007,
      "grad_norm": 0.14982321858406067,
      "learning_rate": 3.2502870663493445e-05,
      "loss": 0.0798,
      "num_input_tokens_seen": 30530560,
      "step": 16120
    },
    {
      "epoch": 2.6305571416918183,
      "grad_norm": 0.7457962036132812,
      "learning_rate": 3.249350516360203e-05,
      "loss": 0.0303,
      "num_input_tokens_seen": 30540448,
      "step": 16125
    },
    {
      "epoch": 2.631372868912636,
      "grad_norm": 0.3220919072628021,
      "learning_rate": 3.248413850811797e-05,
      "loss": 0.0913,
      "num_input_tokens_seen": 30550944,
      "step": 16130
    },
    {
      "epoch": 2.632188596133453,
      "grad_norm": 0.7050184607505798,
      "learning_rate": 3.2474770698485677e-05,
      "loss": 0.1028,
      "num_input_tokens_seen": 30560368,
      "step": 16135
    },
    {
      "epoch": 2.63300432335427,
      "grad_norm": 0.34754258394241333,
      "learning_rate": 3.246540173614983e-05,
      "loss": 0.1037,
      "num_input_tokens_seen": 30569024,
      "step": 16140
    },
    {
      "epoch": 2.6338200505750877,
      "grad_norm": 1.1574792861938477,
      "learning_rate": 3.2456031622555197e-05,
      "loss": 0.098,
      "num_input_tokens_seen": 30580304,
      "step": 16145
    },
    {
      "epoch": 2.6346357777959053,
      "grad_norm": 1.6001098155975342,
      "learning_rate": 3.2446660359146794e-05,
      "loss": 0.2756,
      "num_input_tokens_seen": 30590096,
      "step": 16150
    },
    {
      "epoch": 2.6354515050167224,
      "grad_norm": 0.074282206594944,
      "learning_rate": 3.2437287947369786e-05,
      "loss": 0.1352,
      "num_input_tokens_seen": 30600032,
      "step": 16155
    },
    {
      "epoch": 2.6362672322375396,
      "grad_norm": 0.27315300703048706,
      "learning_rate": 3.2427914388669525e-05,
      "loss": 0.0759,
      "num_input_tokens_seen": 30609744,
      "step": 16160
    },
    {
      "epoch": 2.637082959458357,
      "grad_norm": 0.720628559589386,
      "learning_rate": 3.241853968449151e-05,
      "loss": 0.1591,
      "num_input_tokens_seen": 30619072,
      "step": 16165
    },
    {
      "epoch": 2.6378986866791747,
      "grad_norm": 1.609959363937378,
      "learning_rate": 3.240916383628144e-05,
      "loss": 0.2181,
      "num_input_tokens_seen": 30628656,
      "step": 16170
    },
    {
      "epoch": 2.638714413899992,
      "grad_norm": 0.4187566041946411,
      "learning_rate": 3.239978684548521e-05,
      "loss": 0.0875,
      "num_input_tokens_seen": 30638304,
      "step": 16175
    },
    {
      "epoch": 2.639530141120809,
      "grad_norm": 0.26589852571487427,
      "learning_rate": 3.239040871354885e-05,
      "loss": 0.0781,
      "num_input_tokens_seen": 30648336,
      "step": 16180
    },
    {
      "epoch": 2.6403458683416265,
      "grad_norm": 1.0661122798919678,
      "learning_rate": 3.2381029441918596e-05,
      "loss": 0.0484,
      "num_input_tokens_seen": 30658000,
      "step": 16185
    },
    {
      "epoch": 2.641161595562444,
      "grad_norm": 0.14037616550922394,
      "learning_rate": 3.2371649032040845e-05,
      "loss": 0.0838,
      "num_input_tokens_seen": 30667920,
      "step": 16190
    },
    {
      "epoch": 2.6419773227832613,
      "grad_norm": 0.29273396730422974,
      "learning_rate": 3.2362267485362174e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 30677808,
      "step": 16195
    },
    {
      "epoch": 2.6427930500040784,
      "grad_norm": 0.19632530212402344,
      "learning_rate": 3.235288480332934e-05,
      "loss": 0.0921,
      "num_input_tokens_seen": 30688608,
      "step": 16200
    },
    {
      "epoch": 2.6427930500040784,
      "eval_loss": 0.14861468970775604,
      "eval_runtime": 133.306,
      "eval_samples_per_second": 20.442,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 30688608,
      "step": 16200
    },
    {
      "epoch": 2.643608777224896,
      "grad_norm": 0.9615724086761475,
      "learning_rate": 3.234350098738927e-05,
      "loss": 0.0664,
      "num_input_tokens_seen": 30698352,
      "step": 16205
    },
    {
      "epoch": 2.6444245044457135,
      "grad_norm": 1.5938799381256104,
      "learning_rate": 3.233411603898906e-05,
      "loss": 0.2585,
      "num_input_tokens_seen": 30707952,
      "step": 16210
    },
    {
      "epoch": 2.6452402316665307,
      "grad_norm": 1.1350289583206177,
      "learning_rate": 3.232472995957599e-05,
      "loss": 0.1218,
      "num_input_tokens_seen": 30717424,
      "step": 16215
    },
    {
      "epoch": 2.6460559588873482,
      "grad_norm": 1.293976902961731,
      "learning_rate": 3.231534275059751e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 30726720,
      "step": 16220
    },
    {
      "epoch": 2.6468716861081654,
      "grad_norm": 0.48503348231315613,
      "learning_rate": 3.230595441350125e-05,
      "loss": 0.2086,
      "num_input_tokens_seen": 30736480,
      "step": 16225
    },
    {
      "epoch": 2.647687413328983,
      "grad_norm": 0.22796808183193207,
      "learning_rate": 3.2296564949735e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 30745680,
      "step": 16230
    },
    {
      "epoch": 2.6485031405498,
      "grad_norm": 0.6616711020469666,
      "learning_rate": 3.228717436074675e-05,
      "loss": 0.2009,
      "num_input_tokens_seen": 30754736,
      "step": 16235
    },
    {
      "epoch": 2.6493188677706176,
      "grad_norm": 0.057618647813797,
      "learning_rate": 3.227778264798463e-05,
      "loss": 0.1192,
      "num_input_tokens_seen": 30763392,
      "step": 16240
    },
    {
      "epoch": 2.6501345949914348,
      "grad_norm": 0.7745248079299927,
      "learning_rate": 3.226838981289698e-05,
      "loss": 0.1085,
      "num_input_tokens_seen": 30773408,
      "step": 16245
    },
    {
      "epoch": 2.6509503222122524,
      "grad_norm": 0.07966292649507523,
      "learning_rate": 3.225899585693227e-05,
      "loss": 0.0447,
      "num_input_tokens_seen": 30782656,
      "step": 16250
    },
    {
      "epoch": 2.6517660494330695,
      "grad_norm": 0.4043302536010742,
      "learning_rate": 3.224960078153918e-05,
      "loss": 0.1166,
      "num_input_tokens_seen": 30793136,
      "step": 16255
    },
    {
      "epoch": 2.652581776653887,
      "grad_norm": 0.09945771843194962,
      "learning_rate": 3.224020458816655e-05,
      "loss": 0.206,
      "num_input_tokens_seen": 30801760,
      "step": 16260
    },
    {
      "epoch": 2.653397503874704,
      "grad_norm": 2.3694164752960205,
      "learning_rate": 3.223080727826337e-05,
      "loss": 0.2604,
      "num_input_tokens_seen": 30811728,
      "step": 16265
    },
    {
      "epoch": 2.6542132310955218,
      "grad_norm": 0.22048574686050415,
      "learning_rate": 3.222140885327885e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 30821840,
      "step": 16270
    },
    {
      "epoch": 2.655028958316339,
      "grad_norm": 0.30732282996177673,
      "learning_rate": 3.221200931466234e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 30831632,
      "step": 16275
    },
    {
      "epoch": 2.6558446855371565,
      "grad_norm": 1.0366579294204712,
      "learning_rate": 3.220260866386336e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 30840976,
      "step": 16280
    },
    {
      "epoch": 2.6566604127579736,
      "grad_norm": 0.6896691918373108,
      "learning_rate": 3.21932069023316e-05,
      "loss": 0.0587,
      "num_input_tokens_seen": 30850576,
      "step": 16285
    },
    {
      "epoch": 2.657476139978791,
      "grad_norm": 0.15622776746749878,
      "learning_rate": 3.218380403151695e-05,
      "loss": 0.2223,
      "num_input_tokens_seen": 30860032,
      "step": 16290
    },
    {
      "epoch": 2.6582918671996083,
      "grad_norm": 0.22800540924072266,
      "learning_rate": 3.217440005286943e-05,
      "loss": 0.0565,
      "num_input_tokens_seen": 30870224,
      "step": 16295
    },
    {
      "epoch": 2.659107594420426,
      "grad_norm": 0.2764008641242981,
      "learning_rate": 3.216499496783928e-05,
      "loss": 0.1067,
      "num_input_tokens_seen": 30879728,
      "step": 16300
    },
    {
      "epoch": 2.659923321641243,
      "grad_norm": 0.2057814747095108,
      "learning_rate": 3.2155588777876856e-05,
      "loss": 0.0492,
      "num_input_tokens_seen": 30887600,
      "step": 16305
    },
    {
      "epoch": 2.6607390488620606,
      "grad_norm": 0.6264275908470154,
      "learning_rate": 3.214618148443273e-05,
      "loss": 0.0473,
      "num_input_tokens_seen": 30896752,
      "step": 16310
    },
    {
      "epoch": 2.6615547760828777,
      "grad_norm": 0.20306003093719482,
      "learning_rate": 3.2136773088957595e-05,
      "loss": 0.2289,
      "num_input_tokens_seen": 30905664,
      "step": 16315
    },
    {
      "epoch": 2.6623705033036953,
      "grad_norm": 0.26259317994117737,
      "learning_rate": 3.2127363592902374e-05,
      "loss": 0.1921,
      "num_input_tokens_seen": 30916240,
      "step": 16320
    },
    {
      "epoch": 2.663186230524513,
      "grad_norm": 0.9569255113601685,
      "learning_rate": 3.211795299771812e-05,
      "loss": 0.1796,
      "num_input_tokens_seen": 30924656,
      "step": 16325
    },
    {
      "epoch": 2.66400195774533,
      "grad_norm": 0.376699835062027,
      "learning_rate": 3.210854130485605e-05,
      "loss": 0.1088,
      "num_input_tokens_seen": 30934832,
      "step": 16330
    },
    {
      "epoch": 2.664817684966147,
      "grad_norm": 1.6397548913955688,
      "learning_rate": 3.209912851576759e-05,
      "loss": 0.2716,
      "num_input_tokens_seen": 30943248,
      "step": 16335
    },
    {
      "epoch": 2.6656334121869647,
      "grad_norm": 0.06474941223859787,
      "learning_rate": 3.208971463190431e-05,
      "loss": 0.0978,
      "num_input_tokens_seen": 30952720,
      "step": 16340
    },
    {
      "epoch": 2.6664491394077823,
      "grad_norm": 0.6340420246124268,
      "learning_rate": 3.208029965471793e-05,
      "loss": 0.1959,
      "num_input_tokens_seen": 30963056,
      "step": 16345
    },
    {
      "epoch": 2.6672648666285994,
      "grad_norm": 1.0123279094696045,
      "learning_rate": 3.2070883585660364e-05,
      "loss": 0.2399,
      "num_input_tokens_seen": 30973520,
      "step": 16350
    },
    {
      "epoch": 2.6680805938494165,
      "grad_norm": 0.433437317609787,
      "learning_rate": 3.20614664261837e-05,
      "loss": 0.0779,
      "num_input_tokens_seen": 30982064,
      "step": 16355
    },
    {
      "epoch": 2.668896321070234,
      "grad_norm": 0.07287941128015518,
      "learning_rate": 3.205204817774016e-05,
      "loss": 0.0703,
      "num_input_tokens_seen": 30991488,
      "step": 16360
    },
    {
      "epoch": 2.6697120482910517,
      "grad_norm": 0.8922070264816284,
      "learning_rate": 3.204262884178218e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 31002192,
      "step": 16365
    },
    {
      "epoch": 2.670527775511869,
      "grad_norm": 0.2889026701450348,
      "learning_rate": 3.2033208419762314e-05,
      "loss": 0.1645,
      "num_input_tokens_seen": 31011488,
      "step": 16370
    },
    {
      "epoch": 2.671343502732686,
      "grad_norm": 0.2446160465478897,
      "learning_rate": 3.2023786913133344e-05,
      "loss": 0.0871,
      "num_input_tokens_seen": 31020352,
      "step": 16375
    },
    {
      "epoch": 2.6721592299535035,
      "grad_norm": 0.4524873197078705,
      "learning_rate": 3.201436432334816e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 31029824,
      "step": 16380
    },
    {
      "epoch": 2.672974957174321,
      "grad_norm": 0.5480073690414429,
      "learning_rate": 3.2004940651859844e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 31039808,
      "step": 16385
    },
    {
      "epoch": 2.6737906843951382,
      "grad_norm": 2.021374225616455,
      "learning_rate": 3.1995515900121655e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 31049696,
      "step": 16390
    },
    {
      "epoch": 2.6746064116159554,
      "grad_norm": 0.08559805899858475,
      "learning_rate": 3.1986090069587e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 31059584,
      "step": 16395
    },
    {
      "epoch": 2.675422138836773,
      "grad_norm": 0.13706178963184357,
      "learning_rate": 3.1976663161709466e-05,
      "loss": 0.2052,
      "num_input_tokens_seen": 31067744,
      "step": 16400
    },
    {
      "epoch": 2.675422138836773,
      "eval_loss": 0.14604440331459045,
      "eval_runtime": 133.36,
      "eval_samples_per_second": 20.433,
      "eval_steps_per_second": 10.22,
      "num_input_tokens_seen": 31067744,
      "step": 16400
    },
    {
      "epoch": 2.6762378660575905,
      "grad_norm": 0.25292181968688965,
      "learning_rate": 3.196723517794279e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 31077808,
      "step": 16405
    },
    {
      "epoch": 2.6770535932784076,
      "grad_norm": 0.6958163976669312,
      "learning_rate": 3.19578061197409e-05,
      "loss": 0.2671,
      "num_input_tokens_seen": 31087936,
      "step": 16410
    },
    {
      "epoch": 2.677869320499225,
      "grad_norm": 0.5129421353340149,
      "learning_rate": 3.194837598855787e-05,
      "loss": 0.2954,
      "num_input_tokens_seen": 31097312,
      "step": 16415
    },
    {
      "epoch": 2.6786850477200423,
      "grad_norm": 0.3482704162597656,
      "learning_rate": 3.193894478584794e-05,
      "loss": 0.2357,
      "num_input_tokens_seen": 31106816,
      "step": 16420
    },
    {
      "epoch": 2.67950077494086,
      "grad_norm": 0.950901985168457,
      "learning_rate": 3.192951251306553e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 31116640,
      "step": 16425
    },
    {
      "epoch": 2.680316502161677,
      "grad_norm": 0.07921092957258224,
      "learning_rate": 3.192007917166521e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 31126736,
      "step": 16430
    },
    {
      "epoch": 2.6811322293824946,
      "grad_norm": 0.30422112345695496,
      "learning_rate": 3.191064476310171e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 31136512,
      "step": 16435
    },
    {
      "epoch": 2.6819479566033118,
      "grad_norm": 0.1961301863193512,
      "learning_rate": 3.1901209288829944e-05,
      "loss": 0.4455,
      "num_input_tokens_seen": 31146544,
      "step": 16440
    },
    {
      "epoch": 2.6827636838241293,
      "grad_norm": 0.08656345307826996,
      "learning_rate": 3.1891772750304985e-05,
      "loss": 0.1107,
      "num_input_tokens_seen": 31155744,
      "step": 16445
    },
    {
      "epoch": 2.6835794110449465,
      "grad_norm": 1.3819586038589478,
      "learning_rate": 3.188233514898206e-05,
      "loss": 0.1827,
      "num_input_tokens_seen": 31165136,
      "step": 16450
    },
    {
      "epoch": 2.684395138265764,
      "grad_norm": 1.6916691064834595,
      "learning_rate": 3.187289648631657e-05,
      "loss": 0.2455,
      "num_input_tokens_seen": 31175600,
      "step": 16455
    },
    {
      "epoch": 2.685210865486581,
      "grad_norm": 0.25494834780693054,
      "learning_rate": 3.186345676376406e-05,
      "loss": 0.1218,
      "num_input_tokens_seen": 31185584,
      "step": 16460
    },
    {
      "epoch": 2.6860265927073987,
      "grad_norm": 0.7150965929031372,
      "learning_rate": 3.1854015982780275e-05,
      "loss": 0.1808,
      "num_input_tokens_seen": 31195824,
      "step": 16465
    },
    {
      "epoch": 2.686842319928216,
      "grad_norm": 0.07273370772600174,
      "learning_rate": 3.1844574144821084e-05,
      "loss": 0.0433,
      "num_input_tokens_seen": 31205760,
      "step": 16470
    },
    {
      "epoch": 2.6876580471490334,
      "grad_norm": 0.32850033044815063,
      "learning_rate": 3.1835131251342554e-05,
      "loss": 0.0242,
      "num_input_tokens_seen": 31215824,
      "step": 16475
    },
    {
      "epoch": 2.6884737743698506,
      "grad_norm": 0.7584575414657593,
      "learning_rate": 3.182568730380089e-05,
      "loss": 0.0949,
      "num_input_tokens_seen": 31224016,
      "step": 16480
    },
    {
      "epoch": 2.689289501590668,
      "grad_norm": 0.4026753008365631,
      "learning_rate": 3.181624230365245e-05,
      "loss": 0.1094,
      "num_input_tokens_seen": 31233600,
      "step": 16485
    },
    {
      "epoch": 2.6901052288114853,
      "grad_norm": 0.553360641002655,
      "learning_rate": 3.180679625235381e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 31242784,
      "step": 16490
    },
    {
      "epoch": 2.690920956032303,
      "grad_norm": 1.1546695232391357,
      "learning_rate": 3.1797349151361646e-05,
      "loss": 0.0716,
      "num_input_tokens_seen": 31252688,
      "step": 16495
    },
    {
      "epoch": 2.6917366832531204,
      "grad_norm": 0.33215537667274475,
      "learning_rate": 3.178790100213281e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 31262240,
      "step": 16500
    },
    {
      "epoch": 2.6925524104739376,
      "grad_norm": 0.2056816965341568,
      "learning_rate": 3.1778451806124346e-05,
      "loss": 0.2153,
      "num_input_tokens_seen": 31272880,
      "step": 16505
    },
    {
      "epoch": 2.6933681376947547,
      "grad_norm": 0.9516927599906921,
      "learning_rate": 3.176900156479342e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 31283600,
      "step": 16510
    },
    {
      "epoch": 2.6941838649155723,
      "grad_norm": 1.0862133502960205,
      "learning_rate": 3.17595502795974e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 31292528,
      "step": 16515
    },
    {
      "epoch": 2.69499959213639,
      "grad_norm": 1.5822159051895142,
      "learning_rate": 3.175009795199377e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 31301584,
      "step": 16520
    },
    {
      "epoch": 2.695815319357207,
      "grad_norm": 0.9862978458404541,
      "learning_rate": 3.1740644583440224e-05,
      "loss": 0.1174,
      "num_input_tokens_seen": 31311472,
      "step": 16525
    },
    {
      "epoch": 2.696631046578024,
      "grad_norm": 1.394661784172058,
      "learning_rate": 3.173119017539457e-05,
      "loss": 0.2515,
      "num_input_tokens_seen": 31321456,
      "step": 16530
    },
    {
      "epoch": 2.6974467737988417,
      "grad_norm": 0.5457894802093506,
      "learning_rate": 3.172173472931479e-05,
      "loss": 0.0717,
      "num_input_tokens_seen": 31331472,
      "step": 16535
    },
    {
      "epoch": 2.6982625010196593,
      "grad_norm": 0.2502395808696747,
      "learning_rate": 3.1712278246659055e-05,
      "loss": 0.0845,
      "num_input_tokens_seen": 31340416,
      "step": 16540
    },
    {
      "epoch": 2.6990782282404764,
      "grad_norm": 0.2121029645204544,
      "learning_rate": 3.170282072888566e-05,
      "loss": 0.0746,
      "num_input_tokens_seen": 31348784,
      "step": 16545
    },
    {
      "epoch": 2.6998939554612935,
      "grad_norm": 0.9518917798995972,
      "learning_rate": 3.169336217745307e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 31358784,
      "step": 16550
    },
    {
      "epoch": 2.700709682682111,
      "grad_norm": 0.02664671465754509,
      "learning_rate": 3.1683902593819924e-05,
      "loss": 0.0482,
      "num_input_tokens_seen": 31369376,
      "step": 16555
    },
    {
      "epoch": 2.7015254099029287,
      "grad_norm": 0.10263797640800476,
      "learning_rate": 3.1674441979445e-05,
      "loss": 0.0271,
      "num_input_tokens_seen": 31379408,
      "step": 16560
    },
    {
      "epoch": 2.702341137123746,
      "grad_norm": 1.4678608179092407,
      "learning_rate": 3.166498033578725e-05,
      "loss": 0.1982,
      "num_input_tokens_seen": 31389248,
      "step": 16565
    },
    {
      "epoch": 2.703156864344563,
      "grad_norm": 0.3473377823829651,
      "learning_rate": 3.165551766430578e-05,
      "loss": 0.2057,
      "num_input_tokens_seen": 31398720,
      "step": 16570
    },
    {
      "epoch": 2.7039725915653805,
      "grad_norm": 1.1789568662643433,
      "learning_rate": 3.164605396645984e-05,
      "loss": 0.1102,
      "num_input_tokens_seen": 31409232,
      "step": 16575
    },
    {
      "epoch": 2.704788318786198,
      "grad_norm": 0.8230952620506287,
      "learning_rate": 3.163658924370886e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 31419584,
      "step": 16580
    },
    {
      "epoch": 2.705604046007015,
      "grad_norm": 0.18593324720859528,
      "learning_rate": 3.1627123497512415e-05,
      "loss": 0.2195,
      "num_input_tokens_seen": 31429088,
      "step": 16585
    },
    {
      "epoch": 2.7064197732278323,
      "grad_norm": 0.4116280674934387,
      "learning_rate": 3.1617656729330245e-05,
      "loss": 0.1918,
      "num_input_tokens_seen": 31437360,
      "step": 16590
    },
    {
      "epoch": 2.70723550044865,
      "grad_norm": 0.7173841595649719,
      "learning_rate": 3.1608188940622255e-05,
      "loss": 0.0772,
      "num_input_tokens_seen": 31445760,
      "step": 16595
    },
    {
      "epoch": 2.7080512276694675,
      "grad_norm": 1.7682186365127563,
      "learning_rate": 3.159872013284847e-05,
      "loss": 0.104,
      "num_input_tokens_seen": 31455328,
      "step": 16600
    },
    {
      "epoch": 2.7080512276694675,
      "eval_loss": 0.14793835580348969,
      "eval_runtime": 133.2939,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 31455328,
      "step": 16600
    },
    {
      "epoch": 2.7088669548902846,
      "grad_norm": 0.037861548364162445,
      "learning_rate": 3.1589250307469134e-05,
      "loss": 0.1256,
      "num_input_tokens_seen": 31464944,
      "step": 16605
    },
    {
      "epoch": 2.709682682111102,
      "grad_norm": 0.13041073083877563,
      "learning_rate": 3.1579779465944586e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 31475088,
      "step": 16610
    },
    {
      "epoch": 2.7104984093319193,
      "grad_norm": 0.16270345449447632,
      "learning_rate": 3.1570307609735363e-05,
      "loss": 0.167,
      "num_input_tokens_seen": 31484880,
      "step": 16615
    },
    {
      "epoch": 2.711314136552737,
      "grad_norm": 1.0203757286071777,
      "learning_rate": 3.156083474030213e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 31495216,
      "step": 16620
    },
    {
      "epoch": 2.712129863773554,
      "grad_norm": 0.033942729234695435,
      "learning_rate": 3.155136085910573e-05,
      "loss": 0.0717,
      "num_input_tokens_seen": 31505264,
      "step": 16625
    },
    {
      "epoch": 2.7129455909943716,
      "grad_norm": 1.3817708492279053,
      "learning_rate": 3.154188596760717e-05,
      "loss": 0.2864,
      "num_input_tokens_seen": 31514624,
      "step": 16630
    },
    {
      "epoch": 2.7137613182151887,
      "grad_norm": 1.0524338483810425,
      "learning_rate": 3.153241006726757e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 31523984,
      "step": 16635
    },
    {
      "epoch": 2.7145770454360063,
      "grad_norm": 0.08077523112297058,
      "learning_rate": 3.152293315954825e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 31533600,
      "step": 16640
    },
    {
      "epoch": 2.7153927726568234,
      "grad_norm": 0.3327764868736267,
      "learning_rate": 3.1513455245910666e-05,
      "loss": 0.2631,
      "num_input_tokens_seen": 31542208,
      "step": 16645
    },
    {
      "epoch": 2.716208499877641,
      "grad_norm": 0.5153088569641113,
      "learning_rate": 3.150397632781643e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 31552112,
      "step": 16650
    },
    {
      "epoch": 2.717024227098458,
      "grad_norm": 0.044475968927145004,
      "learning_rate": 3.149449640672731e-05,
      "loss": 0.0589,
      "num_input_tokens_seen": 31560864,
      "step": 16655
    },
    {
      "epoch": 2.7178399543192757,
      "grad_norm": 0.392887145280838,
      "learning_rate": 3.148501548410523e-05,
      "loss": 0.2833,
      "num_input_tokens_seen": 31570608,
      "step": 16660
    },
    {
      "epoch": 2.718655681540093,
      "grad_norm": 0.05198994651436806,
      "learning_rate": 3.1475533561412256e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 31579536,
      "step": 16665
    },
    {
      "epoch": 2.7194714087609104,
      "grad_norm": 0.9432244300842285,
      "learning_rate": 3.146605064011065e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 31588944,
      "step": 16670
    },
    {
      "epoch": 2.7202871359817276,
      "grad_norm": 0.17730455100536346,
      "learning_rate": 3.145656672166277e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 31597344,
      "step": 16675
    },
    {
      "epoch": 2.721102863202545,
      "grad_norm": 0.3415593206882477,
      "learning_rate": 3.144708180753116e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 31606848,
      "step": 16680
    },
    {
      "epoch": 2.7219185904233623,
      "grad_norm": 0.3058873116970062,
      "learning_rate": 3.143759589917851e-05,
      "loss": 0.0566,
      "num_input_tokens_seen": 31616288,
      "step": 16685
    },
    {
      "epoch": 2.72273431764418,
      "grad_norm": 0.38844990730285645,
      "learning_rate": 3.142810899806768e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 31625424,
      "step": 16690
    },
    {
      "epoch": 2.7235500448649974,
      "grad_norm": 1.8019276857376099,
      "learning_rate": 3.141862110566166e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 31633376,
      "step": 16695
    },
    {
      "epoch": 2.7243657720858145,
      "grad_norm": 0.035663194954395294,
      "learning_rate": 3.1409132223423606e-05,
      "loss": 0.0712,
      "num_input_tokens_seen": 31643520,
      "step": 16700
    },
    {
      "epoch": 2.7251814993066317,
      "grad_norm": 0.30535393953323364,
      "learning_rate": 3.139964235281682e-05,
      "loss": 0.2567,
      "num_input_tokens_seen": 31653216,
      "step": 16705
    },
    {
      "epoch": 2.7259972265274492,
      "grad_norm": 0.6096274852752686,
      "learning_rate": 3.139015149530476e-05,
      "loss": 0.0904,
      "num_input_tokens_seen": 31662352,
      "step": 16710
    },
    {
      "epoch": 2.726812953748267,
      "grad_norm": 1.3457057476043701,
      "learning_rate": 3.1380659652351034e-05,
      "loss": 0.0643,
      "num_input_tokens_seen": 31670880,
      "step": 16715
    },
    {
      "epoch": 2.727628680969084,
      "grad_norm": 0.3158363401889801,
      "learning_rate": 3.137116682541941e-05,
      "loss": 0.2692,
      "num_input_tokens_seen": 31680912,
      "step": 16720
    },
    {
      "epoch": 2.728444408189901,
      "grad_norm": 0.4538556933403015,
      "learning_rate": 3.136167301597379e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 31690016,
      "step": 16725
    },
    {
      "epoch": 2.7292601354107187,
      "grad_norm": 0.22106066346168518,
      "learning_rate": 3.1352178225478254e-05,
      "loss": 0.1287,
      "num_input_tokens_seen": 31699600,
      "step": 16730
    },
    {
      "epoch": 2.7300758626315362,
      "grad_norm": 0.022136177867650986,
      "learning_rate": 3.1342682455396996e-05,
      "loss": 0.0524,
      "num_input_tokens_seen": 31708448,
      "step": 16735
    },
    {
      "epoch": 2.7308915898523534,
      "grad_norm": 0.21196426451206207,
      "learning_rate": 3.133318570719441e-05,
      "loss": 0.1333,
      "num_input_tokens_seen": 31719312,
      "step": 16740
    },
    {
      "epoch": 2.7317073170731705,
      "grad_norm": 0.8973103165626526,
      "learning_rate": 3.132368798233499e-05,
      "loss": 0.1027,
      "num_input_tokens_seen": 31728976,
      "step": 16745
    },
    {
      "epoch": 2.732523044293988,
      "grad_norm": 0.19804881513118744,
      "learning_rate": 3.131418928228342e-05,
      "loss": 0.0618,
      "num_input_tokens_seen": 31737952,
      "step": 16750
    },
    {
      "epoch": 2.7333387715148056,
      "grad_norm": 0.07812768965959549,
      "learning_rate": 3.1304689608504514e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 31746912,
      "step": 16755
    },
    {
      "epoch": 2.7341544987356228,
      "grad_norm": 0.8568294048309326,
      "learning_rate": 3.129518896246324e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 31756208,
      "step": 16760
    },
    {
      "epoch": 2.73497022595644,
      "grad_norm": 0.8142222762107849,
      "learning_rate": 3.128568734562472e-05,
      "loss": 0.1916,
      "num_input_tokens_seen": 31767824,
      "step": 16765
    },
    {
      "epoch": 2.7357859531772575,
      "grad_norm": 0.29171523451805115,
      "learning_rate": 3.127618475945421e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 31778416,
      "step": 16770
    },
    {
      "epoch": 2.736601680398075,
      "grad_norm": 1.0506484508514404,
      "learning_rate": 3.126668120541715e-05,
      "loss": 0.0668,
      "num_input_tokens_seen": 31787952,
      "step": 16775
    },
    {
      "epoch": 2.737417407618892,
      "grad_norm": 1.5308951139450073,
      "learning_rate": 3.1257176684979096e-05,
      "loss": 0.2564,
      "num_input_tokens_seen": 31798480,
      "step": 16780
    },
    {
      "epoch": 2.7382331348397098,
      "grad_norm": 0.9204516410827637,
      "learning_rate": 3.124767119960576e-05,
      "loss": 0.22,
      "num_input_tokens_seen": 31806544,
      "step": 16785
    },
    {
      "epoch": 2.739048862060527,
      "grad_norm": 0.45823103189468384,
      "learning_rate": 3.123816475076301e-05,
      "loss": 0.066,
      "num_input_tokens_seen": 31816832,
      "step": 16790
    },
    {
      "epoch": 2.7398645892813445,
      "grad_norm": 0.7002341747283936,
      "learning_rate": 3.122865733991687e-05,
      "loss": 0.0427,
      "num_input_tokens_seen": 31825312,
      "step": 16795
    },
    {
      "epoch": 2.7406803165021616,
      "grad_norm": 0.9556106328964233,
      "learning_rate": 3.1219148968533486e-05,
      "loss": 0.0636,
      "num_input_tokens_seen": 31833136,
      "step": 16800
    },
    {
      "epoch": 2.7406803165021616,
      "eval_loss": 0.14530940353870392,
      "eval_runtime": 133.3801,
      "eval_samples_per_second": 20.43,
      "eval_steps_per_second": 10.219,
      "num_input_tokens_seen": 31833136,
      "step": 16800
    },
    {
      "epoch": 2.741496043722979,
      "grad_norm": 0.3801877200603485,
      "learning_rate": 3.120963963807918e-05,
      "loss": 0.09,
      "num_input_tokens_seen": 31842912,
      "step": 16805
    },
    {
      "epoch": 2.7423117709437963,
      "grad_norm": 1.094624638557434,
      "learning_rate": 3.12001293500204e-05,
      "loss": 0.1969,
      "num_input_tokens_seen": 31852624,
      "step": 16810
    },
    {
      "epoch": 2.743127498164614,
      "grad_norm": 1.005027174949646,
      "learning_rate": 3.1190618105823765e-05,
      "loss": 0.262,
      "num_input_tokens_seen": 31861440,
      "step": 16815
    },
    {
      "epoch": 2.743943225385431,
      "grad_norm": 0.05884275957942009,
      "learning_rate": 3.118110590695603e-05,
      "loss": 0.0808,
      "num_input_tokens_seen": 31870176,
      "step": 16820
    },
    {
      "epoch": 2.7447589526062486,
      "grad_norm": 0.1432800143957138,
      "learning_rate": 3.117159275488407e-05,
      "loss": 0.2306,
      "num_input_tokens_seen": 31879856,
      "step": 16825
    },
    {
      "epoch": 2.7455746798270657,
      "grad_norm": 0.6553974151611328,
      "learning_rate": 3.1162078651074956e-05,
      "loss": 0.2229,
      "num_input_tokens_seen": 31889120,
      "step": 16830
    },
    {
      "epoch": 2.7463904070478833,
      "grad_norm": 0.24978125095367432,
      "learning_rate": 3.1152563596995885e-05,
      "loss": 0.1455,
      "num_input_tokens_seen": 31898816,
      "step": 16835
    },
    {
      "epoch": 2.7472061342687004,
      "grad_norm": 0.6513045430183411,
      "learning_rate": 3.1143047594114186e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 31908512,
      "step": 16840
    },
    {
      "epoch": 2.748021861489518,
      "grad_norm": 0.7587801814079285,
      "learning_rate": 3.113353064389734e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 31917552,
      "step": 16845
    },
    {
      "epoch": 2.748837588710335,
      "grad_norm": 0.7937371134757996,
      "learning_rate": 3.1124012747812993e-05,
      "loss": 0.0578,
      "num_input_tokens_seen": 31927696,
      "step": 16850
    },
    {
      "epoch": 2.7496533159311527,
      "grad_norm": 0.07162167131900787,
      "learning_rate": 3.1114493907328936e-05,
      "loss": 0.0379,
      "num_input_tokens_seen": 31937424,
      "step": 16855
    },
    {
      "epoch": 2.75046904315197,
      "grad_norm": 0.5850275158882141,
      "learning_rate": 3.110497412391306e-05,
      "loss": 0.0884,
      "num_input_tokens_seen": 31946912,
      "step": 16860
    },
    {
      "epoch": 2.7512847703727874,
      "grad_norm": 0.8987415432929993,
      "learning_rate": 3.1095453399033466e-05,
      "loss": 0.094,
      "num_input_tokens_seen": 31954640,
      "step": 16865
    },
    {
      "epoch": 2.7521004975936045,
      "grad_norm": 0.24437817931175232,
      "learning_rate": 3.108593173415835e-05,
      "loss": 0.2685,
      "num_input_tokens_seen": 31965312,
      "step": 16870
    },
    {
      "epoch": 2.752916224814422,
      "grad_norm": 0.03573188930749893,
      "learning_rate": 3.107640913075609e-05,
      "loss": 0.2677,
      "num_input_tokens_seen": 31974816,
      "step": 16875
    },
    {
      "epoch": 2.7537319520352392,
      "grad_norm": 1.359655737876892,
      "learning_rate": 3.106688559029517e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 31983968,
      "step": 16880
    },
    {
      "epoch": 2.754547679256057,
      "grad_norm": 0.3071117401123047,
      "learning_rate": 3.105736111424425e-05,
      "loss": 0.1764,
      "num_input_tokens_seen": 31992576,
      "step": 16885
    },
    {
      "epoch": 2.7553634064768744,
      "grad_norm": 1.5046489238739014,
      "learning_rate": 3.1047835704072136e-05,
      "loss": 0.189,
      "num_input_tokens_seen": 32000944,
      "step": 16890
    },
    {
      "epoch": 2.7561791336976915,
      "grad_norm": 1.1584372520446777,
      "learning_rate": 3.103830936124775e-05,
      "loss": 0.2317,
      "num_input_tokens_seen": 32010384,
      "step": 16895
    },
    {
      "epoch": 2.7569948609185086,
      "grad_norm": 1.0859489440917969,
      "learning_rate": 3.102878208724018e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 32021040,
      "step": 16900
    },
    {
      "epoch": 2.7578105881393262,
      "grad_norm": 1.431986689567566,
      "learning_rate": 3.101925388351865e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 32030624,
      "step": 16905
    },
    {
      "epoch": 2.758626315360144,
      "grad_norm": 0.41476359963417053,
      "learning_rate": 3.1009724751552515e-05,
      "loss": 0.0714,
      "num_input_tokens_seen": 32039840,
      "step": 16910
    },
    {
      "epoch": 2.759442042580961,
      "grad_norm": 0.39936816692352295,
      "learning_rate": 3.100019469281131e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 32050064,
      "step": 16915
    },
    {
      "epoch": 2.760257769801778,
      "grad_norm": 0.12036274373531342,
      "learning_rate": 3.0990663708764685e-05,
      "loss": 0.1646,
      "num_input_tokens_seen": 32060208,
      "step": 16920
    },
    {
      "epoch": 2.7610734970225956,
      "grad_norm": 0.31895267963409424,
      "learning_rate": 3.098113180088243e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 32069984,
      "step": 16925
    },
    {
      "epoch": 2.761889224243413,
      "grad_norm": 0.4457817077636719,
      "learning_rate": 3.097159897063448e-05,
      "loss": 0.0996,
      "num_input_tokens_seen": 32079840,
      "step": 16930
    },
    {
      "epoch": 2.7627049514642303,
      "grad_norm": 0.4198065996170044,
      "learning_rate": 3.096206521949094e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 32088656,
      "step": 16935
    },
    {
      "epoch": 2.7635206786850475,
      "grad_norm": 0.45653975009918213,
      "learning_rate": 3.0952530548922006e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 32098176,
      "step": 16940
    },
    {
      "epoch": 2.764336405905865,
      "grad_norm": 0.9463157653808594,
      "learning_rate": 3.0942994960398064e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 32107952,
      "step": 16945
    },
    {
      "epoch": 2.7651521331266826,
      "grad_norm": 1.5749748945236206,
      "learning_rate": 3.093345845538961e-05,
      "loss": 0.196,
      "num_input_tokens_seen": 32117184,
      "step": 16950
    },
    {
      "epoch": 2.7659678603474998,
      "grad_norm": 1.0187697410583496,
      "learning_rate": 3.09239210353673e-05,
      "loss": 0.0908,
      "num_input_tokens_seen": 32126416,
      "step": 16955
    },
    {
      "epoch": 2.766783587568317,
      "grad_norm": 0.7007352113723755,
      "learning_rate": 3.0914382701801926e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 32136080,
      "step": 16960
    },
    {
      "epoch": 2.7675993147891345,
      "grad_norm": 1.5067836046218872,
      "learning_rate": 3.090484345616441e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 32145712,
      "step": 16965
    },
    {
      "epoch": 2.768415042009952,
      "grad_norm": 0.089383065700531,
      "learning_rate": 3.0895303299925825e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 32154528,
      "step": 16970
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.9974232912063599,
      "learning_rate": 3.0885762234557393e-05,
      "loss": 0.267,
      "num_input_tokens_seen": 32164672,
      "step": 16975
    },
    {
      "epoch": 2.7700464964515867,
      "grad_norm": 2.068284749984741,
      "learning_rate": 3.087622026153045e-05,
      "loss": 0.0837,
      "num_input_tokens_seen": 32174544,
      "step": 16980
    },
    {
      "epoch": 2.770862223672404,
      "grad_norm": 0.6299343109130859,
      "learning_rate": 3.086667738231651e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 32183248,
      "step": 16985
    },
    {
      "epoch": 2.7716779508932214,
      "grad_norm": 1.3442399501800537,
      "learning_rate": 3.085713359838718e-05,
      "loss": 0.1927,
      "num_input_tokens_seen": 32192768,
      "step": 16990
    },
    {
      "epoch": 2.7724936781140386,
      "grad_norm": 0.5071068406105042,
      "learning_rate": 3.084758891121425e-05,
      "loss": 0.0972,
      "num_input_tokens_seen": 32202880,
      "step": 16995
    },
    {
      "epoch": 2.773309405334856,
      "grad_norm": 0.13305234909057617,
      "learning_rate": 3.083804332226963e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 32213296,
      "step": 17000
    },
    {
      "epoch": 2.773309405334856,
      "eval_loss": 0.1447463482618332,
      "eval_runtime": 133.2536,
      "eval_samples_per_second": 20.45,
      "eval_steps_per_second": 10.229,
      "num_input_tokens_seen": 32213296,
      "step": 17000
    },
    {
      "epoch": 2.7741251325556733,
      "grad_norm": 0.9317333102226257,
      "learning_rate": 3.082849683302536e-05,
      "loss": 0.2585,
      "num_input_tokens_seen": 32223056,
      "step": 17005
    },
    {
      "epoch": 2.774940859776491,
      "grad_norm": 0.2720954418182373,
      "learning_rate": 3.081894944495363e-05,
      "loss": 0.0938,
      "num_input_tokens_seen": 32232336,
      "step": 17010
    },
    {
      "epoch": 2.775756586997308,
      "grad_norm": 0.31607580184936523,
      "learning_rate": 3.080940115952677e-05,
      "loss": 0.0774,
      "num_input_tokens_seen": 32240512,
      "step": 17015
    },
    {
      "epoch": 2.7765723142181256,
      "grad_norm": 0.3942442834377289,
      "learning_rate": 3.0799851978217245e-05,
      "loss": 0.0515,
      "num_input_tokens_seen": 32249728,
      "step": 17020
    },
    {
      "epoch": 2.7773880414389427,
      "grad_norm": 0.20427539944648743,
      "learning_rate": 3.0790301902497666e-05,
      "loss": 0.1056,
      "num_input_tokens_seen": 32258944,
      "step": 17025
    },
    {
      "epoch": 2.7782037686597603,
      "grad_norm": 0.7915567755699158,
      "learning_rate": 3.078075093384076e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 32268896,
      "step": 17030
    },
    {
      "epoch": 2.7790194958805774,
      "grad_norm": 1.0965520143508911,
      "learning_rate": 3.077119907371942e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 32279568,
      "step": 17035
    },
    {
      "epoch": 2.779835223101395,
      "grad_norm": 0.32262060046195984,
      "learning_rate": 3.076164632360666e-05,
      "loss": 0.2395,
      "num_input_tokens_seen": 32288784,
      "step": 17040
    },
    {
      "epoch": 2.780650950322212,
      "grad_norm": 0.08621180057525635,
      "learning_rate": 3.075209268497563e-05,
      "loss": 0.0998,
      "num_input_tokens_seen": 32298784,
      "step": 17045
    },
    {
      "epoch": 2.7814666775430297,
      "grad_norm": 0.4190996289253235,
      "learning_rate": 3.074253815929961e-05,
      "loss": 0.0912,
      "num_input_tokens_seen": 32308480,
      "step": 17050
    },
    {
      "epoch": 2.782282404763847,
      "grad_norm": 1.2657065391540527,
      "learning_rate": 3.0732982748052054e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 32318960,
      "step": 17055
    },
    {
      "epoch": 2.7830981319846644,
      "grad_norm": 1.8398364782333374,
      "learning_rate": 3.072342645270651e-05,
      "loss": 0.2964,
      "num_input_tokens_seen": 32329232,
      "step": 17060
    },
    {
      "epoch": 2.783913859205482,
      "grad_norm": 1.0700432062149048,
      "learning_rate": 3.071386927473668e-05,
      "loss": 0.1907,
      "num_input_tokens_seen": 32337184,
      "step": 17065
    },
    {
      "epoch": 2.784729586426299,
      "grad_norm": 1.2068489789962769,
      "learning_rate": 3.0704311215616404e-05,
      "loss": 0.2337,
      "num_input_tokens_seen": 32346544,
      "step": 17070
    },
    {
      "epoch": 2.785545313647116,
      "grad_norm": 0.17700791358947754,
      "learning_rate": 3.0694752276819656e-05,
      "loss": 0.1238,
      "num_input_tokens_seen": 32354896,
      "step": 17075
    },
    {
      "epoch": 2.786361040867934,
      "grad_norm": 0.8170201778411865,
      "learning_rate": 3.068519245982054e-05,
      "loss": 0.1316,
      "num_input_tokens_seen": 32364224,
      "step": 17080
    },
    {
      "epoch": 2.7871767680887514,
      "grad_norm": 1.3726375102996826,
      "learning_rate": 3.0675631766093304e-05,
      "loss": 0.3541,
      "num_input_tokens_seen": 32372688,
      "step": 17085
    },
    {
      "epoch": 2.7879924953095685,
      "grad_norm": 0.2647177278995514,
      "learning_rate": 3.066607019711232e-05,
      "loss": 0.092,
      "num_input_tokens_seen": 32382512,
      "step": 17090
    },
    {
      "epoch": 2.7888082225303856,
      "grad_norm": 0.053780447691679,
      "learning_rate": 3.065650775435211e-05,
      "loss": 0.1895,
      "num_input_tokens_seen": 32392032,
      "step": 17095
    },
    {
      "epoch": 2.789623949751203,
      "grad_norm": 0.05338728055357933,
      "learning_rate": 3.0646944439287326e-05,
      "loss": 0.0725,
      "num_input_tokens_seen": 32400432,
      "step": 17100
    },
    {
      "epoch": 2.7904396769720208,
      "grad_norm": 1.380506157875061,
      "learning_rate": 3.0637380253392736e-05,
      "loss": 0.1002,
      "num_input_tokens_seen": 32409520,
      "step": 17105
    },
    {
      "epoch": 2.791255404192838,
      "grad_norm": 0.44946980476379395,
      "learning_rate": 3.062781519814327e-05,
      "loss": 0.1088,
      "num_input_tokens_seen": 32418080,
      "step": 17110
    },
    {
      "epoch": 2.792071131413655,
      "grad_norm": 0.5671525001525879,
      "learning_rate": 3.0618249275013985e-05,
      "loss": 0.3747,
      "num_input_tokens_seen": 32427712,
      "step": 17115
    },
    {
      "epoch": 2.7928868586344726,
      "grad_norm": 0.7107559442520142,
      "learning_rate": 3.060868248548005e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 32437504,
      "step": 17120
    },
    {
      "epoch": 2.79370258585529,
      "grad_norm": 0.8837308287620544,
      "learning_rate": 3.0599114831016796e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 32447456,
      "step": 17125
    },
    {
      "epoch": 2.7945183130761073,
      "grad_norm": 0.22787420451641083,
      "learning_rate": 3.0589546313099666e-05,
      "loss": 0.0532,
      "num_input_tokens_seen": 32455168,
      "step": 17130
    },
    {
      "epoch": 2.7953340402969244,
      "grad_norm": 0.17967917025089264,
      "learning_rate": 3.0579976933204255e-05,
      "loss": 0.0855,
      "num_input_tokens_seen": 32464240,
      "step": 17135
    },
    {
      "epoch": 2.796149767517742,
      "grad_norm": 1.3505991697311401,
      "learning_rate": 3.0570406692806284e-05,
      "loss": 0.1287,
      "num_input_tokens_seen": 32474576,
      "step": 17140
    },
    {
      "epoch": 2.7969654947385596,
      "grad_norm": 1.2062429189682007,
      "learning_rate": 3.05608355933816e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 32483984,
      "step": 17145
    },
    {
      "epoch": 2.7977812219593767,
      "grad_norm": 0.14646093547344208,
      "learning_rate": 3.055126363640618e-05,
      "loss": 0.0522,
      "num_input_tokens_seen": 32493056,
      "step": 17150
    },
    {
      "epoch": 2.7985969491801943,
      "grad_norm": 0.5240901112556458,
      "learning_rate": 3.0541690823356146e-05,
      "loss": 0.1094,
      "num_input_tokens_seen": 32502960,
      "step": 17155
    },
    {
      "epoch": 2.7994126764010114,
      "grad_norm": 0.37345126271247864,
      "learning_rate": 3.053211715570775e-05,
      "loss": 0.1015,
      "num_input_tokens_seen": 32512832,
      "step": 17160
    },
    {
      "epoch": 2.800228403621829,
      "grad_norm": 2.432927370071411,
      "learning_rate": 3.052254263493736e-05,
      "loss": 0.2839,
      "num_input_tokens_seen": 32522224,
      "step": 17165
    },
    {
      "epoch": 2.801044130842646,
      "grad_norm": 1.2683039903640747,
      "learning_rate": 3.0512967262521498e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 32530176,
      "step": 17170
    },
    {
      "epoch": 2.8018598580634637,
      "grad_norm": 0.8442355394363403,
      "learning_rate": 3.0503391039936803e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 32538976,
      "step": 17175
    },
    {
      "epoch": 2.802675585284281,
      "grad_norm": 0.36124423146247864,
      "learning_rate": 3.0493813968660056e-05,
      "loss": 0.1519,
      "num_input_tokens_seen": 32548944,
      "step": 17180
    },
    {
      "epoch": 2.8034913125050984,
      "grad_norm": 0.5644629597663879,
      "learning_rate": 3.0484236050168153e-05,
      "loss": 0.0702,
      "num_input_tokens_seen": 32558976,
      "step": 17185
    },
    {
      "epoch": 2.8043070397259156,
      "grad_norm": 0.21678990125656128,
      "learning_rate": 3.0474657285938123e-05,
      "loss": 0.2062,
      "num_input_tokens_seen": 32569200,
      "step": 17190
    },
    {
      "epoch": 2.805122766946733,
      "grad_norm": 1.8461406230926514,
      "learning_rate": 3.046507767744715e-05,
      "loss": 0.3354,
      "num_input_tokens_seen": 32578688,
      "step": 17195
    },
    {
      "epoch": 2.8059384941675503,
      "grad_norm": 0.7551963925361633,
      "learning_rate": 3.045549722617252e-05,
      "loss": 0.1023,
      "num_input_tokens_seen": 32588128,
      "step": 17200
    },
    {
      "epoch": 2.8059384941675503,
      "eval_loss": 0.147405207157135,
      "eval_runtime": 133.3032,
      "eval_samples_per_second": 20.442,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 32588128,
      "step": 17200
    },
    {
      "epoch": 2.806754221388368,
      "grad_norm": 0.4325866997241974,
      "learning_rate": 3.0445915933591658e-05,
      "loss": 0.1951,
      "num_input_tokens_seen": 32598000,
      "step": 17205
    },
    {
      "epoch": 2.807569948609185,
      "grad_norm": 0.23293918371200562,
      "learning_rate": 3.0436333801182114e-05,
      "loss": 0.0778,
      "num_input_tokens_seen": 32607984,
      "step": 17210
    },
    {
      "epoch": 2.8083856758300025,
      "grad_norm": 0.40934014320373535,
      "learning_rate": 3.0426750830421596e-05,
      "loss": 0.0837,
      "num_input_tokens_seen": 32617104,
      "step": 17215
    },
    {
      "epoch": 2.8092014030508197,
      "grad_norm": 0.46679651737213135,
      "learning_rate": 3.0417167022787897e-05,
      "loss": 0.0661,
      "num_input_tokens_seen": 32625456,
      "step": 17220
    },
    {
      "epoch": 2.8100171302716372,
      "grad_norm": 0.052508123219013214,
      "learning_rate": 3.0407582379758966e-05,
      "loss": 0.1005,
      "num_input_tokens_seen": 32635280,
      "step": 17225
    },
    {
      "epoch": 2.8108328574924544,
      "grad_norm": 0.7513095140457153,
      "learning_rate": 3.039799690281287e-05,
      "loss": 0.0844,
      "num_input_tokens_seen": 32643136,
      "step": 17230
    },
    {
      "epoch": 2.811648584713272,
      "grad_norm": 0.15242432057857513,
      "learning_rate": 3.0388410593427823e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 32651600,
      "step": 17235
    },
    {
      "epoch": 2.812464311934089,
      "grad_norm": 0.18510428071022034,
      "learning_rate": 3.0378823453082146e-05,
      "loss": 0.1882,
      "num_input_tokens_seen": 32661264,
      "step": 17240
    },
    {
      "epoch": 2.8132800391549067,
      "grad_norm": 1.5284806489944458,
      "learning_rate": 3.03692354832543e-05,
      "loss": 0.1961,
      "num_input_tokens_seen": 32672032,
      "step": 17245
    },
    {
      "epoch": 2.814095766375724,
      "grad_norm": 0.2254219800233841,
      "learning_rate": 3.0359646685422865e-05,
      "loss": 0.2268,
      "num_input_tokens_seen": 32683024,
      "step": 17250
    },
    {
      "epoch": 2.8149114935965414,
      "grad_norm": 1.3155782222747803,
      "learning_rate": 3.035005706106656e-05,
      "loss": 0.2705,
      "num_input_tokens_seen": 32692512,
      "step": 17255
    },
    {
      "epoch": 2.815727220817359,
      "grad_norm": 0.2639642655849457,
      "learning_rate": 3.034046661166422e-05,
      "loss": 0.2057,
      "num_input_tokens_seen": 32702384,
      "step": 17260
    },
    {
      "epoch": 2.816542948038176,
      "grad_norm": 0.4845694899559021,
      "learning_rate": 3.033087533869482e-05,
      "loss": 0.1829,
      "num_input_tokens_seen": 32712352,
      "step": 17265
    },
    {
      "epoch": 2.817358675258993,
      "grad_norm": 0.11194822937250137,
      "learning_rate": 3.0321283243637444e-05,
      "loss": 0.1097,
      "num_input_tokens_seen": 32722448,
      "step": 17270
    },
    {
      "epoch": 2.8181744024798108,
      "grad_norm": 0.0716741532087326,
      "learning_rate": 3.0311690327971326e-05,
      "loss": 0.1138,
      "num_input_tokens_seen": 32732016,
      "step": 17275
    },
    {
      "epoch": 2.8189901297006283,
      "grad_norm": 2.1121792793273926,
      "learning_rate": 3.030209659317581e-05,
      "loss": 0.1896,
      "num_input_tokens_seen": 32741904,
      "step": 17280
    },
    {
      "epoch": 2.8198058569214455,
      "grad_norm": 0.5649389028549194,
      "learning_rate": 3.0292502040730362e-05,
      "loss": 0.1122,
      "num_input_tokens_seen": 32751904,
      "step": 17285
    },
    {
      "epoch": 2.8206215841422626,
      "grad_norm": 0.1667235642671585,
      "learning_rate": 3.0282906672114597e-05,
      "loss": 0.0891,
      "num_input_tokens_seen": 32762096,
      "step": 17290
    },
    {
      "epoch": 2.82143731136308,
      "grad_norm": 0.09161605685949326,
      "learning_rate": 3.027331048880823e-05,
      "loss": 0.1272,
      "num_input_tokens_seen": 32772272,
      "step": 17295
    },
    {
      "epoch": 2.8222530385838978,
      "grad_norm": 0.4403226673603058,
      "learning_rate": 3.0263713492291123e-05,
      "loss": 0.0636,
      "num_input_tokens_seen": 32781664,
      "step": 17300
    },
    {
      "epoch": 2.823068765804715,
      "grad_norm": 1.8300957679748535,
      "learning_rate": 3.0254115684043242e-05,
      "loss": 0.1694,
      "num_input_tokens_seen": 32792128,
      "step": 17305
    },
    {
      "epoch": 2.823884493025532,
      "grad_norm": 0.9998231530189514,
      "learning_rate": 3.024451706554469e-05,
      "loss": 0.2095,
      "num_input_tokens_seen": 32801248,
      "step": 17310
    },
    {
      "epoch": 2.8247002202463496,
      "grad_norm": 3.278477907180786,
      "learning_rate": 3.0234917638275705e-05,
      "loss": 0.2387,
      "num_input_tokens_seen": 32810304,
      "step": 17315
    },
    {
      "epoch": 2.825515947467167,
      "grad_norm": 0.394405335187912,
      "learning_rate": 3.0225317403716635e-05,
      "loss": 0.1164,
      "num_input_tokens_seen": 32820496,
      "step": 17320
    },
    {
      "epoch": 2.8263316746879843,
      "grad_norm": 0.35757189989089966,
      "learning_rate": 3.0215716363347956e-05,
      "loss": 0.09,
      "num_input_tokens_seen": 32829968,
      "step": 17325
    },
    {
      "epoch": 2.8271474019088014,
      "grad_norm": 0.22203080356121063,
      "learning_rate": 3.0206114518650275e-05,
      "loss": 0.0922,
      "num_input_tokens_seen": 32840432,
      "step": 17330
    },
    {
      "epoch": 2.827963129129619,
      "grad_norm": 0.821916401386261,
      "learning_rate": 3.0196511871104304e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 32850864,
      "step": 17335
    },
    {
      "epoch": 2.8287788563504366,
      "grad_norm": 0.06372971832752228,
      "learning_rate": 3.01869084221909e-05,
      "loss": 0.0661,
      "num_input_tokens_seen": 32859664,
      "step": 17340
    },
    {
      "epoch": 2.8295945835712537,
      "grad_norm": 0.5138905048370361,
      "learning_rate": 3.0177304173391037e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 32869008,
      "step": 17345
    },
    {
      "epoch": 2.8304103107920713,
      "grad_norm": 0.14793001115322113,
      "learning_rate": 3.01676991261858e-05,
      "loss": 0.2454,
      "num_input_tokens_seen": 32878832,
      "step": 17350
    },
    {
      "epoch": 2.8312260380128884,
      "grad_norm": 1.8305823802947998,
      "learning_rate": 3.015809328205642e-05,
      "loss": 0.2542,
      "num_input_tokens_seen": 32887280,
      "step": 17355
    },
    {
      "epoch": 2.832041765233706,
      "grad_norm": 1.4394896030426025,
      "learning_rate": 3.0148486642484248e-05,
      "loss": 0.2722,
      "num_input_tokens_seen": 32896208,
      "step": 17360
    },
    {
      "epoch": 2.832857492454523,
      "grad_norm": 0.5263702869415283,
      "learning_rate": 3.0138879208950722e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 32905072,
      "step": 17365
    },
    {
      "epoch": 2.8336732196753407,
      "grad_norm": 0.4467626214027405,
      "learning_rate": 3.012927098293744e-05,
      "loss": 0.1641,
      "num_input_tokens_seen": 32914912,
      "step": 17370
    },
    {
      "epoch": 2.834488946896158,
      "grad_norm": 1.278093934059143,
      "learning_rate": 3.0119661965926123e-05,
      "loss": 0.2363,
      "num_input_tokens_seen": 32923136,
      "step": 17375
    },
    {
      "epoch": 2.8353046741169754,
      "grad_norm": 1.3314152956008911,
      "learning_rate": 3.0110052159398587e-05,
      "loss": 0.2068,
      "num_input_tokens_seen": 32932304,
      "step": 17380
    },
    {
      "epoch": 2.8361204013377925,
      "grad_norm": 0.5220938324928284,
      "learning_rate": 3.0100441564836802e-05,
      "loss": 0.2331,
      "num_input_tokens_seen": 32942016,
      "step": 17385
    },
    {
      "epoch": 2.83693612855861,
      "grad_norm": 0.20661865174770355,
      "learning_rate": 3.0090830183722817e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 32951872,
      "step": 17390
    },
    {
      "epoch": 2.8377518557794272,
      "grad_norm": 0.8054172992706299,
      "learning_rate": 3.0081218017538852e-05,
      "loss": 0.0872,
      "num_input_tokens_seen": 32962032,
      "step": 17395
    },
    {
      "epoch": 2.838567583000245,
      "grad_norm": 0.10725194960832596,
      "learning_rate": 3.0071605067767212e-05,
      "loss": 0.2427,
      "num_input_tokens_seen": 32971552,
      "step": 17400
    },
    {
      "epoch": 2.838567583000245,
      "eval_loss": 0.14417928457260132,
      "eval_runtime": 133.3901,
      "eval_samples_per_second": 20.429,
      "eval_steps_per_second": 10.218,
      "num_input_tokens_seen": 32971552,
      "step": 17400
    },
    {
      "epoch": 2.839383310221062,
      "grad_norm": 0.3716790974140167,
      "learning_rate": 3.006199133589034e-05,
      "loss": 0.1205,
      "num_input_tokens_seen": 32981872,
      "step": 17405
    },
    {
      "epoch": 2.8401990374418795,
      "grad_norm": 0.1656465381383896,
      "learning_rate": 3.005237682339079e-05,
      "loss": 0.2439,
      "num_input_tokens_seen": 32991136,
      "step": 17410
    },
    {
      "epoch": 2.8410147646626966,
      "grad_norm": 0.1840868890285492,
      "learning_rate": 3.0042761531751228e-05,
      "loss": 0.1744,
      "num_input_tokens_seen": 33000384,
      "step": 17415
    },
    {
      "epoch": 2.841830491883514,
      "grad_norm": 0.5655860304832458,
      "learning_rate": 3.0033145462454482e-05,
      "loss": 0.1502,
      "num_input_tokens_seen": 33008912,
      "step": 17420
    },
    {
      "epoch": 2.8426462191043314,
      "grad_norm": 1.610358715057373,
      "learning_rate": 3.002352861698345e-05,
      "loss": 0.2144,
      "num_input_tokens_seen": 33018608,
      "step": 17425
    },
    {
      "epoch": 2.843461946325149,
      "grad_norm": 0.18179009854793549,
      "learning_rate": 3.0013910996821178e-05,
      "loss": 0.0623,
      "num_input_tokens_seen": 33028464,
      "step": 17430
    },
    {
      "epoch": 2.8442776735459665,
      "grad_norm": 0.4127155840396881,
      "learning_rate": 3.0004292603450817e-05,
      "loss": 0.1008,
      "num_input_tokens_seen": 33038752,
      "step": 17435
    },
    {
      "epoch": 2.8450934007667836,
      "grad_norm": 0.19951385259628296,
      "learning_rate": 2.9994673438355653e-05,
      "loss": 0.0971,
      "num_input_tokens_seen": 33048400,
      "step": 17440
    },
    {
      "epoch": 2.8459091279876008,
      "grad_norm": 0.10394182801246643,
      "learning_rate": 2.9985053503019078e-05,
      "loss": 0.0588,
      "num_input_tokens_seen": 33057840,
      "step": 17445
    },
    {
      "epoch": 2.8467248552084183,
      "grad_norm": 0.7434885501861572,
      "learning_rate": 2.99754327989246e-05,
      "loss": 0.1019,
      "num_input_tokens_seen": 33067568,
      "step": 17450
    },
    {
      "epoch": 2.847540582429236,
      "grad_norm": 0.17933261394500732,
      "learning_rate": 2.9965811327555864e-05,
      "loss": 0.0657,
      "num_input_tokens_seen": 33077248,
      "step": 17455
    },
    {
      "epoch": 2.848356309650053,
      "grad_norm": 0.10144105553627014,
      "learning_rate": 2.995618909039662e-05,
      "loss": 0.0263,
      "num_input_tokens_seen": 33087168,
      "step": 17460
    },
    {
      "epoch": 2.84917203687087,
      "grad_norm": 1.5694772005081177,
      "learning_rate": 2.9946566088930727e-05,
      "loss": 0.2103,
      "num_input_tokens_seen": 33096704,
      "step": 17465
    },
    {
      "epoch": 2.8499877640916877,
      "grad_norm": 0.30092284083366394,
      "learning_rate": 2.9936942324642192e-05,
      "loss": 0.1095,
      "num_input_tokens_seen": 33105456,
      "step": 17470
    },
    {
      "epoch": 2.8508034913125053,
      "grad_norm": 0.23258830606937408,
      "learning_rate": 2.9927317799015097e-05,
      "loss": 0.0794,
      "num_input_tokens_seen": 33114752,
      "step": 17475
    },
    {
      "epoch": 2.8516192185333225,
      "grad_norm": 1.5072518587112427,
      "learning_rate": 2.9917692513533685e-05,
      "loss": 0.2059,
      "num_input_tokens_seen": 33124144,
      "step": 17480
    },
    {
      "epoch": 2.8524349457541396,
      "grad_norm": 0.11356618255376816,
      "learning_rate": 2.990806646968229e-05,
      "loss": 0.087,
      "num_input_tokens_seen": 33134336,
      "step": 17485
    },
    {
      "epoch": 2.853250672974957,
      "grad_norm": 0.18288184702396393,
      "learning_rate": 2.989843966894536e-05,
      "loss": 0.2656,
      "num_input_tokens_seen": 33143872,
      "step": 17490
    },
    {
      "epoch": 2.8540664001957747,
      "grad_norm": 0.6094569563865662,
      "learning_rate": 2.9888812112807472e-05,
      "loss": 0.1682,
      "num_input_tokens_seen": 33152784,
      "step": 17495
    },
    {
      "epoch": 2.854882127416592,
      "grad_norm": 0.20705445110797882,
      "learning_rate": 2.987918380275333e-05,
      "loss": 0.0976,
      "num_input_tokens_seen": 33162432,
      "step": 17500
    },
    {
      "epoch": 2.855697854637409,
      "grad_norm": 0.1583642214536667,
      "learning_rate": 2.9869554740267724e-05,
      "loss": 0.1199,
      "num_input_tokens_seen": 33171984,
      "step": 17505
    },
    {
      "epoch": 2.8565135818582266,
      "grad_norm": 0.201979398727417,
      "learning_rate": 2.9859924926835585e-05,
      "loss": 0.1978,
      "num_input_tokens_seen": 33180736,
      "step": 17510
    },
    {
      "epoch": 2.857329309079044,
      "grad_norm": 1.2873812913894653,
      "learning_rate": 2.9850294363941944e-05,
      "loss": 0.2018,
      "num_input_tokens_seen": 33189552,
      "step": 17515
    },
    {
      "epoch": 2.8581450362998613,
      "grad_norm": 0.5605859756469727,
      "learning_rate": 2.9840663053071967e-05,
      "loss": 0.1247,
      "num_input_tokens_seen": 33200192,
      "step": 17520
    },
    {
      "epoch": 2.858960763520679,
      "grad_norm": 0.35610857605934143,
      "learning_rate": 2.983103099571091e-05,
      "loss": 0.0889,
      "num_input_tokens_seen": 33210304,
      "step": 17525
    },
    {
      "epoch": 2.859776490741496,
      "grad_norm": 0.7591573596000671,
      "learning_rate": 2.9821398193344164e-05,
      "loss": 0.0906,
      "num_input_tokens_seen": 33220400,
      "step": 17530
    },
    {
      "epoch": 2.8605922179623136,
      "grad_norm": 0.5587775707244873,
      "learning_rate": 2.9811764647457226e-05,
      "loss": 0.1011,
      "num_input_tokens_seen": 33229568,
      "step": 17535
    },
    {
      "epoch": 2.8614079451831307,
      "grad_norm": 1.5268423557281494,
      "learning_rate": 2.9802130359535714e-05,
      "loss": 0.0916,
      "num_input_tokens_seen": 33240080,
      "step": 17540
    },
    {
      "epoch": 2.8622236724039483,
      "grad_norm": 0.6919792294502258,
      "learning_rate": 2.979249533106535e-05,
      "loss": 0.1917,
      "num_input_tokens_seen": 33248864,
      "step": 17545
    },
    {
      "epoch": 2.8630393996247654,
      "grad_norm": 0.629262387752533,
      "learning_rate": 2.9782859563531986e-05,
      "loss": 0.2601,
      "num_input_tokens_seen": 33258512,
      "step": 17550
    },
    {
      "epoch": 2.863855126845583,
      "grad_norm": 0.14484240114688873,
      "learning_rate": 2.977322305842156e-05,
      "loss": 0.082,
      "num_input_tokens_seen": 33268240,
      "step": 17555
    },
    {
      "epoch": 2.8646708540664,
      "grad_norm": 0.6713433265686035,
      "learning_rate": 2.9763585817220162e-05,
      "loss": 0.0804,
      "num_input_tokens_seen": 33277280,
      "step": 17560
    },
    {
      "epoch": 2.8654865812872177,
      "grad_norm": 0.5305413007736206,
      "learning_rate": 2.975394784141397e-05,
      "loss": 0.1864,
      "num_input_tokens_seen": 33286880,
      "step": 17565
    },
    {
      "epoch": 2.866302308508035,
      "grad_norm": 0.5080809593200684,
      "learning_rate": 2.974430913248928e-05,
      "loss": 0.2125,
      "num_input_tokens_seen": 33297136,
      "step": 17570
    },
    {
      "epoch": 2.8671180357288524,
      "grad_norm": 0.45030900835990906,
      "learning_rate": 2.9734669691932497e-05,
      "loss": 0.1825,
      "num_input_tokens_seen": 33306880,
      "step": 17575
    },
    {
      "epoch": 2.8679337629496695,
      "grad_norm": 0.026702256873250008,
      "learning_rate": 2.9725029521230147e-05,
      "loss": 0.0661,
      "num_input_tokens_seen": 33316752,
      "step": 17580
    },
    {
      "epoch": 2.868749490170487,
      "grad_norm": 0.8783713579177856,
      "learning_rate": 2.9715388621868873e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 33326240,
      "step": 17585
    },
    {
      "epoch": 2.869565217391304,
      "grad_norm": 0.1704959124326706,
      "learning_rate": 2.970574699533541e-05,
      "loss": 0.1027,
      "num_input_tokens_seen": 33336320,
      "step": 17590
    },
    {
      "epoch": 2.870380944612122,
      "grad_norm": 0.19212891161441803,
      "learning_rate": 2.969610464311662e-05,
      "loss": 0.0692,
      "num_input_tokens_seen": 33345584,
      "step": 17595
    },
    {
      "epoch": 2.871196671832939,
      "grad_norm": 0.5409834384918213,
      "learning_rate": 2.9686461566699487e-05,
      "loss": 0.0961,
      "num_input_tokens_seen": 33356064,
      "step": 17600
    },
    {
      "epoch": 2.871196671832939,
      "eval_loss": 0.14295439422130585,
      "eval_runtime": 133.2819,
      "eval_samples_per_second": 20.445,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 33356064,
      "step": 17600
    },
    {
      "epoch": 2.8720123990537565,
      "grad_norm": 1.2926971912384033,
      "learning_rate": 2.9676817767571086e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 33366688,
      "step": 17605
    },
    {
      "epoch": 2.8728281262745736,
      "grad_norm": 0.08972817659378052,
      "learning_rate": 2.966717324721861e-05,
      "loss": 0.2562,
      "num_input_tokens_seen": 33376464,
      "step": 17610
    },
    {
      "epoch": 2.873643853495391,
      "grad_norm": 0.3331056237220764,
      "learning_rate": 2.9657528007129366e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 33385872,
      "step": 17615
    },
    {
      "epoch": 2.8744595807162083,
      "grad_norm": 0.14518801867961884,
      "learning_rate": 2.9647882048790777e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 33396432,
      "step": 17620
    },
    {
      "epoch": 2.875275307937026,
      "grad_norm": 1.3820608854293823,
      "learning_rate": 2.963823537369037e-05,
      "loss": 0.1098,
      "num_input_tokens_seen": 33406208,
      "step": 17625
    },
    {
      "epoch": 2.8760910351578435,
      "grad_norm": 0.8576138615608215,
      "learning_rate": 2.9628587983315775e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 33414944,
      "step": 17630
    },
    {
      "epoch": 2.8769067623786606,
      "grad_norm": 0.7240233421325684,
      "learning_rate": 2.9618939879154746e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 33424816,
      "step": 17635
    },
    {
      "epoch": 2.8777224895994777,
      "grad_norm": 0.0316508486866951,
      "learning_rate": 2.9609291062695143e-05,
      "loss": 0.2309,
      "num_input_tokens_seen": 33433520,
      "step": 17640
    },
    {
      "epoch": 2.8785382168202953,
      "grad_norm": 0.9196802973747253,
      "learning_rate": 2.9599641535424938e-05,
      "loss": 0.0751,
      "num_input_tokens_seen": 33442576,
      "step": 17645
    },
    {
      "epoch": 2.879353944041113,
      "grad_norm": 0.2385273575782776,
      "learning_rate": 2.9589991298832202e-05,
      "loss": 0.0842,
      "num_input_tokens_seen": 33450432,
      "step": 17650
    },
    {
      "epoch": 2.88016967126193,
      "grad_norm": 0.6971878409385681,
      "learning_rate": 2.958034035440513e-05,
      "loss": 0.0867,
      "num_input_tokens_seen": 33459920,
      "step": 17655
    },
    {
      "epoch": 2.880985398482747,
      "grad_norm": 0.8246817588806152,
      "learning_rate": 2.957068870363201e-05,
      "loss": 0.1202,
      "num_input_tokens_seen": 33468784,
      "step": 17660
    },
    {
      "epoch": 2.8818011257035647,
      "grad_norm": 2.973921298980713,
      "learning_rate": 2.956103634800126e-05,
      "loss": 0.3605,
      "num_input_tokens_seen": 33478080,
      "step": 17665
    },
    {
      "epoch": 2.8826168529243823,
      "grad_norm": 1.2169365882873535,
      "learning_rate": 2.9551383289001384e-05,
      "loss": 0.2482,
      "num_input_tokens_seen": 33488240,
      "step": 17670
    },
    {
      "epoch": 2.8834325801451994,
      "grad_norm": 1.0347795486450195,
      "learning_rate": 2.9541729528121005e-05,
      "loss": 0.1969,
      "num_input_tokens_seen": 33497952,
      "step": 17675
    },
    {
      "epoch": 2.8842483073660166,
      "grad_norm": 0.8655484318733215,
      "learning_rate": 2.9532075066848856e-05,
      "loss": 0.0974,
      "num_input_tokens_seen": 33508048,
      "step": 17680
    },
    {
      "epoch": 2.885064034586834,
      "grad_norm": 0.8125998973846436,
      "learning_rate": 2.9522419906673786e-05,
      "loss": 0.0534,
      "num_input_tokens_seen": 33516464,
      "step": 17685
    },
    {
      "epoch": 2.8858797618076517,
      "grad_norm": 1.001678705215454,
      "learning_rate": 2.951276404908474e-05,
      "loss": 0.1073,
      "num_input_tokens_seen": 33526224,
      "step": 17690
    },
    {
      "epoch": 2.886695489028469,
      "grad_norm": 0.07737500220537186,
      "learning_rate": 2.9503107495570752e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 33536672,
      "step": 17695
    },
    {
      "epoch": 2.887511216249286,
      "grad_norm": 0.1266665756702423,
      "learning_rate": 2.9493450247621003e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 33545184,
      "step": 17700
    },
    {
      "epoch": 2.8883269434701035,
      "grad_norm": 0.574131965637207,
      "learning_rate": 2.948379230672476e-05,
      "loss": 0.0643,
      "num_input_tokens_seen": 33555392,
      "step": 17705
    },
    {
      "epoch": 2.889142670690921,
      "grad_norm": 1.7109999656677246,
      "learning_rate": 2.9474133674371396e-05,
      "loss": 0.192,
      "num_input_tokens_seen": 33565024,
      "step": 17710
    },
    {
      "epoch": 2.8899583979117383,
      "grad_norm": 0.0939188003540039,
      "learning_rate": 2.9464474352050387e-05,
      "loss": 0.0869,
      "num_input_tokens_seen": 33574192,
      "step": 17715
    },
    {
      "epoch": 2.890774125132556,
      "grad_norm": 0.8813491463661194,
      "learning_rate": 2.9454814341251336e-05,
      "loss": 0.0921,
      "num_input_tokens_seen": 33583776,
      "step": 17720
    },
    {
      "epoch": 2.891589852353373,
      "grad_norm": 0.19257594645023346,
      "learning_rate": 2.9445153643463942e-05,
      "loss": 0.0556,
      "num_input_tokens_seen": 33594032,
      "step": 17725
    },
    {
      "epoch": 2.8924055795741905,
      "grad_norm": 0.0648842453956604,
      "learning_rate": 2.943549226017798e-05,
      "loss": 0.0549,
      "num_input_tokens_seen": 33604960,
      "step": 17730
    },
    {
      "epoch": 2.8932213067950077,
      "grad_norm": 0.036788832396268845,
      "learning_rate": 2.942583019288337e-05,
      "loss": 0.1215,
      "num_input_tokens_seen": 33614160,
      "step": 17735
    },
    {
      "epoch": 2.8940370340158252,
      "grad_norm": 0.7989298105239868,
      "learning_rate": 2.9416167443070132e-05,
      "loss": 0.1203,
      "num_input_tokens_seen": 33623696,
      "step": 17740
    },
    {
      "epoch": 2.8948527612366424,
      "grad_norm": 0.6916013360023499,
      "learning_rate": 2.9406504012228375e-05,
      "loss": 0.2317,
      "num_input_tokens_seen": 33632672,
      "step": 17745
    },
    {
      "epoch": 2.89566848845746,
      "grad_norm": 1.3272178173065186,
      "learning_rate": 2.939683990184832e-05,
      "loss": 0.0653,
      "num_input_tokens_seen": 33642720,
      "step": 17750
    },
    {
      "epoch": 2.896484215678277,
      "grad_norm": 0.6900888085365295,
      "learning_rate": 2.93871751134203e-05,
      "loss": 0.0795,
      "num_input_tokens_seen": 33651936,
      "step": 17755
    },
    {
      "epoch": 2.8972999428990946,
      "grad_norm": 0.6291124224662781,
      "learning_rate": 2.9377509648434752e-05,
      "loss": 0.0746,
      "num_input_tokens_seen": 33662288,
      "step": 17760
    },
    {
      "epoch": 2.898115670119912,
      "grad_norm": 0.3871579170227051,
      "learning_rate": 2.9367843508382203e-05,
      "loss": 0.0751,
      "num_input_tokens_seen": 33670736,
      "step": 17765
    },
    {
      "epoch": 2.8989313973407294,
      "grad_norm": 0.05065181478857994,
      "learning_rate": 2.9358176694753293e-05,
      "loss": 0.0538,
      "num_input_tokens_seen": 33680096,
      "step": 17770
    },
    {
      "epoch": 2.8997471245615465,
      "grad_norm": 1.3862793445587158,
      "learning_rate": 2.9348509209038766e-05,
      "loss": 0.1071,
      "num_input_tokens_seen": 33690240,
      "step": 17775
    },
    {
      "epoch": 2.900562851782364,
      "grad_norm": 1.0930399894714355,
      "learning_rate": 2.933884105272947e-05,
      "loss": 0.135,
      "num_input_tokens_seen": 33699680,
      "step": 17780
    },
    {
      "epoch": 2.901378579003181,
      "grad_norm": 0.37142860889434814,
      "learning_rate": 2.9329172227316366e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 33710656,
      "step": 17785
    },
    {
      "epoch": 2.9021943062239988,
      "grad_norm": 0.18196067214012146,
      "learning_rate": 2.93195027342905e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 33720240,
      "step": 17790
    },
    {
      "epoch": 2.903010033444816,
      "grad_norm": 1.211427927017212,
      "learning_rate": 2.9309832575143024e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 33730640,
      "step": 17795
    },
    {
      "epoch": 2.9038257606656335,
      "grad_norm": 0.4795752167701721,
      "learning_rate": 2.930016175136521e-05,
      "loss": 0.1,
      "num_input_tokens_seen": 33739984,
      "step": 17800
    },
    {
      "epoch": 2.9038257606656335,
      "eval_loss": 0.144253209233284,
      "eval_runtime": 133.2699,
      "eval_samples_per_second": 20.447,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 33739984,
      "step": 17800
    },
    {
      "epoch": 2.904641487886451,
      "grad_norm": 0.5663754343986511,
      "learning_rate": 2.9290490264448412e-05,
      "loss": 0.0549,
      "num_input_tokens_seen": 33748704,
      "step": 17805
    },
    {
      "epoch": 2.905457215107268,
      "grad_norm": 0.630117654800415,
      "learning_rate": 2.9280818115884094e-05,
      "loss": 0.1977,
      "num_input_tokens_seen": 33757984,
      "step": 17810
    },
    {
      "epoch": 2.9062729423280853,
      "grad_norm": 0.27005451917648315,
      "learning_rate": 2.9271145307163828e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 33767808,
      "step": 17815
    },
    {
      "epoch": 2.907088669548903,
      "grad_norm": 1.9960891008377075,
      "learning_rate": 2.9261471839779287e-05,
      "loss": 0.2164,
      "num_input_tokens_seen": 33777296,
      "step": 17820
    },
    {
      "epoch": 2.9079043967697205,
      "grad_norm": 0.5388262867927551,
      "learning_rate": 2.925179771522223e-05,
      "loss": 0.0671,
      "num_input_tokens_seen": 33786592,
      "step": 17825
    },
    {
      "epoch": 2.9087201239905376,
      "grad_norm": 0.9479618668556213,
      "learning_rate": 2.9242122934984535e-05,
      "loss": 0.2263,
      "num_input_tokens_seen": 33796512,
      "step": 17830
    },
    {
      "epoch": 2.9095358512113547,
      "grad_norm": 1.5906480550765991,
      "learning_rate": 2.9232447500558176e-05,
      "loss": 0.2014,
      "num_input_tokens_seen": 33807280,
      "step": 17835
    },
    {
      "epoch": 2.9103515784321723,
      "grad_norm": 0.0787779912352562,
      "learning_rate": 2.9222771413435225e-05,
      "loss": 0.1216,
      "num_input_tokens_seen": 33816144,
      "step": 17840
    },
    {
      "epoch": 2.91116730565299,
      "grad_norm": 0.5789855122566223,
      "learning_rate": 2.9213094675107848e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 33824880,
      "step": 17845
    },
    {
      "epoch": 2.911983032873807,
      "grad_norm": 1.75790274143219,
      "learning_rate": 2.9203417287068335e-05,
      "loss": 0.2259,
      "num_input_tokens_seen": 33833920,
      "step": 17850
    },
    {
      "epoch": 2.912798760094624,
      "grad_norm": 0.8375682830810547,
      "learning_rate": 2.9193739250809042e-05,
      "loss": 0.2048,
      "num_input_tokens_seen": 33843008,
      "step": 17855
    },
    {
      "epoch": 2.9136144873154417,
      "grad_norm": 1.017781376838684,
      "learning_rate": 2.9184060567822463e-05,
      "loss": 0.2498,
      "num_input_tokens_seen": 33854464,
      "step": 17860
    },
    {
      "epoch": 2.9144302145362593,
      "grad_norm": 0.2925262451171875,
      "learning_rate": 2.9174381239601166e-05,
      "loss": 0.1963,
      "num_input_tokens_seen": 33862800,
      "step": 17865
    },
    {
      "epoch": 2.9152459417570764,
      "grad_norm": 0.2461843341588974,
      "learning_rate": 2.916470126763783e-05,
      "loss": 0.0723,
      "num_input_tokens_seen": 33871536,
      "step": 17870
    },
    {
      "epoch": 2.9160616689778935,
      "grad_norm": 0.16865940392017365,
      "learning_rate": 2.9155020653425203e-05,
      "loss": 0.1286,
      "num_input_tokens_seen": 33879792,
      "step": 17875
    },
    {
      "epoch": 2.916877396198711,
      "grad_norm": 0.12342286854982376,
      "learning_rate": 2.9145339398456184e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 33888960,
      "step": 17880
    },
    {
      "epoch": 2.9176931234195287,
      "grad_norm": 0.11574512720108032,
      "learning_rate": 2.913565750422374e-05,
      "loss": 0.3188,
      "num_input_tokens_seen": 33898112,
      "step": 17885
    },
    {
      "epoch": 2.918508850640346,
      "grad_norm": 1.215790867805481,
      "learning_rate": 2.9125974972220938e-05,
      "loss": 0.1823,
      "num_input_tokens_seen": 33907536,
      "step": 17890
    },
    {
      "epoch": 2.919324577861163,
      "grad_norm": 0.5479663014411926,
      "learning_rate": 2.9116291803940932e-05,
      "loss": 0.1107,
      "num_input_tokens_seen": 33916432,
      "step": 17895
    },
    {
      "epoch": 2.9201403050819805,
      "grad_norm": 0.5508849620819092,
      "learning_rate": 2.910660800087701e-05,
      "loss": 0.0682,
      "num_input_tokens_seen": 33925696,
      "step": 17900
    },
    {
      "epoch": 2.920956032302798,
      "grad_norm": 2.127878189086914,
      "learning_rate": 2.909692356452254e-05,
      "loss": 0.4238,
      "num_input_tokens_seen": 33935632,
      "step": 17905
    },
    {
      "epoch": 2.9217717595236152,
      "grad_norm": 0.10391651093959808,
      "learning_rate": 2.9087238496370962e-05,
      "loss": 0.1075,
      "num_input_tokens_seen": 33945696,
      "step": 17910
    },
    {
      "epoch": 2.922587486744433,
      "grad_norm": 0.2311338484287262,
      "learning_rate": 2.907755279791583e-05,
      "loss": 0.1043,
      "num_input_tokens_seen": 33954928,
      "step": 17915
    },
    {
      "epoch": 2.92340321396525,
      "grad_norm": 0.45768821239471436,
      "learning_rate": 2.906786647065083e-05,
      "loss": 0.2375,
      "num_input_tokens_seen": 33965536,
      "step": 17920
    },
    {
      "epoch": 2.9242189411860675,
      "grad_norm": 0.2007332742214203,
      "learning_rate": 2.9058179516069695e-05,
      "loss": 0.0898,
      "num_input_tokens_seen": 33974336,
      "step": 17925
    },
    {
      "epoch": 2.9250346684068846,
      "grad_norm": 0.051049184054136276,
      "learning_rate": 2.9048491935666282e-05,
      "loss": 0.0709,
      "num_input_tokens_seen": 33983696,
      "step": 17930
    },
    {
      "epoch": 2.925850395627702,
      "grad_norm": 2.1914143562316895,
      "learning_rate": 2.9038803730934534e-05,
      "loss": 0.26,
      "num_input_tokens_seen": 33993600,
      "step": 17935
    },
    {
      "epoch": 2.9266661228485193,
      "grad_norm": 0.09416591376066208,
      "learning_rate": 2.9029114903368503e-05,
      "loss": 0.0421,
      "num_input_tokens_seen": 34003664,
      "step": 17940
    },
    {
      "epoch": 2.927481850069337,
      "grad_norm": 0.18998895585536957,
      "learning_rate": 2.9019425454462318e-05,
      "loss": 0.106,
      "num_input_tokens_seen": 34013840,
      "step": 17945
    },
    {
      "epoch": 2.928297577290154,
      "grad_norm": 0.6722466349601746,
      "learning_rate": 2.9009735385710212e-05,
      "loss": 0.1536,
      "num_input_tokens_seen": 34022576,
      "step": 17950
    },
    {
      "epoch": 2.9291133045109716,
      "grad_norm": 0.6397283673286438,
      "learning_rate": 2.900004469860652e-05,
      "loss": 0.1335,
      "num_input_tokens_seen": 34032432,
      "step": 17955
    },
    {
      "epoch": 2.9299290317317888,
      "grad_norm": 0.11357611417770386,
      "learning_rate": 2.8990353394645668e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 34042512,
      "step": 17960
    },
    {
      "epoch": 2.9307447589526063,
      "grad_norm": 0.1723993569612503,
      "learning_rate": 2.8980661475322186e-05,
      "loss": 0.0456,
      "num_input_tokens_seen": 34052848,
      "step": 17965
    },
    {
      "epoch": 2.9315604861734235,
      "grad_norm": 0.3182675838470459,
      "learning_rate": 2.897096894213067e-05,
      "loss": 0.1225,
      "num_input_tokens_seen": 34062432,
      "step": 17970
    },
    {
      "epoch": 2.932376213394241,
      "grad_norm": 1.5048415660858154,
      "learning_rate": 2.8961275796565845e-05,
      "loss": 0.3042,
      "num_input_tokens_seen": 34071616,
      "step": 17975
    },
    {
      "epoch": 2.933191940615058,
      "grad_norm": 1.1212903261184692,
      "learning_rate": 2.8951582040122517e-05,
      "loss": 0.1694,
      "num_input_tokens_seen": 34081808,
      "step": 17980
    },
    {
      "epoch": 2.9340076678358757,
      "grad_norm": 1.197619080543518,
      "learning_rate": 2.894188767429557e-05,
      "loss": 0.1433,
      "num_input_tokens_seen": 34090704,
      "step": 17985
    },
    {
      "epoch": 2.934823395056693,
      "grad_norm": 0.9613530039787292,
      "learning_rate": 2.8932192700580014e-05,
      "loss": 0.2136,
      "num_input_tokens_seen": 34100624,
      "step": 17990
    },
    {
      "epoch": 2.9356391222775104,
      "grad_norm": 0.26312577724456787,
      "learning_rate": 2.8922497120470916e-05,
      "loss": 0.0984,
      "num_input_tokens_seen": 34110688,
      "step": 17995
    },
    {
      "epoch": 2.936454849498328,
      "grad_norm": 0.19471098482608795,
      "learning_rate": 2.891280093546348e-05,
      "loss": 0.0681,
      "num_input_tokens_seen": 34121824,
      "step": 18000
    },
    {
      "epoch": 2.936454849498328,
      "eval_loss": 0.14418886601924896,
      "eval_runtime": 133.2596,
      "eval_samples_per_second": 20.449,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 34121824,
      "step": 18000
    },
    {
      "epoch": 2.937270576719145,
      "grad_norm": 0.3371475338935852,
      "learning_rate": 2.890310414705297e-05,
      "loss": 0.0761,
      "num_input_tokens_seen": 34131984,
      "step": 18005
    },
    {
      "epoch": 2.9380863039399623,
      "grad_norm": 1.3230489492416382,
      "learning_rate": 2.8893406756734742e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 34141376,
      "step": 18010
    },
    {
      "epoch": 2.93890203116078,
      "grad_norm": 0.11709468066692352,
      "learning_rate": 2.888370876600427e-05,
      "loss": 0.0897,
      "num_input_tokens_seen": 34150864,
      "step": 18015
    },
    {
      "epoch": 2.9397177583815974,
      "grad_norm": 0.11299438029527664,
      "learning_rate": 2.8874010176357104e-05,
      "loss": 0.0492,
      "num_input_tokens_seen": 34160640,
      "step": 18020
    },
    {
      "epoch": 2.9405334856024146,
      "grad_norm": 0.6455227136611938,
      "learning_rate": 2.886431098928888e-05,
      "loss": 0.0754,
      "num_input_tokens_seen": 34170672,
      "step": 18025
    },
    {
      "epoch": 2.9413492128232317,
      "grad_norm": 1.639887809753418,
      "learning_rate": 2.885461120629534e-05,
      "loss": 0.293,
      "num_input_tokens_seen": 34179936,
      "step": 18030
    },
    {
      "epoch": 2.9421649400440493,
      "grad_norm": 0.8445166945457458,
      "learning_rate": 2.8844910828872317e-05,
      "loss": 0.0484,
      "num_input_tokens_seen": 34187808,
      "step": 18035
    },
    {
      "epoch": 2.942980667264867,
      "grad_norm": 1.4006587266921997,
      "learning_rate": 2.8835209858515715e-05,
      "loss": 0.2021,
      "num_input_tokens_seen": 34197504,
      "step": 18040
    },
    {
      "epoch": 2.943796394485684,
      "grad_norm": 0.1790137141942978,
      "learning_rate": 2.8825508296721566e-05,
      "loss": 0.0993,
      "num_input_tokens_seen": 34208048,
      "step": 18045
    },
    {
      "epoch": 2.944612121706501,
      "grad_norm": 0.5324270129203796,
      "learning_rate": 2.881580614498596e-05,
      "loss": 0.1194,
      "num_input_tokens_seen": 34217616,
      "step": 18050
    },
    {
      "epoch": 2.9454278489273187,
      "grad_norm": 0.11881746351718903,
      "learning_rate": 2.8806103404805103e-05,
      "loss": 0.0894,
      "num_input_tokens_seen": 34227824,
      "step": 18055
    },
    {
      "epoch": 2.9462435761481363,
      "grad_norm": 2.0620436668395996,
      "learning_rate": 2.8796400077675257e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 34238016,
      "step": 18060
    },
    {
      "epoch": 2.9470593033689534,
      "grad_norm": 0.07929282635450363,
      "learning_rate": 2.8786696165092812e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 34248208,
      "step": 18065
    },
    {
      "epoch": 2.9478750305897705,
      "grad_norm": 0.18954788148403168,
      "learning_rate": 2.8776991668554236e-05,
      "loss": 0.1232,
      "num_input_tokens_seen": 34257472,
      "step": 18070
    },
    {
      "epoch": 2.948690757810588,
      "grad_norm": 0.8343013525009155,
      "learning_rate": 2.876728658955608e-05,
      "loss": 0.081,
      "num_input_tokens_seen": 34267344,
      "step": 18075
    },
    {
      "epoch": 2.9495064850314057,
      "grad_norm": 0.12129337340593338,
      "learning_rate": 2.8757580929594986e-05,
      "loss": 0.0302,
      "num_input_tokens_seen": 34276896,
      "step": 18080
    },
    {
      "epoch": 2.950322212252223,
      "grad_norm": 0.7083820700645447,
      "learning_rate": 2.87478746901677e-05,
      "loss": 0.0928,
      "num_input_tokens_seen": 34284960,
      "step": 18085
    },
    {
      "epoch": 2.9511379394730404,
      "grad_norm": 0.5147595405578613,
      "learning_rate": 2.873816787277103e-05,
      "loss": 0.0407,
      "num_input_tokens_seen": 34294352,
      "step": 18090
    },
    {
      "epoch": 2.9519536666938575,
      "grad_norm": 0.7328729033470154,
      "learning_rate": 2.8728460478901903e-05,
      "loss": 0.3567,
      "num_input_tokens_seen": 34301680,
      "step": 18095
    },
    {
      "epoch": 2.952769393914675,
      "grad_norm": 0.1878916472196579,
      "learning_rate": 2.8718752510057307e-05,
      "loss": 0.1494,
      "num_input_tokens_seen": 34311024,
      "step": 18100
    },
    {
      "epoch": 2.953585121135492,
      "grad_norm": 0.3621782064437866,
      "learning_rate": 2.870904396773435e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 34320224,
      "step": 18105
    },
    {
      "epoch": 2.95440084835631,
      "grad_norm": 0.37216803431510925,
      "learning_rate": 2.86993348534302e-05,
      "loss": 0.077,
      "num_input_tokens_seen": 34330112,
      "step": 18110
    },
    {
      "epoch": 2.955216575577127,
      "grad_norm": 0.9153695702552795,
      "learning_rate": 2.868962516864212e-05,
      "loss": 0.1688,
      "num_input_tokens_seen": 34339024,
      "step": 18115
    },
    {
      "epoch": 2.9560323027979445,
      "grad_norm": 1.9762808084487915,
      "learning_rate": 2.8679914914867477e-05,
      "loss": 0.1699,
      "num_input_tokens_seen": 34348272,
      "step": 18120
    },
    {
      "epoch": 2.9568480300187616,
      "grad_norm": 0.841729462146759,
      "learning_rate": 2.8670204093603713e-05,
      "loss": 0.1082,
      "num_input_tokens_seen": 34358512,
      "step": 18125
    },
    {
      "epoch": 2.957663757239579,
      "grad_norm": 0.8883981704711914,
      "learning_rate": 2.8660492706348357e-05,
      "loss": 0.0719,
      "num_input_tokens_seen": 34367184,
      "step": 18130
    },
    {
      "epoch": 2.9584794844603963,
      "grad_norm": 0.186110258102417,
      "learning_rate": 2.8650780754599022e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 34377136,
      "step": 18135
    },
    {
      "epoch": 2.959295211681214,
      "grad_norm": 0.22917573153972626,
      "learning_rate": 2.8641068239853407e-05,
      "loss": 0.0955,
      "num_input_tokens_seen": 34385568,
      "step": 18140
    },
    {
      "epoch": 2.960110938902031,
      "grad_norm": 0.21871934831142426,
      "learning_rate": 2.863135516360932e-05,
      "loss": 0.0477,
      "num_input_tokens_seen": 34395904,
      "step": 18145
    },
    {
      "epoch": 2.9609266661228486,
      "grad_norm": 1.6137531995773315,
      "learning_rate": 2.8621641527364633e-05,
      "loss": 0.2443,
      "num_input_tokens_seen": 34406592,
      "step": 18150
    },
    {
      "epoch": 2.9617423933436657,
      "grad_norm": 0.5145681500434875,
      "learning_rate": 2.8611927332617313e-05,
      "loss": 0.0803,
      "num_input_tokens_seen": 34416224,
      "step": 18155
    },
    {
      "epoch": 2.9625581205644833,
      "grad_norm": 0.3324536383152008,
      "learning_rate": 2.8602212580865405e-05,
      "loss": 0.1067,
      "num_input_tokens_seen": 34425104,
      "step": 18160
    },
    {
      "epoch": 2.9633738477853004,
      "grad_norm": 1.8857675790786743,
      "learning_rate": 2.859249727360705e-05,
      "loss": 0.2731,
      "num_input_tokens_seen": 34435040,
      "step": 18165
    },
    {
      "epoch": 2.964189575006118,
      "grad_norm": 2.2969493865966797,
      "learning_rate": 2.8582781412340465e-05,
      "loss": 0.1182,
      "num_input_tokens_seen": 34445680,
      "step": 18170
    },
    {
      "epoch": 2.965005302226935,
      "grad_norm": 0.0357184000313282,
      "learning_rate": 2.857306499856397e-05,
      "loss": 0.1486,
      "num_input_tokens_seen": 34455168,
      "step": 18175
    },
    {
      "epoch": 2.9658210294477527,
      "grad_norm": 0.32676970958709717,
      "learning_rate": 2.856334803377594e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 34463904,
      "step": 18180
    },
    {
      "epoch": 2.96663675666857,
      "grad_norm": 1.4103642702102661,
      "learning_rate": 2.8553630519474867e-05,
      "loss": 0.1006,
      "num_input_tokens_seen": 34473312,
      "step": 18185
    },
    {
      "epoch": 2.9674524838893874,
      "grad_norm": 1.224892020225525,
      "learning_rate": 2.8543912457159317e-05,
      "loss": 0.2123,
      "num_input_tokens_seen": 34481680,
      "step": 18190
    },
    {
      "epoch": 2.968268211110205,
      "grad_norm": 1.5075538158416748,
      "learning_rate": 2.853419384832792e-05,
      "loss": 0.1867,
      "num_input_tokens_seen": 34489472,
      "step": 18195
    },
    {
      "epoch": 2.969083938331022,
      "grad_norm": 1.7951922416687012,
      "learning_rate": 2.8524474694479423e-05,
      "loss": 0.0941,
      "num_input_tokens_seen": 34498368,
      "step": 18200
    },
    {
      "epoch": 2.969083938331022,
      "eval_loss": 0.1481918841600418,
      "eval_runtime": 133.1395,
      "eval_samples_per_second": 20.467,
      "eval_steps_per_second": 10.237,
      "num_input_tokens_seen": 34498368,
      "step": 18200
    },
    {
      "epoch": 2.9698996655518393,
      "grad_norm": 1.5271544456481934,
      "learning_rate": 2.851475499711264e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 34507008,
      "step": 18205
    },
    {
      "epoch": 2.970715392772657,
      "grad_norm": 0.16829200088977814,
      "learning_rate": 2.8505034757726468e-05,
      "loss": 0.0537,
      "num_input_tokens_seen": 34515840,
      "step": 18210
    },
    {
      "epoch": 2.9715311199934744,
      "grad_norm": 0.4709819257259369,
      "learning_rate": 2.8495313977819886e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 34525648,
      "step": 18215
    },
    {
      "epoch": 2.9723468472142915,
      "grad_norm": 0.36378058791160583,
      "learning_rate": 2.8485592658891956e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 34534416,
      "step": 18220
    },
    {
      "epoch": 2.9731625744351087,
      "grad_norm": 0.39452293515205383,
      "learning_rate": 2.8475870802441844e-05,
      "loss": 0.1969,
      "num_input_tokens_seen": 34543984,
      "step": 18225
    },
    {
      "epoch": 2.9739783016559262,
      "grad_norm": 1.3170620203018188,
      "learning_rate": 2.8466148409968774e-05,
      "loss": 0.339,
      "num_input_tokens_seen": 34551984,
      "step": 18230
    },
    {
      "epoch": 2.974794028876744,
      "grad_norm": 0.20533356070518494,
      "learning_rate": 2.8456425482972067e-05,
      "loss": 0.0883,
      "num_input_tokens_seen": 34560752,
      "step": 18235
    },
    {
      "epoch": 2.975609756097561,
      "grad_norm": 1.764917254447937,
      "learning_rate": 2.84467020229511e-05,
      "loss": 0.1144,
      "num_input_tokens_seen": 34569472,
      "step": 18240
    },
    {
      "epoch": 2.976425483318378,
      "grad_norm": 0.10694712400436401,
      "learning_rate": 2.8436978031405375e-05,
      "loss": 0.1183,
      "num_input_tokens_seen": 34578528,
      "step": 18245
    },
    {
      "epoch": 2.9772412105391957,
      "grad_norm": 0.5818459987640381,
      "learning_rate": 2.842725350983445e-05,
      "loss": 0.126,
      "num_input_tokens_seen": 34588000,
      "step": 18250
    },
    {
      "epoch": 2.9780569377600132,
      "grad_norm": 0.6547546982765198,
      "learning_rate": 2.8417528459737957e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 34596096,
      "step": 18255
    },
    {
      "epoch": 2.9788726649808304,
      "grad_norm": 0.3591856360435486,
      "learning_rate": 2.8407802882615624e-05,
      "loss": 0.0903,
      "num_input_tokens_seen": 34605920,
      "step": 18260
    },
    {
      "epoch": 2.9796883922016475,
      "grad_norm": 0.9323648810386658,
      "learning_rate": 2.8398076779967277e-05,
      "loss": 0.0768,
      "num_input_tokens_seen": 34614752,
      "step": 18265
    },
    {
      "epoch": 2.980504119422465,
      "grad_norm": 0.07681816816329956,
      "learning_rate": 2.8388350153292774e-05,
      "loss": 0.0788,
      "num_input_tokens_seen": 34624112,
      "step": 18270
    },
    {
      "epoch": 2.9813198466432826,
      "grad_norm": 0.5417879819869995,
      "learning_rate": 2.8378623004092103e-05,
      "loss": 0.0802,
      "num_input_tokens_seen": 34633456,
      "step": 18275
    },
    {
      "epoch": 2.9821355738640998,
      "grad_norm": 0.6622781753540039,
      "learning_rate": 2.8368895333865302e-05,
      "loss": 0.2559,
      "num_input_tokens_seen": 34642176,
      "step": 18280
    },
    {
      "epoch": 2.9829513010849174,
      "grad_norm": 0.4040444493293762,
      "learning_rate": 2.835916714411251e-05,
      "loss": 0.0877,
      "num_input_tokens_seen": 34650976,
      "step": 18285
    },
    {
      "epoch": 2.9837670283057345,
      "grad_norm": 0.1351557970046997,
      "learning_rate": 2.8349438436333926e-05,
      "loss": 0.2189,
      "num_input_tokens_seen": 34660576,
      "step": 18290
    },
    {
      "epoch": 2.984582755526552,
      "grad_norm": 0.2918570637702942,
      "learning_rate": 2.833970921202984e-05,
      "loss": 0.0825,
      "num_input_tokens_seen": 34671744,
      "step": 18295
    },
    {
      "epoch": 2.985398482747369,
      "grad_norm": 0.49552440643310547,
      "learning_rate": 2.8329979472700628e-05,
      "loss": 0.1955,
      "num_input_tokens_seen": 34680544,
      "step": 18300
    },
    {
      "epoch": 2.9862142099681868,
      "grad_norm": 1.648369550704956,
      "learning_rate": 2.832024921984674e-05,
      "loss": 0.4116,
      "num_input_tokens_seen": 34689040,
      "step": 18305
    },
    {
      "epoch": 2.987029937189004,
      "grad_norm": 0.30196940898895264,
      "learning_rate": 2.8310518454968693e-05,
      "loss": 0.054,
      "num_input_tokens_seen": 34699568,
      "step": 18310
    },
    {
      "epoch": 2.9878456644098215,
      "grad_norm": 0.1601170152425766,
      "learning_rate": 2.8300787179567095e-05,
      "loss": 0.05,
      "num_input_tokens_seen": 34709120,
      "step": 18315
    },
    {
      "epoch": 2.9886613916306386,
      "grad_norm": 0.364993691444397,
      "learning_rate": 2.8291055395142636e-05,
      "loss": 0.0997,
      "num_input_tokens_seen": 34716496,
      "step": 18320
    },
    {
      "epoch": 2.989477118851456,
      "grad_norm": 0.6079339385032654,
      "learning_rate": 2.8281323103196073e-05,
      "loss": 0.0478,
      "num_input_tokens_seen": 34727248,
      "step": 18325
    },
    {
      "epoch": 2.9902928460722733,
      "grad_norm": 0.4147399961948395,
      "learning_rate": 2.8271590305228256e-05,
      "loss": 0.192,
      "num_input_tokens_seen": 34737760,
      "step": 18330
    },
    {
      "epoch": 2.991108573293091,
      "grad_norm": 0.5360404253005981,
      "learning_rate": 2.82618570027401e-05,
      "loss": 0.2322,
      "num_input_tokens_seen": 34746800,
      "step": 18335
    },
    {
      "epoch": 2.991924300513908,
      "grad_norm": 1.613867163658142,
      "learning_rate": 2.8252123197232604e-05,
      "loss": 0.0624,
      "num_input_tokens_seen": 34756624,
      "step": 18340
    },
    {
      "epoch": 2.9927400277347256,
      "grad_norm": 1.3283237218856812,
      "learning_rate": 2.8242388890206843e-05,
      "loss": 0.1084,
      "num_input_tokens_seen": 34766224,
      "step": 18345
    },
    {
      "epoch": 2.9935557549555427,
      "grad_norm": 0.36583608388900757,
      "learning_rate": 2.8232654083163967e-05,
      "loss": 0.081,
      "num_input_tokens_seen": 34775664,
      "step": 18350
    },
    {
      "epoch": 2.9943714821763603,
      "grad_norm": 2.63325572013855,
      "learning_rate": 2.822291877760521e-05,
      "loss": 0.2511,
      "num_input_tokens_seen": 34785440,
      "step": 18355
    },
    {
      "epoch": 2.9951872093971774,
      "grad_norm": 0.5791681408882141,
      "learning_rate": 2.8213182975031864e-05,
      "loss": 0.0666,
      "num_input_tokens_seen": 34795040,
      "step": 18360
    },
    {
      "epoch": 2.996002936617995,
      "grad_norm": 0.5349048376083374,
      "learning_rate": 2.8203446676945337e-05,
      "loss": 0.1025,
      "num_input_tokens_seen": 34804080,
      "step": 18365
    },
    {
      "epoch": 2.9968186638388126,
      "grad_norm": 0.22636033594608307,
      "learning_rate": 2.8193709884847075e-05,
      "loss": 0.123,
      "num_input_tokens_seen": 34813856,
      "step": 18370
    },
    {
      "epoch": 2.9976343910596297,
      "grad_norm": 1.1098507642745972,
      "learning_rate": 2.8183972600238605e-05,
      "loss": 0.0815,
      "num_input_tokens_seen": 34823056,
      "step": 18375
    },
    {
      "epoch": 2.998450118280447,
      "grad_norm": 1.9944432973861694,
      "learning_rate": 2.817423482462156e-05,
      "loss": 0.1211,
      "num_input_tokens_seen": 34831376,
      "step": 18380
    },
    {
      "epoch": 2.9992658455012644,
      "grad_norm": 0.557998776435852,
      "learning_rate": 2.8164496559497605e-05,
      "loss": 0.0399,
      "num_input_tokens_seen": 34839488,
      "step": 18385
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.18677577376365662,
      "learning_rate": 2.815475780636852e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 34846688,
      "step": 18390
    },
    {
      "epoch": 3.0008157272208176,
      "grad_norm": 0.0853983461856842,
      "learning_rate": 2.814501856673613e-05,
      "loss": 0.0186,
      "num_input_tokens_seen": 34855408,
      "step": 18395
    },
    {
      "epoch": 3.0016314544416347,
      "grad_norm": 0.09182015061378479,
      "learning_rate": 2.8135278842102353e-05,
      "loss": 0.0614,
      "num_input_tokens_seen": 34866272,
      "step": 18400
    },
    {
      "epoch": 3.0016314544416347,
      "eval_loss": 0.14648741483688354,
      "eval_runtime": 133.2185,
      "eval_samples_per_second": 20.455,
      "eval_steps_per_second": 10.231,
      "num_input_tokens_seen": 34866272,
      "step": 18400
    },
    {
      "epoch": 3.0024471816624523,
      "grad_norm": 0.10318555682897568,
      "learning_rate": 2.8125538633969183e-05,
      "loss": 0.0824,
      "num_input_tokens_seen": 34876304,
      "step": 18405
    },
    {
      "epoch": 3.0032629088832694,
      "grad_norm": 1.285258412361145,
      "learning_rate": 2.8115797943838677e-05,
      "loss": 0.2288,
      "num_input_tokens_seen": 34884768,
      "step": 18410
    },
    {
      "epoch": 3.004078636104087,
      "grad_norm": 1.5477107763290405,
      "learning_rate": 2.810605677321298e-05,
      "loss": 0.3306,
      "num_input_tokens_seen": 34895248,
      "step": 18415
    },
    {
      "epoch": 3.004894363324904,
      "grad_norm": 1.1421581506729126,
      "learning_rate": 2.809631512359428e-05,
      "loss": 0.0759,
      "num_input_tokens_seen": 34904032,
      "step": 18420
    },
    {
      "epoch": 3.0057100905457217,
      "grad_norm": 2.3783936500549316,
      "learning_rate": 2.8086572996484884e-05,
      "loss": 0.0952,
      "num_input_tokens_seen": 34914288,
      "step": 18425
    },
    {
      "epoch": 3.006525817766539,
      "grad_norm": 0.5608576536178589,
      "learning_rate": 2.8076830393387143e-05,
      "loss": 0.0953,
      "num_input_tokens_seen": 34923520,
      "step": 18430
    },
    {
      "epoch": 3.0073415449873564,
      "grad_norm": 0.2094935178756714,
      "learning_rate": 2.8067087315803497e-05,
      "loss": 0.0172,
      "num_input_tokens_seen": 34933392,
      "step": 18435
    },
    {
      "epoch": 3.0081572722081735,
      "grad_norm": 1.5942788124084473,
      "learning_rate": 2.8057343765236433e-05,
      "loss": 0.2002,
      "num_input_tokens_seen": 34943360,
      "step": 18440
    },
    {
      "epoch": 3.008972999428991,
      "grad_norm": 1.8055521249771118,
      "learning_rate": 2.804759974318854e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 34953472,
      "step": 18445
    },
    {
      "epoch": 3.0097887266498082,
      "grad_norm": 0.2382371574640274,
      "learning_rate": 2.8037855251162482e-05,
      "loss": 0.1502,
      "num_input_tokens_seen": 34963536,
      "step": 18450
    },
    {
      "epoch": 3.010604453870626,
      "grad_norm": 0.43729713559150696,
      "learning_rate": 2.802811029066096e-05,
      "loss": 0.2125,
      "num_input_tokens_seen": 34974448,
      "step": 18455
    },
    {
      "epoch": 3.011420181091443,
      "grad_norm": 0.6144698262214661,
      "learning_rate": 2.8018364863186764e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 34985008,
      "step": 18460
    },
    {
      "epoch": 3.0122359083122605,
      "grad_norm": 0.15894486010074615,
      "learning_rate": 2.800861897024279e-05,
      "loss": 0.0714,
      "num_input_tokens_seen": 34994832,
      "step": 18465
    },
    {
      "epoch": 3.0130516355330776,
      "grad_norm": 0.6797276735305786,
      "learning_rate": 2.799887261333196e-05,
      "loss": 0.1006,
      "num_input_tokens_seen": 35004544,
      "step": 18470
    },
    {
      "epoch": 3.013867362753895,
      "grad_norm": 0.1073317676782608,
      "learning_rate": 2.798912579395728e-05,
      "loss": 0.2904,
      "num_input_tokens_seen": 35013552,
      "step": 18475
    },
    {
      "epoch": 3.0146830899747123,
      "grad_norm": 0.1459168940782547,
      "learning_rate": 2.797937851362185e-05,
      "loss": 0.0741,
      "num_input_tokens_seen": 35023312,
      "step": 18480
    },
    {
      "epoch": 3.01549881719553,
      "grad_norm": 0.22977884113788605,
      "learning_rate": 2.7969630773828802e-05,
      "loss": 0.0497,
      "num_input_tokens_seen": 35031568,
      "step": 18485
    },
    {
      "epoch": 3.016314544416347,
      "grad_norm": 0.27335938811302185,
      "learning_rate": 2.7959882576081382e-05,
      "loss": 0.0857,
      "num_input_tokens_seen": 35040784,
      "step": 18490
    },
    {
      "epoch": 3.0171302716371646,
      "grad_norm": 0.7452739477157593,
      "learning_rate": 2.795013392188286e-05,
      "loss": 0.0927,
      "num_input_tokens_seen": 35050592,
      "step": 18495
    },
    {
      "epoch": 3.0179459988579818,
      "grad_norm": 0.6865739822387695,
      "learning_rate": 2.7940384812736614e-05,
      "loss": 0.1255,
      "num_input_tokens_seen": 35060784,
      "step": 18500
    },
    {
      "epoch": 3.0187617260787993,
      "grad_norm": 0.17749929428100586,
      "learning_rate": 2.7930635250146087e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 35070464,
      "step": 18505
    },
    {
      "epoch": 3.0195774532996165,
      "grad_norm": 0.7636072635650635,
      "learning_rate": 2.792088523561477e-05,
      "loss": 0.0661,
      "num_input_tokens_seen": 35081456,
      "step": 18510
    },
    {
      "epoch": 3.020393180520434,
      "grad_norm": 0.853136420249939,
      "learning_rate": 2.7911134770646246e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 35091680,
      "step": 18515
    },
    {
      "epoch": 3.021208907741251,
      "grad_norm": 0.7042958736419678,
      "learning_rate": 2.7901383856744157e-05,
      "loss": 0.1178,
      "num_input_tokens_seen": 35101216,
      "step": 18520
    },
    {
      "epoch": 3.0220246349620687,
      "grad_norm": 1.2045464515686035,
      "learning_rate": 2.7891632495412217e-05,
      "loss": 0.0759,
      "num_input_tokens_seen": 35111360,
      "step": 18525
    },
    {
      "epoch": 3.022840362182886,
      "grad_norm": 0.06543738394975662,
      "learning_rate": 2.7881880688154205e-05,
      "loss": 0.0385,
      "num_input_tokens_seen": 35121760,
      "step": 18530
    },
    {
      "epoch": 3.0236560894037035,
      "grad_norm": 0.16917161643505096,
      "learning_rate": 2.7872128436473977e-05,
      "loss": 0.2554,
      "num_input_tokens_seen": 35131504,
      "step": 18535
    },
    {
      "epoch": 3.0244718166245206,
      "grad_norm": 0.7122541666030884,
      "learning_rate": 2.7862375741875448e-05,
      "loss": 0.2607,
      "num_input_tokens_seen": 35141472,
      "step": 18540
    },
    {
      "epoch": 3.025287543845338,
      "grad_norm": 0.7663626074790955,
      "learning_rate": 2.785262260586261e-05,
      "loss": 0.1532,
      "num_input_tokens_seen": 35151824,
      "step": 18545
    },
    {
      "epoch": 3.0261032710661553,
      "grad_norm": 0.5567265748977661,
      "learning_rate": 2.7842869029939517e-05,
      "loss": 0.0925,
      "num_input_tokens_seen": 35160704,
      "step": 18550
    },
    {
      "epoch": 3.026918998286973,
      "grad_norm": 0.11251068115234375,
      "learning_rate": 2.7833115015610296e-05,
      "loss": 0.1125,
      "num_input_tokens_seen": 35170304,
      "step": 18555
    },
    {
      "epoch": 3.02773472550779,
      "grad_norm": 0.19517336785793304,
      "learning_rate": 2.7823360564379136e-05,
      "loss": 0.0364,
      "num_input_tokens_seen": 35178880,
      "step": 18560
    },
    {
      "epoch": 3.0285504527286076,
      "grad_norm": 0.09017453342676163,
      "learning_rate": 2.7813605677750297e-05,
      "loss": 0.1574,
      "num_input_tokens_seen": 35187104,
      "step": 18565
    },
    {
      "epoch": 3.0293661799494247,
      "grad_norm": 0.712607204914093,
      "learning_rate": 2.7803850357228102e-05,
      "loss": 0.1055,
      "num_input_tokens_seen": 35198032,
      "step": 18570
    },
    {
      "epoch": 3.0301819071702423,
      "grad_norm": 0.9714052081108093,
      "learning_rate": 2.779409460431695e-05,
      "loss": 0.1017,
      "num_input_tokens_seen": 35208224,
      "step": 18575
    },
    {
      "epoch": 3.03099763439106,
      "grad_norm": 0.13365301489830017,
      "learning_rate": 2.778433842052129e-05,
      "loss": 0.0379,
      "num_input_tokens_seen": 35218144,
      "step": 18580
    },
    {
      "epoch": 3.031813361611877,
      "grad_norm": 0.5565543174743652,
      "learning_rate": 2.7774581807345664e-05,
      "loss": 0.1773,
      "num_input_tokens_seen": 35228720,
      "step": 18585
    },
    {
      "epoch": 3.0326290888326946,
      "grad_norm": 0.38568374514579773,
      "learning_rate": 2.776482476629465e-05,
      "loss": 0.0443,
      "num_input_tokens_seen": 35238576,
      "step": 18590
    },
    {
      "epoch": 3.0334448160535117,
      "grad_norm": 1.5081193447113037,
      "learning_rate": 2.7755067298872924e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 35247552,
      "step": 18595
    },
    {
      "epoch": 3.0342605432743293,
      "grad_norm": 0.030844802036881447,
      "learning_rate": 2.774530940658518e-05,
      "loss": 0.0636,
      "num_input_tokens_seen": 35258768,
      "step": 18600
    },
    {
      "epoch": 3.0342605432743293,
      "eval_loss": 0.14593404531478882,
      "eval_runtime": 133.2916,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 35258768,
      "step": 18600
    },
    {
      "epoch": 3.0350762704951464,
      "grad_norm": 1.181764006614685,
      "learning_rate": 2.7735551090936236e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 35268160,
      "step": 18605
    },
    {
      "epoch": 3.035891997715964,
      "grad_norm": 0.4736557900905609,
      "learning_rate": 2.7725792353430934e-05,
      "loss": 0.2049,
      "num_input_tokens_seen": 35278176,
      "step": 18610
    },
    {
      "epoch": 3.036707724936781,
      "grad_norm": 0.5690107345581055,
      "learning_rate": 2.77160331955742e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 35287856,
      "step": 18615
    },
    {
      "epoch": 3.0375234521575987,
      "grad_norm": 0.4115108549594879,
      "learning_rate": 2.7706273618871008e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 35298128,
      "step": 18620
    },
    {
      "epoch": 3.038339179378416,
      "grad_norm": 0.25542980432510376,
      "learning_rate": 2.769651362482642e-05,
      "loss": 0.1944,
      "num_input_tokens_seen": 35307312,
      "step": 18625
    },
    {
      "epoch": 3.0391549065992334,
      "grad_norm": 1.0905897617340088,
      "learning_rate": 2.768675321494555e-05,
      "loss": 0.1146,
      "num_input_tokens_seen": 35316608,
      "step": 18630
    },
    {
      "epoch": 3.0399706338200505,
      "grad_norm": 0.511404275894165,
      "learning_rate": 2.7676992390733565e-05,
      "loss": 0.0468,
      "num_input_tokens_seen": 35326512,
      "step": 18635
    },
    {
      "epoch": 3.040786361040868,
      "grad_norm": 2.3643131256103516,
      "learning_rate": 2.766723115369571e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 35335504,
      "step": 18640
    },
    {
      "epoch": 3.041602088261685,
      "grad_norm": 0.06742221862077713,
      "learning_rate": 2.765746950533729e-05,
      "loss": 0.0865,
      "num_input_tokens_seen": 35344800,
      "step": 18645
    },
    {
      "epoch": 3.042417815482503,
      "grad_norm": 1.299147605895996,
      "learning_rate": 2.7647707447163684e-05,
      "loss": 0.2987,
      "num_input_tokens_seen": 35355616,
      "step": 18650
    },
    {
      "epoch": 3.04323354270332,
      "grad_norm": 0.631173849105835,
      "learning_rate": 2.7637944980680315e-05,
      "loss": 0.0904,
      "num_input_tokens_seen": 35364960,
      "step": 18655
    },
    {
      "epoch": 3.0440492699241375,
      "grad_norm": 0.3031715154647827,
      "learning_rate": 2.762818210739268e-05,
      "loss": 0.1858,
      "num_input_tokens_seen": 35374544,
      "step": 18660
    },
    {
      "epoch": 3.0448649971449546,
      "grad_norm": 0.7223507761955261,
      "learning_rate": 2.7618418828806332e-05,
      "loss": 0.0906,
      "num_input_tokens_seen": 35383392,
      "step": 18665
    },
    {
      "epoch": 3.045680724365772,
      "grad_norm": 1.2832283973693848,
      "learning_rate": 2.76086551464269e-05,
      "loss": 0.0849,
      "num_input_tokens_seen": 35392320,
      "step": 18670
    },
    {
      "epoch": 3.0464964515865893,
      "grad_norm": 0.875906765460968,
      "learning_rate": 2.759889106176006e-05,
      "loss": 0.1066,
      "num_input_tokens_seen": 35403248,
      "step": 18675
    },
    {
      "epoch": 3.047312178807407,
      "grad_norm": 0.41800162196159363,
      "learning_rate": 2.758912657631156e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 35413248,
      "step": 18680
    },
    {
      "epoch": 3.048127906028224,
      "grad_norm": 0.48220720887184143,
      "learning_rate": 2.7579361691587198e-05,
      "loss": 0.0813,
      "num_input_tokens_seen": 35423936,
      "step": 18685
    },
    {
      "epoch": 3.0489436332490416,
      "grad_norm": 0.9803263545036316,
      "learning_rate": 2.756959640909285e-05,
      "loss": 0.1131,
      "num_input_tokens_seen": 35434448,
      "step": 18690
    },
    {
      "epoch": 3.0497593604698587,
      "grad_norm": 1.0977559089660645,
      "learning_rate": 2.7559830730334452e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 35443840,
      "step": 18695
    },
    {
      "epoch": 3.0505750876906763,
      "grad_norm": 1.616128921508789,
      "learning_rate": 2.7550064656817988e-05,
      "loss": 0.0916,
      "num_input_tokens_seen": 35452672,
      "step": 18700
    },
    {
      "epoch": 3.0513908149114934,
      "grad_norm": 0.921881914138794,
      "learning_rate": 2.7540298190049503e-05,
      "loss": 0.1316,
      "num_input_tokens_seen": 35463072,
      "step": 18705
    },
    {
      "epoch": 3.052206542132311,
      "grad_norm": 0.638408899307251,
      "learning_rate": 2.7530531331535107e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 35472640,
      "step": 18710
    },
    {
      "epoch": 3.053022269353128,
      "grad_norm": 0.8500091433525085,
      "learning_rate": 2.752076408278099e-05,
      "loss": 0.1121,
      "num_input_tokens_seen": 35481744,
      "step": 18715
    },
    {
      "epoch": 3.0538379965739457,
      "grad_norm": 1.20624577999115,
      "learning_rate": 2.751099644529337e-05,
      "loss": 0.095,
      "num_input_tokens_seen": 35492560,
      "step": 18720
    },
    {
      "epoch": 3.054653723794763,
      "grad_norm": 0.02568722702562809,
      "learning_rate": 2.7501228420578533e-05,
      "loss": 0.0834,
      "num_input_tokens_seen": 35500736,
      "step": 18725
    },
    {
      "epoch": 3.0554694510155804,
      "grad_norm": 0.5612598061561584,
      "learning_rate": 2.7491460010142857e-05,
      "loss": 0.2444,
      "num_input_tokens_seen": 35510448,
      "step": 18730
    },
    {
      "epoch": 3.0562851782363976,
      "grad_norm": 0.41782715916633606,
      "learning_rate": 2.7481691215492727e-05,
      "loss": 0.0433,
      "num_input_tokens_seen": 35519712,
      "step": 18735
    },
    {
      "epoch": 3.057100905457215,
      "grad_norm": 1.4420666694641113,
      "learning_rate": 2.747192203813463e-05,
      "loss": 0.0856,
      "num_input_tokens_seen": 35529776,
      "step": 18740
    },
    {
      "epoch": 3.0579166326780323,
      "grad_norm": 3.087578058242798,
      "learning_rate": 2.7462152479575087e-05,
      "loss": 0.2194,
      "num_input_tokens_seen": 35539056,
      "step": 18745
    },
    {
      "epoch": 3.05873235989885,
      "grad_norm": 1.0736262798309326,
      "learning_rate": 2.7452382541320697e-05,
      "loss": 0.2302,
      "num_input_tokens_seen": 35548624,
      "step": 18750
    },
    {
      "epoch": 3.059548087119667,
      "grad_norm": 0.445639044046402,
      "learning_rate": 2.7442612224878096e-05,
      "loss": 0.244,
      "num_input_tokens_seen": 35558416,
      "step": 18755
    },
    {
      "epoch": 3.0603638143404845,
      "grad_norm": 0.6862762570381165,
      "learning_rate": 2.7432841531753994e-05,
      "loss": 0.0644,
      "num_input_tokens_seen": 35568512,
      "step": 18760
    },
    {
      "epoch": 3.0611795415613017,
      "grad_norm": 0.34475311636924744,
      "learning_rate": 2.7423070463455147e-05,
      "loss": 0.2083,
      "num_input_tokens_seen": 35578016,
      "step": 18765
    },
    {
      "epoch": 3.0619952687821193,
      "grad_norm": 1.7734761238098145,
      "learning_rate": 2.7413299021488397e-05,
      "loss": 0.2004,
      "num_input_tokens_seen": 35586432,
      "step": 18770
    },
    {
      "epoch": 3.062810996002937,
      "grad_norm": 0.7703030109405518,
      "learning_rate": 2.7403527207360615e-05,
      "loss": 0.0939,
      "num_input_tokens_seen": 35596448,
      "step": 18775
    },
    {
      "epoch": 3.063626723223754,
      "grad_norm": 0.20917831361293793,
      "learning_rate": 2.7393755022578722e-05,
      "loss": 0.1718,
      "num_input_tokens_seen": 35605200,
      "step": 18780
    },
    {
      "epoch": 3.0644424504445715,
      "grad_norm": 0.6996634602546692,
      "learning_rate": 2.7383982468649714e-05,
      "loss": 0.063,
      "num_input_tokens_seen": 35615184,
      "step": 18785
    },
    {
      "epoch": 3.0652581776653887,
      "grad_norm": 0.5879335403442383,
      "learning_rate": 2.7374209547080665e-05,
      "loss": 0.0697,
      "num_input_tokens_seen": 35624224,
      "step": 18790
    },
    {
      "epoch": 3.0660739048862062,
      "grad_norm": 1.8012531995773315,
      "learning_rate": 2.7364436259378663e-05,
      "loss": 0.232,
      "num_input_tokens_seen": 35634608,
      "step": 18795
    },
    {
      "epoch": 3.0668896321070234,
      "grad_norm": 0.9470393061637878,
      "learning_rate": 2.735466260705088e-05,
      "loss": 0.1946,
      "num_input_tokens_seen": 35644416,
      "step": 18800
    },
    {
      "epoch": 3.0668896321070234,
      "eval_loss": 0.1436465084552765,
      "eval_runtime": 133.3374,
      "eval_samples_per_second": 20.437,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 35644416,
      "step": 18800
    },
    {
      "epoch": 3.067705359327841,
      "grad_norm": 0.9808899760246277,
      "learning_rate": 2.7344888591604524e-05,
      "loss": 0.1219,
      "num_input_tokens_seen": 35654160,
      "step": 18805
    },
    {
      "epoch": 3.068521086548658,
      "grad_norm": 0.6103624701499939,
      "learning_rate": 2.7335114214546893e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 35662480,
      "step": 18810
    },
    {
      "epoch": 3.0693368137694756,
      "grad_norm": 0.8837024569511414,
      "learning_rate": 2.7325339477385293e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 35671648,
      "step": 18815
    },
    {
      "epoch": 3.0701525409902928,
      "grad_norm": 1.1614993810653687,
      "learning_rate": 2.7315564381627128e-05,
      "loss": 0.2104,
      "num_input_tokens_seen": 35681552,
      "step": 18820
    },
    {
      "epoch": 3.0709682682111104,
      "grad_norm": 1.052341103553772,
      "learning_rate": 2.7305788928779835e-05,
      "loss": 0.2195,
      "num_input_tokens_seen": 35691056,
      "step": 18825
    },
    {
      "epoch": 3.0717839954319275,
      "grad_norm": 0.3076252341270447,
      "learning_rate": 2.729601312035091e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 35698768,
      "step": 18830
    },
    {
      "epoch": 3.072599722652745,
      "grad_norm": 0.8611893653869629,
      "learning_rate": 2.7286236957847915e-05,
      "loss": 0.0742,
      "num_input_tokens_seen": 35707312,
      "step": 18835
    },
    {
      "epoch": 3.073415449873562,
      "grad_norm": 0.7332620024681091,
      "learning_rate": 2.7276460442778446e-05,
      "loss": 0.0564,
      "num_input_tokens_seen": 35716656,
      "step": 18840
    },
    {
      "epoch": 3.0742311770943798,
      "grad_norm": 1.9567338228225708,
      "learning_rate": 2.726668357665017e-05,
      "loss": 0.2107,
      "num_input_tokens_seen": 35727184,
      "step": 18845
    },
    {
      "epoch": 3.075046904315197,
      "grad_norm": 0.6292383670806885,
      "learning_rate": 2.7256906360970808e-05,
      "loss": 0.0743,
      "num_input_tokens_seen": 35736992,
      "step": 18850
    },
    {
      "epoch": 3.0758626315360145,
      "grad_norm": 1.62075674533844,
      "learning_rate": 2.7247128797248117e-05,
      "loss": 0.1945,
      "num_input_tokens_seen": 35745776,
      "step": 18855
    },
    {
      "epoch": 3.0766783587568316,
      "grad_norm": 0.43547365069389343,
      "learning_rate": 2.7237350886989925e-05,
      "loss": 0.1833,
      "num_input_tokens_seen": 35754288,
      "step": 18860
    },
    {
      "epoch": 3.077494085977649,
      "grad_norm": 2.37772798538208,
      "learning_rate": 2.7227572631704107e-05,
      "loss": 0.1973,
      "num_input_tokens_seen": 35763392,
      "step": 18865
    },
    {
      "epoch": 3.0783098131984663,
      "grad_norm": 0.8120166659355164,
      "learning_rate": 2.7217794032898596e-05,
      "loss": 0.0833,
      "num_input_tokens_seen": 35773104,
      "step": 18870
    },
    {
      "epoch": 3.079125540419284,
      "grad_norm": 0.238632932305336,
      "learning_rate": 2.7208015092081384e-05,
      "loss": 0.1868,
      "num_input_tokens_seen": 35783104,
      "step": 18875
    },
    {
      "epoch": 3.079941267640101,
      "grad_norm": 0.4528937041759491,
      "learning_rate": 2.719823581076049e-05,
      "loss": 0.2127,
      "num_input_tokens_seen": 35791232,
      "step": 18880
    },
    {
      "epoch": 3.0807569948609186,
      "grad_norm": 1.4852184057235718,
      "learning_rate": 2.718845619044401e-05,
      "loss": 0.1012,
      "num_input_tokens_seen": 35800400,
      "step": 18885
    },
    {
      "epoch": 3.0815727220817357,
      "grad_norm": 0.28320759534835815,
      "learning_rate": 2.7178676232640088e-05,
      "loss": 0.0286,
      "num_input_tokens_seen": 35809920,
      "step": 18890
    },
    {
      "epoch": 3.0823884493025533,
      "grad_norm": 0.3391270041465759,
      "learning_rate": 2.716889593885691e-05,
      "loss": 0.0652,
      "num_input_tokens_seen": 35817984,
      "step": 18895
    },
    {
      "epoch": 3.0832041765233704,
      "grad_norm": 0.09946509450674057,
      "learning_rate": 2.7159115310602716e-05,
      "loss": 0.0737,
      "num_input_tokens_seen": 35828672,
      "step": 18900
    },
    {
      "epoch": 3.084019903744188,
      "grad_norm": 0.6992310881614685,
      "learning_rate": 2.7149334349385814e-05,
      "loss": 0.0776,
      "num_input_tokens_seen": 35838208,
      "step": 18905
    },
    {
      "epoch": 3.084835630965005,
      "grad_norm": 0.6084586381912231,
      "learning_rate": 2.713955305671454e-05,
      "loss": 0.1153,
      "num_input_tokens_seen": 35847504,
      "step": 18910
    },
    {
      "epoch": 3.0856513581858227,
      "grad_norm": 1.0006930828094482,
      "learning_rate": 2.71297714340973e-05,
      "loss": 0.2389,
      "num_input_tokens_seen": 35857136,
      "step": 18915
    },
    {
      "epoch": 3.08646708540664,
      "grad_norm": 1.682196855545044,
      "learning_rate": 2.7119989483042545e-05,
      "loss": 0.1595,
      "num_input_tokens_seen": 35866976,
      "step": 18920
    },
    {
      "epoch": 3.0872828126274574,
      "grad_norm": 0.1960950493812561,
      "learning_rate": 2.7110207205058768e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 35875744,
      "step": 18925
    },
    {
      "epoch": 3.0880985398482745,
      "grad_norm": 0.2854374051094055,
      "learning_rate": 2.7100424601654517e-05,
      "loss": 0.0685,
      "num_input_tokens_seen": 35884480,
      "step": 18930
    },
    {
      "epoch": 3.088914267069092,
      "grad_norm": 0.6432121992111206,
      "learning_rate": 2.7090641674338403e-05,
      "loss": 0.0821,
      "num_input_tokens_seen": 35893408,
      "step": 18935
    },
    {
      "epoch": 3.0897299942899092,
      "grad_norm": 1.8675224781036377,
      "learning_rate": 2.7080858424619072e-05,
      "loss": 0.1697,
      "num_input_tokens_seen": 35903488,
      "step": 18940
    },
    {
      "epoch": 3.090545721510727,
      "grad_norm": 1.0138046741485596,
      "learning_rate": 2.707107485400521e-05,
      "loss": 0.1225,
      "num_input_tokens_seen": 35912160,
      "step": 18945
    },
    {
      "epoch": 3.0913614487315444,
      "grad_norm": 0.1792621910572052,
      "learning_rate": 2.7061290964005586e-05,
      "loss": 0.0568,
      "num_input_tokens_seen": 35921760,
      "step": 18950
    },
    {
      "epoch": 3.0921771759523615,
      "grad_norm": 0.15901759266853333,
      "learning_rate": 2.7051506756129e-05,
      "loss": 0.0965,
      "num_input_tokens_seen": 35932208,
      "step": 18955
    },
    {
      "epoch": 3.092992903173179,
      "grad_norm": 0.5942751169204712,
      "learning_rate": 2.704172223188428e-05,
      "loss": 0.067,
      "num_input_tokens_seen": 35940800,
      "step": 18960
    },
    {
      "epoch": 3.0938086303939962,
      "grad_norm": 0.2629292607307434,
      "learning_rate": 2.7031937392780334e-05,
      "loss": 0.1048,
      "num_input_tokens_seen": 35950224,
      "step": 18965
    },
    {
      "epoch": 3.094624357614814,
      "grad_norm": 0.9377290606498718,
      "learning_rate": 2.702215224032611e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 35959248,
      "step": 18970
    },
    {
      "epoch": 3.095440084835631,
      "grad_norm": 1.0978859663009644,
      "learning_rate": 2.70123667760306e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 35968944,
      "step": 18975
    },
    {
      "epoch": 3.0962558120564485,
      "grad_norm": 0.0592348538339138,
      "learning_rate": 2.7002581001402845e-05,
      "loss": 0.0375,
      "num_input_tokens_seen": 35979520,
      "step": 18980
    },
    {
      "epoch": 3.0970715392772656,
      "grad_norm": 0.9018471240997314,
      "learning_rate": 2.6992794917951923e-05,
      "loss": 0.1232,
      "num_input_tokens_seen": 35989056,
      "step": 18985
    },
    {
      "epoch": 3.097887266498083,
      "grad_norm": 0.160149484872818,
      "learning_rate": 2.6983008527187e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 35998480,
      "step": 18990
    },
    {
      "epoch": 3.0987029937189003,
      "grad_norm": 0.06903714686632156,
      "learning_rate": 2.697322183061723e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 36008512,
      "step": 18995
    },
    {
      "epoch": 3.099518720939718,
      "grad_norm": 1.827720284461975,
      "learning_rate": 2.696343482975186e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 36017808,
      "step": 19000
    },
    {
      "epoch": 3.099518720939718,
      "eval_loss": 0.14507602155208588,
      "eval_runtime": 133.3588,
      "eval_samples_per_second": 20.434,
      "eval_steps_per_second": 10.221,
      "num_input_tokens_seen": 36017808,
      "step": 19000
    },
    {
      "epoch": 3.100334448160535,
      "grad_norm": 0.12229188531637192,
      "learning_rate": 2.695364752610016e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 36027632,
      "step": 19005
    },
    {
      "epoch": 3.1011501753813526,
      "grad_norm": 0.4686278700828552,
      "learning_rate": 2.6943859921171467e-05,
      "loss": 0.087,
      "num_input_tokens_seen": 36037088,
      "step": 19010
    },
    {
      "epoch": 3.1019659026021698,
      "grad_norm": 0.17180460691452026,
      "learning_rate": 2.6934072016475143e-05,
      "loss": 0.1747,
      "num_input_tokens_seen": 36046320,
      "step": 19015
    },
    {
      "epoch": 3.1027816298229873,
      "grad_norm": 0.37844085693359375,
      "learning_rate": 2.6924283813520606e-05,
      "loss": 0.0216,
      "num_input_tokens_seen": 36055616,
      "step": 19020
    },
    {
      "epoch": 3.1035973570438045,
      "grad_norm": 0.034448251128196716,
      "learning_rate": 2.691449531381733e-05,
      "loss": 0.2249,
      "num_input_tokens_seen": 36064864,
      "step": 19025
    },
    {
      "epoch": 3.104413084264622,
      "grad_norm": 0.08221510052680969,
      "learning_rate": 2.6904706518874816e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 36075296,
      "step": 19030
    },
    {
      "epoch": 3.105228811485439,
      "grad_norm": 0.564163088798523,
      "learning_rate": 2.6894917430202615e-05,
      "loss": 0.1051,
      "num_input_tokens_seen": 36084176,
      "step": 19035
    },
    {
      "epoch": 3.1060445387062567,
      "grad_norm": 0.0997392013669014,
      "learning_rate": 2.6885128049310343e-05,
      "loss": 0.2829,
      "num_input_tokens_seen": 36095712,
      "step": 19040
    },
    {
      "epoch": 3.106860265927074,
      "grad_norm": 0.3897484540939331,
      "learning_rate": 2.687533837770762e-05,
      "loss": 0.0999,
      "num_input_tokens_seen": 36105056,
      "step": 19045
    },
    {
      "epoch": 3.1076759931478914,
      "grad_norm": 0.45900383591651917,
      "learning_rate": 2.6865548416904162e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 36114688,
      "step": 19050
    },
    {
      "epoch": 3.1084917203687086,
      "grad_norm": 0.4809972047805786,
      "learning_rate": 2.68557581684097e-05,
      "loss": 0.2099,
      "num_input_tokens_seen": 36122784,
      "step": 19055
    },
    {
      "epoch": 3.109307447589526,
      "grad_norm": 1.2096199989318848,
      "learning_rate": 2.6845967633733998e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 36131728,
      "step": 19060
    },
    {
      "epoch": 3.1101231748103433,
      "grad_norm": 0.8602908253669739,
      "learning_rate": 2.683617681438689e-05,
      "loss": 0.1737,
      "num_input_tokens_seen": 36141728,
      "step": 19065
    },
    {
      "epoch": 3.110938902031161,
      "grad_norm": 0.19452370703220367,
      "learning_rate": 2.682638571187825e-05,
      "loss": 0.0305,
      "num_input_tokens_seen": 36149680,
      "step": 19070
    },
    {
      "epoch": 3.111754629251978,
      "grad_norm": 0.9526752233505249,
      "learning_rate": 2.6816594327717976e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 36160112,
      "step": 19075
    },
    {
      "epoch": 3.1125703564727956,
      "grad_norm": 0.06510965526103973,
      "learning_rate": 2.680680266341603e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 36169056,
      "step": 19080
    },
    {
      "epoch": 3.1133860836936127,
      "grad_norm": 0.6411513686180115,
      "learning_rate": 2.67970107204824e-05,
      "loss": 0.0686,
      "num_input_tokens_seen": 36177408,
      "step": 19085
    },
    {
      "epoch": 3.1142018109144303,
      "grad_norm": 1.1699707508087158,
      "learning_rate": 2.6787218500427142e-05,
      "loss": 0.1159,
      "num_input_tokens_seen": 36187120,
      "step": 19090
    },
    {
      "epoch": 3.1150175381352474,
      "grad_norm": 0.5656916499137878,
      "learning_rate": 2.6777426004760332e-05,
      "loss": 0.3204,
      "num_input_tokens_seen": 36198000,
      "step": 19095
    },
    {
      "epoch": 3.115833265356065,
      "grad_norm": 0.6725646257400513,
      "learning_rate": 2.6767633234992094e-05,
      "loss": 0.1003,
      "num_input_tokens_seen": 36206128,
      "step": 19100
    },
    {
      "epoch": 3.116648992576882,
      "grad_norm": 1.5526403188705444,
      "learning_rate": 2.6757840192632598e-05,
      "loss": 0.3323,
      "num_input_tokens_seen": 36215776,
      "step": 19105
    },
    {
      "epoch": 3.1174647197976997,
      "grad_norm": 0.2701238691806793,
      "learning_rate": 2.6748046879192052e-05,
      "loss": 0.1744,
      "num_input_tokens_seen": 36224272,
      "step": 19110
    },
    {
      "epoch": 3.118280447018517,
      "grad_norm": 1.308807134628296,
      "learning_rate": 2.673825329618071e-05,
      "loss": 0.1065,
      "num_input_tokens_seen": 36233760,
      "step": 19115
    },
    {
      "epoch": 3.1190961742393344,
      "grad_norm": 0.20973926782608032,
      "learning_rate": 2.6728459445108866e-05,
      "loss": 0.187,
      "num_input_tokens_seen": 36243584,
      "step": 19120
    },
    {
      "epoch": 3.1199119014601515,
      "grad_norm": 3.1745917797088623,
      "learning_rate": 2.6718665327486854e-05,
      "loss": 0.3159,
      "num_input_tokens_seen": 36253360,
      "step": 19125
    },
    {
      "epoch": 3.120727628680969,
      "grad_norm": 0.131002277135849,
      "learning_rate": 2.6708870944825048e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 36263040,
      "step": 19130
    },
    {
      "epoch": 3.121543355901786,
      "grad_norm": 0.4033568203449249,
      "learning_rate": 2.6699076298633874e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 36272624,
      "step": 19135
    },
    {
      "epoch": 3.122359083122604,
      "grad_norm": 0.5862872004508972,
      "learning_rate": 2.6689281390423788e-05,
      "loss": 0.1055,
      "num_input_tokens_seen": 36281952,
      "step": 19140
    },
    {
      "epoch": 3.1231748103434214,
      "grad_norm": 1.2867509126663208,
      "learning_rate": 2.667948622170527e-05,
      "loss": 0.0793,
      "num_input_tokens_seen": 36291808,
      "step": 19145
    },
    {
      "epoch": 3.1239905375642385,
      "grad_norm": 0.29157349467277527,
      "learning_rate": 2.6669690793988873e-05,
      "loss": 0.1839,
      "num_input_tokens_seen": 36300720,
      "step": 19150
    },
    {
      "epoch": 3.124806264785056,
      "grad_norm": 0.6889610886573792,
      "learning_rate": 2.665989510878518e-05,
      "loss": 0.1215,
      "num_input_tokens_seen": 36309552,
      "step": 19155
    },
    {
      "epoch": 3.125621992005873,
      "grad_norm": 0.18516317009925842,
      "learning_rate": 2.6650099167604793e-05,
      "loss": 0.036,
      "num_input_tokens_seen": 36319392,
      "step": 19160
    },
    {
      "epoch": 3.126437719226691,
      "grad_norm": 0.4903431534767151,
      "learning_rate": 2.6640302971958376e-05,
      "loss": 0.1113,
      "num_input_tokens_seen": 36327856,
      "step": 19165
    },
    {
      "epoch": 3.127253446447508,
      "grad_norm": 0.688447892665863,
      "learning_rate": 2.6630506523356635e-05,
      "loss": 0.1796,
      "num_input_tokens_seen": 36337184,
      "step": 19170
    },
    {
      "epoch": 3.1280691736683255,
      "grad_norm": 1.5775010585784912,
      "learning_rate": 2.6620709823310297e-05,
      "loss": 0.3692,
      "num_input_tokens_seen": 36347600,
      "step": 19175
    },
    {
      "epoch": 3.1288849008891426,
      "grad_norm": 1.0502007007598877,
      "learning_rate": 2.661091287333014e-05,
      "loss": 0.1438,
      "num_input_tokens_seen": 36356144,
      "step": 19180
    },
    {
      "epoch": 3.12970062810996,
      "grad_norm": 0.0821511298418045,
      "learning_rate": 2.660111567492696e-05,
      "loss": 0.1141,
      "num_input_tokens_seen": 36366240,
      "step": 19185
    },
    {
      "epoch": 3.1305163553307773,
      "grad_norm": 1.3233156204223633,
      "learning_rate": 2.6591318229611635e-05,
      "loss": 0.2348,
      "num_input_tokens_seen": 36375360,
      "step": 19190
    },
    {
      "epoch": 3.131332082551595,
      "grad_norm": 1.4603008031845093,
      "learning_rate": 2.6581520538895037e-05,
      "loss": 0.2935,
      "num_input_tokens_seen": 36385728,
      "step": 19195
    },
    {
      "epoch": 3.132147809772412,
      "grad_norm": 0.3463025391101837,
      "learning_rate": 2.6571722604288102e-05,
      "loss": 0.0472,
      "num_input_tokens_seen": 36393536,
      "step": 19200
    },
    {
      "epoch": 3.132147809772412,
      "eval_loss": 0.14226645231246948,
      "eval_runtime": 133.0704,
      "eval_samples_per_second": 20.478,
      "eval_steps_per_second": 10.243,
      "num_input_tokens_seen": 36393536,
      "step": 19200
    },
    {
      "epoch": 3.1329635369932296,
      "grad_norm": 0.6357391476631165,
      "learning_rate": 2.656192442730179e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 36402784,
      "step": 19205
    },
    {
      "epoch": 3.1337792642140467,
      "grad_norm": 0.9818727374076843,
      "learning_rate": 2.6552126009447098e-05,
      "loss": 0.0972,
      "num_input_tokens_seen": 36411568,
      "step": 19210
    },
    {
      "epoch": 3.1345949914348643,
      "grad_norm": 0.20253366231918335,
      "learning_rate": 2.654232735223507e-05,
      "loss": 0.043,
      "num_input_tokens_seen": 36420160,
      "step": 19215
    },
    {
      "epoch": 3.1354107186556814,
      "grad_norm": 0.3334088623523712,
      "learning_rate": 2.6532528457176787e-05,
      "loss": 0.1421,
      "num_input_tokens_seen": 36429040,
      "step": 19220
    },
    {
      "epoch": 3.136226445876499,
      "grad_norm": 0.8582351207733154,
      "learning_rate": 2.6522729325783348e-05,
      "loss": 0.1366,
      "num_input_tokens_seen": 36438112,
      "step": 19225
    },
    {
      "epoch": 3.137042173097316,
      "grad_norm": 0.12890063226222992,
      "learning_rate": 2.6512929959565914e-05,
      "loss": 0.1639,
      "num_input_tokens_seen": 36448080,
      "step": 19230
    },
    {
      "epoch": 3.1378579003181337,
      "grad_norm": 1.154517650604248,
      "learning_rate": 2.6503130360035673e-05,
      "loss": 0.1134,
      "num_input_tokens_seen": 36457568,
      "step": 19235
    },
    {
      "epoch": 3.138673627538951,
      "grad_norm": 0.6144842505455017,
      "learning_rate": 2.6493330528703835e-05,
      "loss": 0.2011,
      "num_input_tokens_seen": 36466608,
      "step": 19240
    },
    {
      "epoch": 3.1394893547597684,
      "grad_norm": 0.9543153643608093,
      "learning_rate": 2.648353046708167e-05,
      "loss": 0.0897,
      "num_input_tokens_seen": 36478256,
      "step": 19245
    },
    {
      "epoch": 3.1403050819805856,
      "grad_norm": 1.0708950757980347,
      "learning_rate": 2.647373017668046e-05,
      "loss": 0.0912,
      "num_input_tokens_seen": 36488384,
      "step": 19250
    },
    {
      "epoch": 3.141120809201403,
      "grad_norm": 0.4020916521549225,
      "learning_rate": 2.6463929659011537e-05,
      "loss": 0.3005,
      "num_input_tokens_seen": 36497808,
      "step": 19255
    },
    {
      "epoch": 3.1419365364222203,
      "grad_norm": 1.1109470129013062,
      "learning_rate": 2.6454128915586262e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 36506240,
      "step": 19260
    },
    {
      "epoch": 3.142752263643038,
      "grad_norm": 0.13494642078876495,
      "learning_rate": 2.6444327947916036e-05,
      "loss": 0.0488,
      "num_input_tokens_seen": 36515632,
      "step": 19265
    },
    {
      "epoch": 3.143567990863855,
      "grad_norm": 0.8851801753044128,
      "learning_rate": 2.6434526757512292e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 36525472,
      "step": 19270
    },
    {
      "epoch": 3.1443837180846725,
      "grad_norm": 0.22423966228961945,
      "learning_rate": 2.6424725345886486e-05,
      "loss": 0.0599,
      "num_input_tokens_seen": 36534880,
      "step": 19275
    },
    {
      "epoch": 3.1451994453054897,
      "grad_norm": 0.03484120965003967,
      "learning_rate": 2.641492371455014e-05,
      "loss": 0.108,
      "num_input_tokens_seen": 36543744,
      "step": 19280
    },
    {
      "epoch": 3.1460151725263072,
      "grad_norm": 0.23887157440185547,
      "learning_rate": 2.640512186501477e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 36553328,
      "step": 19285
    },
    {
      "epoch": 3.1468308997471244,
      "grad_norm": 1.5419766902923584,
      "learning_rate": 2.639531979879195e-05,
      "loss": 0.1418,
      "num_input_tokens_seen": 36562496,
      "step": 19290
    },
    {
      "epoch": 3.147646626967942,
      "grad_norm": 1.338732361793518,
      "learning_rate": 2.638551751739328e-05,
      "loss": 0.0966,
      "num_input_tokens_seen": 36572704,
      "step": 19295
    },
    {
      "epoch": 3.148462354188759,
      "grad_norm": 2.110773801803589,
      "learning_rate": 2.6375715022330404e-05,
      "loss": 0.1404,
      "num_input_tokens_seen": 36582832,
      "step": 19300
    },
    {
      "epoch": 3.1492780814095767,
      "grad_norm": 0.7149456143379211,
      "learning_rate": 2.6365912315114976e-05,
      "loss": 0.0619,
      "num_input_tokens_seen": 36593024,
      "step": 19305
    },
    {
      "epoch": 3.150093808630394,
      "grad_norm": 1.0819565057754517,
      "learning_rate": 2.6356109397258704e-05,
      "loss": 0.0597,
      "num_input_tokens_seen": 36603744,
      "step": 19310
    },
    {
      "epoch": 3.1509095358512114,
      "grad_norm": 0.2905436158180237,
      "learning_rate": 2.6346306270273325e-05,
      "loss": 0.0554,
      "num_input_tokens_seen": 36613216,
      "step": 19315
    },
    {
      "epoch": 3.151725263072029,
      "grad_norm": 0.03236503526568413,
      "learning_rate": 2.6336502935670608e-05,
      "loss": 0.0886,
      "num_input_tokens_seen": 36623024,
      "step": 19320
    },
    {
      "epoch": 3.152540990292846,
      "grad_norm": 0.3157848119735718,
      "learning_rate": 2.6326699394962333e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 36631616,
      "step": 19325
    },
    {
      "epoch": 3.153356717513663,
      "grad_norm": 1.1502619981765747,
      "learning_rate": 2.6316895649660334e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 36640288,
      "step": 19330
    },
    {
      "epoch": 3.1541724447344808,
      "grad_norm": 0.4124969244003296,
      "learning_rate": 2.6307091701276486e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 36649424,
      "step": 19335
    },
    {
      "epoch": 3.1549881719552983,
      "grad_norm": 1.61601984500885,
      "learning_rate": 2.629728755132267e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 36659728,
      "step": 19340
    },
    {
      "epoch": 3.1558038991761155,
      "grad_norm": 2.180417776107788,
      "learning_rate": 2.628748320131081e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 36668544,
      "step": 19345
    },
    {
      "epoch": 3.156619626396933,
      "grad_norm": 0.2028888761997223,
      "learning_rate": 2.6277678652752856e-05,
      "loss": 0.2381,
      "num_input_tokens_seen": 36677168,
      "step": 19350
    },
    {
      "epoch": 3.15743535361775,
      "grad_norm": 0.9687231779098511,
      "learning_rate": 2.6267873907160807e-05,
      "loss": 0.0805,
      "num_input_tokens_seen": 36686688,
      "step": 19355
    },
    {
      "epoch": 3.1582510808385678,
      "grad_norm": 0.26952072978019714,
      "learning_rate": 2.6258068966046668e-05,
      "loss": 0.0313,
      "num_input_tokens_seen": 36696896,
      "step": 19360
    },
    {
      "epoch": 3.159066808059385,
      "grad_norm": 0.23793846368789673,
      "learning_rate": 2.6248263830922475e-05,
      "loss": 0.0219,
      "num_input_tokens_seen": 36706688,
      "step": 19365
    },
    {
      "epoch": 3.1598825352802025,
      "grad_norm": 3.0379526615142822,
      "learning_rate": 2.6238458503300318e-05,
      "loss": 0.2286,
      "num_input_tokens_seen": 36716368,
      "step": 19370
    },
    {
      "epoch": 3.1606982625010196,
      "grad_norm": 1.127326488494873,
      "learning_rate": 2.6228652984692292e-05,
      "loss": 0.2728,
      "num_input_tokens_seen": 36724832,
      "step": 19375
    },
    {
      "epoch": 3.161513989721837,
      "grad_norm": 0.3502189815044403,
      "learning_rate": 2.621884727661054e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 36733904,
      "step": 19380
    },
    {
      "epoch": 3.1623297169426543,
      "grad_norm": 0.8200361132621765,
      "learning_rate": 2.6209041380567222e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 36743232,
      "step": 19385
    },
    {
      "epoch": 3.163145444163472,
      "grad_norm": 0.5024325251579285,
      "learning_rate": 2.6199235298074527e-05,
      "loss": 0.2937,
      "num_input_tokens_seen": 36752656,
      "step": 19390
    },
    {
      "epoch": 3.163961171384289,
      "grad_norm": 0.5696783065795898,
      "learning_rate": 2.618942903064468e-05,
      "loss": 0.117,
      "num_input_tokens_seen": 36761632,
      "step": 19395
    },
    {
      "epoch": 3.1647768986051066,
      "grad_norm": 1.6421458721160889,
      "learning_rate": 2.6179622579789932e-05,
      "loss": 0.3222,
      "num_input_tokens_seen": 36770432,
      "step": 19400
    },
    {
      "epoch": 3.1647768986051066,
      "eval_loss": 0.1427169293165207,
      "eval_runtime": 133.1674,
      "eval_samples_per_second": 20.463,
      "eval_steps_per_second": 10.235,
      "num_input_tokens_seen": 36770432,
      "step": 19400
    },
    {
      "epoch": 3.1655926258259237,
      "grad_norm": 0.759960412979126,
      "learning_rate": 2.6169815947022553e-05,
      "loss": 0.2889,
      "num_input_tokens_seen": 36780016,
      "step": 19405
    },
    {
      "epoch": 3.1664083530467413,
      "grad_norm": 0.6010296940803528,
      "learning_rate": 2.6160009133854853e-05,
      "loss": 0.0573,
      "num_input_tokens_seen": 36790352,
      "step": 19410
    },
    {
      "epoch": 3.1672240802675584,
      "grad_norm": 0.9881689548492432,
      "learning_rate": 2.6150202141799168e-05,
      "loss": 0.1689,
      "num_input_tokens_seen": 36800704,
      "step": 19415
    },
    {
      "epoch": 3.168039807488376,
      "grad_norm": 0.824562668800354,
      "learning_rate": 2.614039497236786e-05,
      "loss": 0.2153,
      "num_input_tokens_seen": 36810368,
      "step": 19420
    },
    {
      "epoch": 3.168855534709193,
      "grad_norm": 1.5234929323196411,
      "learning_rate": 2.6130587627073315e-05,
      "loss": 0.2428,
      "num_input_tokens_seen": 36820224,
      "step": 19425
    },
    {
      "epoch": 3.1696712619300107,
      "grad_norm": 0.7642940878868103,
      "learning_rate": 2.6120780107427956e-05,
      "loss": 0.0983,
      "num_input_tokens_seen": 36828720,
      "step": 19430
    },
    {
      "epoch": 3.170486989150828,
      "grad_norm": 0.7082297801971436,
      "learning_rate": 2.6110972414944214e-05,
      "loss": 0.2639,
      "num_input_tokens_seen": 36838224,
      "step": 19435
    },
    {
      "epoch": 3.1713027163716454,
      "grad_norm": 1.4088462591171265,
      "learning_rate": 2.6101164551134565e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 36848208,
      "step": 19440
    },
    {
      "epoch": 3.1721184435924625,
      "grad_norm": 0.049436233937740326,
      "learning_rate": 2.6091356517511505e-05,
      "loss": 0.0729,
      "num_input_tokens_seen": 36857424,
      "step": 19445
    },
    {
      "epoch": 3.17293417081328,
      "grad_norm": 0.6676864624023438,
      "learning_rate": 2.608154831558755e-05,
      "loss": 0.0569,
      "num_input_tokens_seen": 36867088,
      "step": 19450
    },
    {
      "epoch": 3.1737498980340972,
      "grad_norm": 1.2183572053909302,
      "learning_rate": 2.607173994687526e-05,
      "loss": 0.2043,
      "num_input_tokens_seen": 36877440,
      "step": 19455
    },
    {
      "epoch": 3.174565625254915,
      "grad_norm": 0.44274961948394775,
      "learning_rate": 2.6061931412887196e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 36885712,
      "step": 19460
    },
    {
      "epoch": 3.175381352475732,
      "grad_norm": 0.9289062023162842,
      "learning_rate": 2.6052122715135973e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 36896048,
      "step": 19465
    },
    {
      "epoch": 3.1761970796965495,
      "grad_norm": 0.35846370458602905,
      "learning_rate": 2.60423138551342e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 36905840,
      "step": 19470
    },
    {
      "epoch": 3.1770128069173666,
      "grad_norm": 0.4709022045135498,
      "learning_rate": 2.6032504834394527e-05,
      "loss": 0.1785,
      "num_input_tokens_seen": 36914848,
      "step": 19475
    },
    {
      "epoch": 3.1778285341381842,
      "grad_norm": 1.3193519115447998,
      "learning_rate": 2.602269565442964e-05,
      "loss": 0.1296,
      "num_input_tokens_seen": 36924928,
      "step": 19480
    },
    {
      "epoch": 3.1786442613590014,
      "grad_norm": 1.1667238473892212,
      "learning_rate": 2.6012886316752227e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 36934672,
      "step": 19485
    },
    {
      "epoch": 3.179459988579819,
      "grad_norm": 0.8621329069137573,
      "learning_rate": 2.6003076822875018e-05,
      "loss": 0.0835,
      "num_input_tokens_seen": 36945056,
      "step": 19490
    },
    {
      "epoch": 3.180275715800636,
      "grad_norm": 0.3050568699836731,
      "learning_rate": 2.5993267174310755e-05,
      "loss": 0.0975,
      "num_input_tokens_seen": 36954224,
      "step": 19495
    },
    {
      "epoch": 3.1810914430214536,
      "grad_norm": 0.9342478513717651,
      "learning_rate": 2.5983457372572218e-05,
      "loss": 0.1804,
      "num_input_tokens_seen": 36962976,
      "step": 19500
    },
    {
      "epoch": 3.1819071702422708,
      "grad_norm": 0.7393316030502319,
      "learning_rate": 2.597364741917219e-05,
      "loss": 0.1441,
      "num_input_tokens_seen": 36971360,
      "step": 19505
    },
    {
      "epoch": 3.1827228974630883,
      "grad_norm": 0.28650087118148804,
      "learning_rate": 2.5963837315623492e-05,
      "loss": 0.1271,
      "num_input_tokens_seen": 36980736,
      "step": 19510
    },
    {
      "epoch": 3.183538624683906,
      "grad_norm": 0.7958032488822937,
      "learning_rate": 2.595402706343897e-05,
      "loss": 0.0739,
      "num_input_tokens_seen": 36990352,
      "step": 19515
    },
    {
      "epoch": 3.184354351904723,
      "grad_norm": 0.46445226669311523,
      "learning_rate": 2.594421666413148e-05,
      "loss": 0.0583,
      "num_input_tokens_seen": 36999600,
      "step": 19520
    },
    {
      "epoch": 3.1851700791255406,
      "grad_norm": 0.9627357125282288,
      "learning_rate": 2.5934406119213928e-05,
      "loss": 0.0676,
      "num_input_tokens_seen": 37010272,
      "step": 19525
    },
    {
      "epoch": 3.1859858063463578,
      "grad_norm": 0.5048984885215759,
      "learning_rate": 2.5924595430199193e-05,
      "loss": 0.0861,
      "num_input_tokens_seen": 37020000,
      "step": 19530
    },
    {
      "epoch": 3.1868015335671753,
      "grad_norm": 1.0248440504074097,
      "learning_rate": 2.5914784598600238e-05,
      "loss": 0.1656,
      "num_input_tokens_seen": 37030288,
      "step": 19535
    },
    {
      "epoch": 3.1876172607879925,
      "grad_norm": 0.3550247550010681,
      "learning_rate": 2.5904973625930002e-05,
      "loss": 0.2785,
      "num_input_tokens_seen": 37040752,
      "step": 19540
    },
    {
      "epoch": 3.18843298800881,
      "grad_norm": 0.3652065098285675,
      "learning_rate": 2.5895162513701456e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 37051344,
      "step": 19545
    },
    {
      "epoch": 3.189248715229627,
      "grad_norm": 0.4317514896392822,
      "learning_rate": 2.5885351263427593e-05,
      "loss": 0.0771,
      "num_input_tokens_seen": 37060224,
      "step": 19550
    },
    {
      "epoch": 3.1900644424504447,
      "grad_norm": 1.7025212049484253,
      "learning_rate": 2.5875539876621448e-05,
      "loss": 0.2316,
      "num_input_tokens_seen": 37070576,
      "step": 19555
    },
    {
      "epoch": 3.190880169671262,
      "grad_norm": 0.0555223673582077,
      "learning_rate": 2.586572835479605e-05,
      "loss": 0.0382,
      "num_input_tokens_seen": 37079616,
      "step": 19560
    },
    {
      "epoch": 3.1916958968920794,
      "grad_norm": 1.2402129173278809,
      "learning_rate": 2.585591669946446e-05,
      "loss": 0.2808,
      "num_input_tokens_seen": 37088368,
      "step": 19565
    },
    {
      "epoch": 3.1925116241128966,
      "grad_norm": 0.476949542760849,
      "learning_rate": 2.5846104912139756e-05,
      "loss": 0.18,
      "num_input_tokens_seen": 37098224,
      "step": 19570
    },
    {
      "epoch": 3.193327351333714,
      "grad_norm": 0.5727080702781677,
      "learning_rate": 2.583629299433505e-05,
      "loss": 0.1234,
      "num_input_tokens_seen": 37107936,
      "step": 19575
    },
    {
      "epoch": 3.1941430785545313,
      "grad_norm": 0.05866629257798195,
      "learning_rate": 2.582648094756345e-05,
      "loss": 0.0968,
      "num_input_tokens_seen": 37117104,
      "step": 19580
    },
    {
      "epoch": 3.194958805775349,
      "grad_norm": 0.06938741356134415,
      "learning_rate": 2.5816668773338098e-05,
      "loss": 0.074,
      "num_input_tokens_seen": 37125424,
      "step": 19585
    },
    {
      "epoch": 3.195774532996166,
      "grad_norm": 0.08667370676994324,
      "learning_rate": 2.580685647317216e-05,
      "loss": 0.0278,
      "num_input_tokens_seen": 37134624,
      "step": 19590
    },
    {
      "epoch": 3.1965902602169836,
      "grad_norm": 0.5709192752838135,
      "learning_rate": 2.5797044048578818e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 37144016,
      "step": 19595
    },
    {
      "epoch": 3.1974059874378007,
      "grad_norm": 0.053866252303123474,
      "learning_rate": 2.5787231501071262e-05,
      "loss": 0.055,
      "num_input_tokens_seen": 37152448,
      "step": 19600
    },
    {
      "epoch": 3.1974059874378007,
      "eval_loss": 0.14219361543655396,
      "eval_runtime": 133.1101,
      "eval_samples_per_second": 20.472,
      "eval_steps_per_second": 10.24,
      "num_input_tokens_seen": 37152448,
      "step": 19600
    },
    {
      "epoch": 3.1982217146586183,
      "grad_norm": 0.20635102689266205,
      "learning_rate": 2.577741883216272e-05,
      "loss": 0.0656,
      "num_input_tokens_seen": 37162288,
      "step": 19605
    },
    {
      "epoch": 3.1990374418794354,
      "grad_norm": 0.39527061581611633,
      "learning_rate": 2.576760604336642e-05,
      "loss": 0.0718,
      "num_input_tokens_seen": 37169856,
      "step": 19610
    },
    {
      "epoch": 3.199853169100253,
      "grad_norm": 1.4313722848892212,
      "learning_rate": 2.575779313619563e-05,
      "loss": 0.2817,
      "num_input_tokens_seen": 37180544,
      "step": 19615
    },
    {
      "epoch": 3.20066889632107,
      "grad_norm": 0.1781388521194458,
      "learning_rate": 2.5747980112163605e-05,
      "loss": 0.0953,
      "num_input_tokens_seen": 37190208,
      "step": 19620
    },
    {
      "epoch": 3.2014846235418877,
      "grad_norm": 0.25433647632598877,
      "learning_rate": 2.5738166972783656e-05,
      "loss": 0.0963,
      "num_input_tokens_seen": 37199712,
      "step": 19625
    },
    {
      "epoch": 3.202300350762705,
      "grad_norm": 0.6134119033813477,
      "learning_rate": 2.5728353719569075e-05,
      "loss": 0.0901,
      "num_input_tokens_seen": 37209760,
      "step": 19630
    },
    {
      "epoch": 3.2031160779835224,
      "grad_norm": 1.4778037071228027,
      "learning_rate": 2.57185403540332e-05,
      "loss": 0.0931,
      "num_input_tokens_seen": 37218352,
      "step": 19635
    },
    {
      "epoch": 3.2039318052043395,
      "grad_norm": 0.18391533195972443,
      "learning_rate": 2.5708726877689375e-05,
      "loss": 0.1078,
      "num_input_tokens_seen": 37228000,
      "step": 19640
    },
    {
      "epoch": 3.204747532425157,
      "grad_norm": 0.023409826681017876,
      "learning_rate": 2.5698913292050964e-05,
      "loss": 0.1121,
      "num_input_tokens_seen": 37236192,
      "step": 19645
    },
    {
      "epoch": 3.205563259645974,
      "grad_norm": 0.20845027267932892,
      "learning_rate": 2.568909959863133e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 37245728,
      "step": 19650
    },
    {
      "epoch": 3.206378986866792,
      "grad_norm": 1.4525089263916016,
      "learning_rate": 2.5679285798943887e-05,
      "loss": 0.2174,
      "num_input_tokens_seen": 37253904,
      "step": 19655
    },
    {
      "epoch": 3.207194714087609,
      "grad_norm": 0.13052058219909668,
      "learning_rate": 2.5669471894502035e-05,
      "loss": 0.0906,
      "num_input_tokens_seen": 37264064,
      "step": 19660
    },
    {
      "epoch": 3.2080104413084265,
      "grad_norm": 0.5022294521331787,
      "learning_rate": 2.56596578868192e-05,
      "loss": 0.0935,
      "num_input_tokens_seen": 37273072,
      "step": 19665
    },
    {
      "epoch": 3.2088261685292436,
      "grad_norm": 1.3715049028396606,
      "learning_rate": 2.564984377740883e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 37282768,
      "step": 19670
    },
    {
      "epoch": 3.209641895750061,
      "grad_norm": 0.4117656648159027,
      "learning_rate": 2.564002956778438e-05,
      "loss": 0.0822,
      "num_input_tokens_seen": 37291248,
      "step": 19675
    },
    {
      "epoch": 3.2104576229708783,
      "grad_norm": 0.5937821269035339,
      "learning_rate": 2.563021525945934e-05,
      "loss": 0.1022,
      "num_input_tokens_seen": 37302016,
      "step": 19680
    },
    {
      "epoch": 3.211273350191696,
      "grad_norm": 1.3157007694244385,
      "learning_rate": 2.562040085394718e-05,
      "loss": 0.237,
      "num_input_tokens_seen": 37310880,
      "step": 19685
    },
    {
      "epoch": 3.2120890774125135,
      "grad_norm": 1.9850025177001953,
      "learning_rate": 2.56105863527614e-05,
      "loss": 0.1239,
      "num_input_tokens_seen": 37321952,
      "step": 19690
    },
    {
      "epoch": 3.2129048046333306,
      "grad_norm": 0.30831727385520935,
      "learning_rate": 2.5600771757415548e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 37331072,
      "step": 19695
    },
    {
      "epoch": 3.2137205318541477,
      "grad_norm": 0.3848450183868408,
      "learning_rate": 2.5590957069423134e-05,
      "loss": 0.0812,
      "num_input_tokens_seen": 37339696,
      "step": 19700
    },
    {
      "epoch": 3.2145362590749653,
      "grad_norm": 2.3364999294281006,
      "learning_rate": 2.5581142290297716e-05,
      "loss": 0.184,
      "num_input_tokens_seen": 37349088,
      "step": 19705
    },
    {
      "epoch": 3.215351986295783,
      "grad_norm": 0.3094170093536377,
      "learning_rate": 2.557132742155285e-05,
      "loss": 0.1989,
      "num_input_tokens_seen": 37359360,
      "step": 19710
    },
    {
      "epoch": 3.2161677135166,
      "grad_norm": 1.2463656663894653,
      "learning_rate": 2.556151246470212e-05,
      "loss": 0.1334,
      "num_input_tokens_seen": 37368688,
      "step": 19715
    },
    {
      "epoch": 3.2169834407374176,
      "grad_norm": 0.11735288798809052,
      "learning_rate": 2.5551697421259114e-05,
      "loss": 0.0792,
      "num_input_tokens_seen": 37378912,
      "step": 19720
    },
    {
      "epoch": 3.2177991679582347,
      "grad_norm": 0.1086508110165596,
      "learning_rate": 2.554188229273743e-05,
      "loss": 0.0384,
      "num_input_tokens_seen": 37388512,
      "step": 19725
    },
    {
      "epoch": 3.2186148951790523,
      "grad_norm": 1.156212568283081,
      "learning_rate": 2.5532067080650678e-05,
      "loss": 0.1677,
      "num_input_tokens_seen": 37398368,
      "step": 19730
    },
    {
      "epoch": 3.2194306223998694,
      "grad_norm": 0.7050184607505798,
      "learning_rate": 2.55222517865125e-05,
      "loss": 0.0619,
      "num_input_tokens_seen": 37407968,
      "step": 19735
    },
    {
      "epoch": 3.220246349620687,
      "grad_norm": 0.05970483645796776,
      "learning_rate": 2.5512436411836538e-05,
      "loss": 0.0753,
      "num_input_tokens_seen": 37418912,
      "step": 19740
    },
    {
      "epoch": 3.221062076841504,
      "grad_norm": 1.368706464767456,
      "learning_rate": 2.5502620958136443e-05,
      "loss": 0.1273,
      "num_input_tokens_seen": 37429264,
      "step": 19745
    },
    {
      "epoch": 3.2218778040623217,
      "grad_norm": 0.14707812666893005,
      "learning_rate": 2.5492805426925874e-05,
      "loss": 0.0669,
      "num_input_tokens_seen": 37438960,
      "step": 19750
    },
    {
      "epoch": 3.222693531283139,
      "grad_norm": 0.2895858883857727,
      "learning_rate": 2.5482989819718523e-05,
      "loss": 0.1339,
      "num_input_tokens_seen": 37448320,
      "step": 19755
    },
    {
      "epoch": 3.2235092585039564,
      "grad_norm": 1.006226658821106,
      "learning_rate": 2.5473174138028065e-05,
      "loss": 0.2152,
      "num_input_tokens_seen": 37457456,
      "step": 19760
    },
    {
      "epoch": 3.2243249857247736,
      "grad_norm": 0.16464906930923462,
      "learning_rate": 2.5463358383368212e-05,
      "loss": 0.0311,
      "num_input_tokens_seen": 37465968,
      "step": 19765
    },
    {
      "epoch": 3.225140712945591,
      "grad_norm": 0.43418946862220764,
      "learning_rate": 2.545354255725267e-05,
      "loss": 0.1245,
      "num_input_tokens_seen": 37473840,
      "step": 19770
    },
    {
      "epoch": 3.2259564401664083,
      "grad_norm": 0.4384429156780243,
      "learning_rate": 2.5443726661195165e-05,
      "loss": 0.116,
      "num_input_tokens_seen": 37483344,
      "step": 19775
    },
    {
      "epoch": 3.226772167387226,
      "grad_norm": 0.9032662510871887,
      "learning_rate": 2.543391069670944e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 37493888,
      "step": 19780
    },
    {
      "epoch": 3.227587894608043,
      "grad_norm": 0.9075506925582886,
      "learning_rate": 2.5424094665309228e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 37503360,
      "step": 19785
    },
    {
      "epoch": 3.2284036218288605,
      "grad_norm": 0.9928440451622009,
      "learning_rate": 2.5414278568508292e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 37513024,
      "step": 19790
    },
    {
      "epoch": 3.2292193490496777,
      "grad_norm": 0.08369394391775131,
      "learning_rate": 2.540446240782039e-05,
      "loss": 0.1163,
      "num_input_tokens_seen": 37521984,
      "step": 19795
    },
    {
      "epoch": 3.2300350762704952,
      "grad_norm": 0.8971673250198364,
      "learning_rate": 2.5394646184759307e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 37532496,
      "step": 19800
    },
    {
      "epoch": 3.2300350762704952,
      "eval_loss": 0.14733749628067017,
      "eval_runtime": 133.2639,
      "eval_samples_per_second": 20.448,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 37532496,
      "step": 19800
    },
    {
      "epoch": 3.2308508034913124,
      "grad_norm": 0.1819232702255249,
      "learning_rate": 2.538482990083882e-05,
      "loss": 0.0389,
      "num_input_tokens_seen": 37540976,
      "step": 19805
    },
    {
      "epoch": 3.23166653071213,
      "grad_norm": 0.14723564684391022,
      "learning_rate": 2.5375013557572725e-05,
      "loss": 0.0227,
      "num_input_tokens_seen": 37550304,
      "step": 19810
    },
    {
      "epoch": 3.232482257932947,
      "grad_norm": 1.3637951612472534,
      "learning_rate": 2.536519715647483e-05,
      "loss": 0.1949,
      "num_input_tokens_seen": 37559600,
      "step": 19815
    },
    {
      "epoch": 3.2332979851537647,
      "grad_norm": 0.8953526020050049,
      "learning_rate": 2.535538069905894e-05,
      "loss": 0.1894,
      "num_input_tokens_seen": 37568576,
      "step": 19820
    },
    {
      "epoch": 3.234113712374582,
      "grad_norm": 1.4107459783554077,
      "learning_rate": 2.534556418683888e-05,
      "loss": 0.1334,
      "num_input_tokens_seen": 37577792,
      "step": 19825
    },
    {
      "epoch": 3.2349294395953994,
      "grad_norm": 0.02356635220348835,
      "learning_rate": 2.5335747621328486e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 37587360,
      "step": 19830
    },
    {
      "epoch": 3.2357451668162165,
      "grad_norm": 0.4127614498138428,
      "learning_rate": 2.5325931004041586e-05,
      "loss": 0.0892,
      "num_input_tokens_seen": 37596816,
      "step": 19835
    },
    {
      "epoch": 3.236560894037034,
      "grad_norm": 0.7127164602279663,
      "learning_rate": 2.5316114336492032e-05,
      "loss": 0.0694,
      "num_input_tokens_seen": 37606928,
      "step": 19840
    },
    {
      "epoch": 3.237376621257851,
      "grad_norm": 0.7028880715370178,
      "learning_rate": 2.530629762019367e-05,
      "loss": 0.2928,
      "num_input_tokens_seen": 37616976,
      "step": 19845
    },
    {
      "epoch": 3.2381923484786688,
      "grad_norm": 0.438127726316452,
      "learning_rate": 2.5296480856660364e-05,
      "loss": 0.0659,
      "num_input_tokens_seen": 37626912,
      "step": 19850
    },
    {
      "epoch": 3.239008075699486,
      "grad_norm": 0.12222267687320709,
      "learning_rate": 2.528666404740599e-05,
      "loss": 0.2488,
      "num_input_tokens_seen": 37636256,
      "step": 19855
    },
    {
      "epoch": 3.2398238029203035,
      "grad_norm": 0.3365391492843628,
      "learning_rate": 2.527684719394442e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 37644048,
      "step": 19860
    },
    {
      "epoch": 3.2406395301411206,
      "grad_norm": 1.8247357606887817,
      "learning_rate": 2.526703029778953e-05,
      "loss": 0.2609,
      "num_input_tokens_seen": 37654064,
      "step": 19865
    },
    {
      "epoch": 3.241455257361938,
      "grad_norm": 0.263907790184021,
      "learning_rate": 2.5257213360455208e-05,
      "loss": 0.0569,
      "num_input_tokens_seen": 37663808,
      "step": 19870
    },
    {
      "epoch": 3.2422709845827553,
      "grad_norm": 1.752128005027771,
      "learning_rate": 2.5247396383455353e-05,
      "loss": 0.1913,
      "num_input_tokens_seen": 37673888,
      "step": 19875
    },
    {
      "epoch": 3.243086711803573,
      "grad_norm": 1.4236655235290527,
      "learning_rate": 2.523757936830387e-05,
      "loss": 0.2508,
      "num_input_tokens_seen": 37683312,
      "step": 19880
    },
    {
      "epoch": 3.2439024390243905,
      "grad_norm": 2.3283135890960693,
      "learning_rate": 2.5227762316514662e-05,
      "loss": 0.2654,
      "num_input_tokens_seen": 37693696,
      "step": 19885
    },
    {
      "epoch": 3.2447181662452076,
      "grad_norm": 0.2534615993499756,
      "learning_rate": 2.5217945229601648e-05,
      "loss": 0.2603,
      "num_input_tokens_seen": 37702992,
      "step": 19890
    },
    {
      "epoch": 3.245533893466025,
      "grad_norm": 0.025732316076755524,
      "learning_rate": 2.5208128109078738e-05,
      "loss": 0.0506,
      "num_input_tokens_seen": 37712736,
      "step": 19895
    },
    {
      "epoch": 3.2463496206868423,
      "grad_norm": 0.14991268515586853,
      "learning_rate": 2.5198310956459853e-05,
      "loss": 0.1853,
      "num_input_tokens_seen": 37722832,
      "step": 19900
    },
    {
      "epoch": 3.24716534790766,
      "grad_norm": 0.45932474732398987,
      "learning_rate": 2.518849377325893e-05,
      "loss": 0.0859,
      "num_input_tokens_seen": 37731584,
      "step": 19905
    },
    {
      "epoch": 3.247981075128477,
      "grad_norm": 1.2296979427337646,
      "learning_rate": 2.51786765609899e-05,
      "loss": 0.1826,
      "num_input_tokens_seen": 37741808,
      "step": 19910
    },
    {
      "epoch": 3.2487968023492946,
      "grad_norm": 0.8546766042709351,
      "learning_rate": 2.5168859321166694e-05,
      "loss": 0.2337,
      "num_input_tokens_seen": 37751056,
      "step": 19915
    },
    {
      "epoch": 3.2496125295701117,
      "grad_norm": 0.1697927862405777,
      "learning_rate": 2.515904205530326e-05,
      "loss": 0.2173,
      "num_input_tokens_seen": 37760768,
      "step": 19920
    },
    {
      "epoch": 3.2504282567909293,
      "grad_norm": 0.45016565918922424,
      "learning_rate": 2.514922476491355e-05,
      "loss": 0.046,
      "num_input_tokens_seen": 37769632,
      "step": 19925
    },
    {
      "epoch": 3.2512439840117464,
      "grad_norm": 0.058526281267404556,
      "learning_rate": 2.51394074515115e-05,
      "loss": 0.0853,
      "num_input_tokens_seen": 37779424,
      "step": 19930
    },
    {
      "epoch": 3.252059711232564,
      "grad_norm": 0.18957491219043732,
      "learning_rate": 2.5129590116611067e-05,
      "loss": 0.2111,
      "num_input_tokens_seen": 37789520,
      "step": 19935
    },
    {
      "epoch": 3.252875438453381,
      "grad_norm": 0.8152214288711548,
      "learning_rate": 2.5119772761726212e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 37799104,
      "step": 19940
    },
    {
      "epoch": 3.2536911656741987,
      "grad_norm": 0.2690587341785431,
      "learning_rate": 2.5109955388370893e-05,
      "loss": 0.1455,
      "num_input_tokens_seen": 37809152,
      "step": 19945
    },
    {
      "epoch": 3.254506892895016,
      "grad_norm": 1.6786010265350342,
      "learning_rate": 2.510013799805907e-05,
      "loss": 0.1843,
      "num_input_tokens_seen": 37817376,
      "step": 19950
    },
    {
      "epoch": 3.2553226201158334,
      "grad_norm": 0.6524873971939087,
      "learning_rate": 2.5090320592304706e-05,
      "loss": 0.079,
      "num_input_tokens_seen": 37825904,
      "step": 19955
    },
    {
      "epoch": 3.2561383473366505,
      "grad_norm": 0.4850170910358429,
      "learning_rate": 2.5080503172621777e-05,
      "loss": 0.0723,
      "num_input_tokens_seen": 37834928,
      "step": 19960
    },
    {
      "epoch": 3.256954074557468,
      "grad_norm": 0.10858715325593948,
      "learning_rate": 2.5070685740524246e-05,
      "loss": 0.2523,
      "num_input_tokens_seen": 37845552,
      "step": 19965
    },
    {
      "epoch": 3.2577698017782852,
      "grad_norm": 1.6013402938842773,
      "learning_rate": 2.5060868297526084e-05,
      "loss": 0.1807,
      "num_input_tokens_seen": 37854688,
      "step": 19970
    },
    {
      "epoch": 3.258585528999103,
      "grad_norm": 1.5282036066055298,
      "learning_rate": 2.5051050845141267e-05,
      "loss": 0.0736,
      "num_input_tokens_seen": 37864880,
      "step": 19975
    },
    {
      "epoch": 3.25940125621992,
      "grad_norm": 0.051290322095155716,
      "learning_rate": 2.5041233384883765e-05,
      "loss": 0.1938,
      "num_input_tokens_seen": 37873488,
      "step": 19980
    },
    {
      "epoch": 3.2602169834407375,
      "grad_norm": 0.885174572467804,
      "learning_rate": 2.5031415918267564e-05,
      "loss": 0.214,
      "num_input_tokens_seen": 37882240,
      "step": 19985
    },
    {
      "epoch": 3.2610327106615546,
      "grad_norm": 0.45332980155944824,
      "learning_rate": 2.5021598446806626e-05,
      "loss": 0.0998,
      "num_input_tokens_seen": 37892592,
      "step": 19990
    },
    {
      "epoch": 3.261848437882372,
      "grad_norm": 1.459374189376831,
      "learning_rate": 2.5011780972014937e-05,
      "loss": 0.2257,
      "num_input_tokens_seen": 37900656,
      "step": 19995
    },
    {
      "epoch": 3.2626641651031894,
      "grad_norm": 1.779240369796753,
      "learning_rate": 2.5001963495406478e-05,
      "loss": 0.2611,
      "num_input_tokens_seen": 37910480,
      "step": 20000
    },
    {
      "epoch": 3.2626641651031894,
      "eval_loss": 0.14190852642059326,
      "eval_runtime": 133.4041,
      "eval_samples_per_second": 20.427,
      "eval_steps_per_second": 10.217,
      "num_input_tokens_seen": 37910480,
      "step": 20000
    },
    {
      "epoch": 3.263479892324007,
      "grad_norm": 1.4543657302856445,
      "learning_rate": 2.499214601849522e-05,
      "loss": 0.2023,
      "num_input_tokens_seen": 37919664,
      "step": 20005
    },
    {
      "epoch": 3.264295619544824,
      "grad_norm": 0.4154205620288849,
      "learning_rate": 2.4982328542795148e-05,
      "loss": 0.2822,
      "num_input_tokens_seen": 37928784,
      "step": 20010
    },
    {
      "epoch": 3.2651113467656416,
      "grad_norm": 0.14695873856544495,
      "learning_rate": 2.497251106982024e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 37938960,
      "step": 20015
    },
    {
      "epoch": 3.2659270739864588,
      "grad_norm": 1.0474299192428589,
      "learning_rate": 2.4962693601084458e-05,
      "loss": 0.1469,
      "num_input_tokens_seen": 37947216,
      "step": 20020
    },
    {
      "epoch": 3.2667428012072763,
      "grad_norm": 1.3464754819869995,
      "learning_rate": 2.4952876138101794e-05,
      "loss": 0.2185,
      "num_input_tokens_seen": 37956576,
      "step": 20025
    },
    {
      "epoch": 3.2675585284280935,
      "grad_norm": 0.5059743523597717,
      "learning_rate": 2.4943058682386233e-05,
      "loss": 0.0876,
      "num_input_tokens_seen": 37966128,
      "step": 20030
    },
    {
      "epoch": 3.268374255648911,
      "grad_norm": 1.1985548734664917,
      "learning_rate": 2.493324123545173e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 37975536,
      "step": 20035
    },
    {
      "epoch": 3.269189982869728,
      "grad_norm": 1.3604146242141724,
      "learning_rate": 2.4923423798812272e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 37986256,
      "step": 20040
    },
    {
      "epoch": 3.2700057100905457,
      "grad_norm": 0.8075903654098511,
      "learning_rate": 2.4913606373981825e-05,
      "loss": 0.2372,
      "num_input_tokens_seen": 37996368,
      "step": 20045
    },
    {
      "epoch": 3.270821437311363,
      "grad_norm": 0.39955034852027893,
      "learning_rate": 2.4903788962474357e-05,
      "loss": 0.0434,
      "num_input_tokens_seen": 38004688,
      "step": 20050
    },
    {
      "epoch": 3.2716371645321805,
      "grad_norm": 0.06523960828781128,
      "learning_rate": 2.489397156580385e-05,
      "loss": 0.0313,
      "num_input_tokens_seen": 38012672,
      "step": 20055
    },
    {
      "epoch": 3.272452891752998,
      "grad_norm": 0.4197540879249573,
      "learning_rate": 2.4884154185484246e-05,
      "loss": 0.0691,
      "num_input_tokens_seen": 38022016,
      "step": 20060
    },
    {
      "epoch": 3.273268618973815,
      "grad_norm": 0.5295142531394958,
      "learning_rate": 2.4874336823029526e-05,
      "loss": 0.0657,
      "num_input_tokens_seen": 38031328,
      "step": 20065
    },
    {
      "epoch": 3.2740843461946323,
      "grad_norm": 0.9143192768096924,
      "learning_rate": 2.4864519479953656e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 38040224,
      "step": 20070
    },
    {
      "epoch": 3.27490007341545,
      "grad_norm": 0.13029636442661285,
      "learning_rate": 2.485470215777058e-05,
      "loss": 0.1271,
      "num_input_tokens_seen": 38048400,
      "step": 20075
    },
    {
      "epoch": 3.2757158006362674,
      "grad_norm": 0.9369627237319946,
      "learning_rate": 2.4844884857994258e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 38058672,
      "step": 20080
    },
    {
      "epoch": 3.2765315278570846,
      "grad_norm": 0.6556028723716736,
      "learning_rate": 2.4835067582138638e-05,
      "loss": 0.1188,
      "num_input_tokens_seen": 38068688,
      "step": 20085
    },
    {
      "epoch": 3.2773472550779017,
      "grad_norm": 1.4158687591552734,
      "learning_rate": 2.4825250331717666e-05,
      "loss": 0.2943,
      "num_input_tokens_seen": 38078304,
      "step": 20090
    },
    {
      "epoch": 3.2781629822987193,
      "grad_norm": 0.502221941947937,
      "learning_rate": 2.4815433108245298e-05,
      "loss": 0.2521,
      "num_input_tokens_seen": 38088688,
      "step": 20095
    },
    {
      "epoch": 3.278978709519537,
      "grad_norm": 0.22112742066383362,
      "learning_rate": 2.4805615913235456e-05,
      "loss": 0.1133,
      "num_input_tokens_seen": 38097680,
      "step": 20100
    },
    {
      "epoch": 3.279794436740354,
      "grad_norm": 0.5699601173400879,
      "learning_rate": 2.479579874820208e-05,
      "loss": 0.193,
      "num_input_tokens_seen": 38108208,
      "step": 20105
    },
    {
      "epoch": 3.2806101639611716,
      "grad_norm": 1.75919508934021,
      "learning_rate": 2.4785981614659115e-05,
      "loss": 0.0582,
      "num_input_tokens_seen": 38117376,
      "step": 20110
    },
    {
      "epoch": 3.2814258911819887,
      "grad_norm": 0.3135209083557129,
      "learning_rate": 2.477616451412047e-05,
      "loss": 0.0707,
      "num_input_tokens_seen": 38127728,
      "step": 20115
    },
    {
      "epoch": 3.2822416184028063,
      "grad_norm": 0.12358581274747849,
      "learning_rate": 2.476634744810007e-05,
      "loss": 0.1185,
      "num_input_tokens_seen": 38136416,
      "step": 20120
    },
    {
      "epoch": 3.2830573456236234,
      "grad_norm": 0.575274646282196,
      "learning_rate": 2.475653041811183e-05,
      "loss": 0.076,
      "num_input_tokens_seen": 38146000,
      "step": 20125
    },
    {
      "epoch": 3.283873072844441,
      "grad_norm": 0.5258607864379883,
      "learning_rate": 2.4746713425669652e-05,
      "loss": 0.0502,
      "num_input_tokens_seen": 38156176,
      "step": 20130
    },
    {
      "epoch": 3.284688800065258,
      "grad_norm": 0.11109481751918793,
      "learning_rate": 2.4736896472287458e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 38164848,
      "step": 20135
    },
    {
      "epoch": 3.2855045272860757,
      "grad_norm": 0.30587878823280334,
      "learning_rate": 2.4727079559479124e-05,
      "loss": 0.0465,
      "num_input_tokens_seen": 38173792,
      "step": 20140
    },
    {
      "epoch": 3.286320254506893,
      "grad_norm": 0.22161798179149628,
      "learning_rate": 2.4717262688758557e-05,
      "loss": 0.2115,
      "num_input_tokens_seen": 38183280,
      "step": 20145
    },
    {
      "epoch": 3.2871359817277104,
      "grad_norm": 0.43665724992752075,
      "learning_rate": 2.4707445861639637e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 38192128,
      "step": 20150
    },
    {
      "epoch": 3.2879517089485275,
      "grad_norm": 0.722845196723938,
      "learning_rate": 2.4697629079636244e-05,
      "loss": 0.1299,
      "num_input_tokens_seen": 38201072,
      "step": 20155
    },
    {
      "epoch": 3.288767436169345,
      "grad_norm": 0.7870884537696838,
      "learning_rate": 2.4687812344262244e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 38210208,
      "step": 20160
    },
    {
      "epoch": 3.289583163390162,
      "grad_norm": 2.089992046356201,
      "learning_rate": 2.46779956570315e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 38219008,
      "step": 20165
    },
    {
      "epoch": 3.29039889061098,
      "grad_norm": 0.762401282787323,
      "learning_rate": 2.466817901945787e-05,
      "loss": 0.0858,
      "num_input_tokens_seen": 38228512,
      "step": 20170
    },
    {
      "epoch": 3.291214617831797,
      "grad_norm": 0.4203280806541443,
      "learning_rate": 2.4658362433055217e-05,
      "loss": 0.1629,
      "num_input_tokens_seen": 38237872,
      "step": 20175
    },
    {
      "epoch": 3.2920303450526145,
      "grad_norm": 1.360281229019165,
      "learning_rate": 2.4648545899337356e-05,
      "loss": 0.2915,
      "num_input_tokens_seen": 38246896,
      "step": 20180
    },
    {
      "epoch": 3.2928460722734316,
      "grad_norm": 0.5677017569541931,
      "learning_rate": 2.4638729419818143e-05,
      "loss": 0.0943,
      "num_input_tokens_seen": 38256208,
      "step": 20185
    },
    {
      "epoch": 3.293661799494249,
      "grad_norm": 0.15761838853359222,
      "learning_rate": 2.46289129960114e-05,
      "loss": 0.1197,
      "num_input_tokens_seen": 38266448,
      "step": 20190
    },
    {
      "epoch": 3.2944775267150663,
      "grad_norm": 1.2737219333648682,
      "learning_rate": 2.4619096629430924e-05,
      "loss": 0.1811,
      "num_input_tokens_seen": 38277136,
      "step": 20195
    },
    {
      "epoch": 3.295293253935884,
      "grad_norm": 0.5720317363739014,
      "learning_rate": 2.4609280321590543e-05,
      "loss": 0.2644,
      "num_input_tokens_seen": 38286080,
      "step": 20200
    },
    {
      "epoch": 3.295293253935884,
      "eval_loss": 0.14224253594875336,
      "eval_runtime": 133.2954,
      "eval_samples_per_second": 20.443,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 38286080,
      "step": 20200
    },
    {
      "epoch": 3.296108981156701,
      "grad_norm": 0.6728852987289429,
      "learning_rate": 2.4599464074004037e-05,
      "loss": 0.1022,
      "num_input_tokens_seen": 38296064,
      "step": 20205
    },
    {
      "epoch": 3.2969247083775186,
      "grad_norm": 1.1100995540618896,
      "learning_rate": 2.4589647888185204e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 38305104,
      "step": 20210
    },
    {
      "epoch": 3.2977404355983357,
      "grad_norm": 0.300886869430542,
      "learning_rate": 2.4579831765647836e-05,
      "loss": 0.1212,
      "num_input_tokens_seen": 38315136,
      "step": 20215
    },
    {
      "epoch": 3.2985561628191533,
      "grad_norm": 0.5140460729598999,
      "learning_rate": 2.4570015707905676e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 38323280,
      "step": 20220
    },
    {
      "epoch": 3.2993718900399704,
      "grad_norm": 0.5830605030059814,
      "learning_rate": 2.4560199716472508e-05,
      "loss": 0.1891,
      "num_input_tokens_seen": 38332528,
      "step": 20225
    },
    {
      "epoch": 3.300187617260788,
      "grad_norm": 0.4281441867351532,
      "learning_rate": 2.455038379286207e-05,
      "loss": 0.1641,
      "num_input_tokens_seen": 38343216,
      "step": 20230
    },
    {
      "epoch": 3.3010033444816056,
      "grad_norm": 1.0035243034362793,
      "learning_rate": 2.4540567938588095e-05,
      "loss": 0.0885,
      "num_input_tokens_seen": 38353856,
      "step": 20235
    },
    {
      "epoch": 3.3018190717024227,
      "grad_norm": 0.42890673875808716,
      "learning_rate": 2.4530752155164328e-05,
      "loss": 0.1699,
      "num_input_tokens_seen": 38363696,
      "step": 20240
    },
    {
      "epoch": 3.30263479892324,
      "grad_norm": 0.10351426899433136,
      "learning_rate": 2.4520936444104463e-05,
      "loss": 0.1234,
      "num_input_tokens_seen": 38372832,
      "step": 20245
    },
    {
      "epoch": 3.3034505261440574,
      "grad_norm": 0.2469605952501297,
      "learning_rate": 2.4511120806922218e-05,
      "loss": 0.2112,
      "num_input_tokens_seen": 38383424,
      "step": 20250
    },
    {
      "epoch": 3.304266253364875,
      "grad_norm": 0.09660451114177704,
      "learning_rate": 2.45013052451313e-05,
      "loss": 0.0489,
      "num_input_tokens_seen": 38393936,
      "step": 20255
    },
    {
      "epoch": 3.305081980585692,
      "grad_norm": 0.13229715824127197,
      "learning_rate": 2.4491489760245376e-05,
      "loss": 0.2056,
      "num_input_tokens_seen": 38402224,
      "step": 20260
    },
    {
      "epoch": 3.3058977078065093,
      "grad_norm": 1.4251164197921753,
      "learning_rate": 2.4481674353778115e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 38411280,
      "step": 20265
    },
    {
      "epoch": 3.306713435027327,
      "grad_norm": 1.8620177507400513,
      "learning_rate": 2.447185902724319e-05,
      "loss": 0.2189,
      "num_input_tokens_seen": 38420384,
      "step": 20270
    },
    {
      "epoch": 3.3075291622481444,
      "grad_norm": 0.1567995846271515,
      "learning_rate": 2.4462043782154233e-05,
      "loss": 0.0633,
      "num_input_tokens_seen": 38429920,
      "step": 20275
    },
    {
      "epoch": 3.3083448894689615,
      "grad_norm": 0.46787405014038086,
      "learning_rate": 2.4452228620024895e-05,
      "loss": 0.2051,
      "num_input_tokens_seen": 38439440,
      "step": 20280
    },
    {
      "epoch": 3.309160616689779,
      "grad_norm": 0.4570968747138977,
      "learning_rate": 2.4442413542368776e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 38448416,
      "step": 20285
    },
    {
      "epoch": 3.3099763439105963,
      "grad_norm": 0.0769435316324234,
      "learning_rate": 2.4432598550699502e-05,
      "loss": 0.0346,
      "num_input_tokens_seen": 38458592,
      "step": 20290
    },
    {
      "epoch": 3.310792071131414,
      "grad_norm": 0.9092785716056824,
      "learning_rate": 2.4422783646530663e-05,
      "loss": 0.0873,
      "num_input_tokens_seen": 38468144,
      "step": 20295
    },
    {
      "epoch": 3.311607798352231,
      "grad_norm": 1.1656519174575806,
      "learning_rate": 2.441296883137584e-05,
      "loss": 0.2368,
      "num_input_tokens_seen": 38478352,
      "step": 20300
    },
    {
      "epoch": 3.3124235255730485,
      "grad_norm": 0.2599679231643677,
      "learning_rate": 2.4403154106748592e-05,
      "loss": 0.078,
      "num_input_tokens_seen": 38487072,
      "step": 20305
    },
    {
      "epoch": 3.3132392527938657,
      "grad_norm": 0.384034126996994,
      "learning_rate": 2.4393339474162494e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 38496224,
      "step": 20310
    },
    {
      "epoch": 3.3140549800146832,
      "grad_norm": 1.151553750038147,
      "learning_rate": 2.4383524935131062e-05,
      "loss": 0.107,
      "num_input_tokens_seen": 38505616,
      "step": 20315
    },
    {
      "epoch": 3.3148707072355004,
      "grad_norm": 0.7223048806190491,
      "learning_rate": 2.437371049116784e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 38515456,
      "step": 20320
    },
    {
      "epoch": 3.315686434456318,
      "grad_norm": 0.2500169277191162,
      "learning_rate": 2.436389614378632e-05,
      "loss": 0.0698,
      "num_input_tokens_seen": 38525408,
      "step": 20325
    },
    {
      "epoch": 3.316502161677135,
      "grad_norm": 2.3102896213531494,
      "learning_rate": 2.435408189450002e-05,
      "loss": 0.1698,
      "num_input_tokens_seen": 38534720,
      "step": 20330
    },
    {
      "epoch": 3.3173178888979526,
      "grad_norm": 0.05743793770670891,
      "learning_rate": 2.4344267744822406e-05,
      "loss": 0.0659,
      "num_input_tokens_seen": 38544464,
      "step": 20335
    },
    {
      "epoch": 3.31813361611877,
      "grad_norm": 0.5825542211532593,
      "learning_rate": 2.4334453696266944e-05,
      "loss": 0.1959,
      "num_input_tokens_seen": 38553824,
      "step": 20340
    },
    {
      "epoch": 3.3189493433395874,
      "grad_norm": 0.48234492540359497,
      "learning_rate": 2.432463975034708e-05,
      "loss": 0.1012,
      "num_input_tokens_seen": 38563488,
      "step": 20345
    },
    {
      "epoch": 3.3197650705604045,
      "grad_norm": 0.10788609087467194,
      "learning_rate": 2.4314825908576265e-05,
      "loss": 0.2959,
      "num_input_tokens_seen": 38572400,
      "step": 20350
    },
    {
      "epoch": 3.320580797781222,
      "grad_norm": 0.5157867670059204,
      "learning_rate": 2.4305012172467897e-05,
      "loss": 0.0964,
      "num_input_tokens_seen": 38581888,
      "step": 20355
    },
    {
      "epoch": 3.321396525002039,
      "grad_norm": 1.2382737398147583,
      "learning_rate": 2.4295198543535393e-05,
      "loss": 0.2339,
      "num_input_tokens_seen": 38591936,
      "step": 20360
    },
    {
      "epoch": 3.3222122522228568,
      "grad_norm": 0.25977325439453125,
      "learning_rate": 2.4285385023292124e-05,
      "loss": 0.0195,
      "num_input_tokens_seen": 38600624,
      "step": 20365
    },
    {
      "epoch": 3.323027979443674,
      "grad_norm": 0.15317776799201965,
      "learning_rate": 2.427557161325147e-05,
      "loss": 0.1894,
      "num_input_tokens_seen": 38610080,
      "step": 20370
    },
    {
      "epoch": 3.3238437066644915,
      "grad_norm": 0.2902660071849823,
      "learning_rate": 2.4265758314926778e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 38620480,
      "step": 20375
    },
    {
      "epoch": 3.3246594338853086,
      "grad_norm": 0.44393372535705566,
      "learning_rate": 2.4255945129831373e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 38629744,
      "step": 20380
    },
    {
      "epoch": 3.325475161106126,
      "grad_norm": 0.6406533122062683,
      "learning_rate": 2.4246132059478578e-05,
      "loss": 0.0924,
      "num_input_tokens_seen": 38639264,
      "step": 20385
    },
    {
      "epoch": 3.3262908883269433,
      "grad_norm": 0.36409270763397217,
      "learning_rate": 2.4236319105381706e-05,
      "loss": 0.0516,
      "num_input_tokens_seen": 38646976,
      "step": 20390
    },
    {
      "epoch": 3.327106615547761,
      "grad_norm": 0.2247934490442276,
      "learning_rate": 2.422650626905401e-05,
      "loss": 0.1184,
      "num_input_tokens_seen": 38655984,
      "step": 20395
    },
    {
      "epoch": 3.327922342768578,
      "grad_norm": 0.11516903340816498,
      "learning_rate": 2.4216693552008785e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 38664512,
      "step": 20400
    },
    {
      "epoch": 3.327922342768578,
      "eval_loss": 0.14266490936279297,
      "eval_runtime": 133.2557,
      "eval_samples_per_second": 20.449,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 38664512,
      "step": 20400
    },
    {
      "epoch": 3.3287380699893956,
      "grad_norm": 0.454201877117157,
      "learning_rate": 2.4206880955759247e-05,
      "loss": 0.1728,
      "num_input_tokens_seen": 38674416,
      "step": 20405
    },
    {
      "epoch": 3.3295537972102127,
      "grad_norm": 2.0270166397094727,
      "learning_rate": 2.419706848181863e-05,
      "loss": 0.2673,
      "num_input_tokens_seen": 38684240,
      "step": 20410
    },
    {
      "epoch": 3.3303695244310303,
      "grad_norm": 0.23007434606552124,
      "learning_rate": 2.4187256131700153e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 38693552,
      "step": 20415
    },
    {
      "epoch": 3.3311852516518474,
      "grad_norm": 0.16217833757400513,
      "learning_rate": 2.4177443906916985e-05,
      "loss": 0.2156,
      "num_input_tokens_seen": 38703648,
      "step": 20420
    },
    {
      "epoch": 3.332000978872665,
      "grad_norm": 0.18038219213485718,
      "learning_rate": 2.4167631808982303e-05,
      "loss": 0.1648,
      "num_input_tokens_seen": 38713408,
      "step": 20425
    },
    {
      "epoch": 3.3328167060934826,
      "grad_norm": 0.28350725769996643,
      "learning_rate": 2.4157819839409264e-05,
      "loss": 0.0214,
      "num_input_tokens_seen": 38723520,
      "step": 20430
    },
    {
      "epoch": 3.3336324333142997,
      "grad_norm": 0.596277117729187,
      "learning_rate": 2.414800799971098e-05,
      "loss": 0.0428,
      "num_input_tokens_seen": 38732736,
      "step": 20435
    },
    {
      "epoch": 3.334448160535117,
      "grad_norm": 0.2490668147802353,
      "learning_rate": 2.4138196291400582e-05,
      "loss": 0.1071,
      "num_input_tokens_seen": 38742816,
      "step": 20440
    },
    {
      "epoch": 3.3352638877559344,
      "grad_norm": 0.26333969831466675,
      "learning_rate": 2.412838471599114e-05,
      "loss": 0.1018,
      "num_input_tokens_seen": 38752336,
      "step": 20445
    },
    {
      "epoch": 3.336079614976752,
      "grad_norm": 1.2500050067901611,
      "learning_rate": 2.411857327499572e-05,
      "loss": 0.0758,
      "num_input_tokens_seen": 38760960,
      "step": 20450
    },
    {
      "epoch": 3.336895342197569,
      "grad_norm": 0.08369175344705582,
      "learning_rate": 2.410876196992739e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 38770592,
      "step": 20455
    },
    {
      "epoch": 3.3377110694183862,
      "grad_norm": 0.14098693430423737,
      "learning_rate": 2.4098950802299156e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 38780624,
      "step": 20460
    },
    {
      "epoch": 3.338526796639204,
      "grad_norm": 0.9598837494850159,
      "learning_rate": 2.4089139773624027e-05,
      "loss": 0.2054,
      "num_input_tokens_seen": 38791152,
      "step": 20465
    },
    {
      "epoch": 3.3393425238600214,
      "grad_norm": 0.25701144337654114,
      "learning_rate": 2.4079328885415007e-05,
      "loss": 0.1108,
      "num_input_tokens_seen": 38800016,
      "step": 20470
    },
    {
      "epoch": 3.3401582510808385,
      "grad_norm": 0.6597967147827148,
      "learning_rate": 2.4069518139185036e-05,
      "loss": 0.2278,
      "num_input_tokens_seen": 38809712,
      "step": 20475
    },
    {
      "epoch": 3.340973978301656,
      "grad_norm": 1.416171669960022,
      "learning_rate": 2.405970753644706e-05,
      "loss": 0.2492,
      "num_input_tokens_seen": 38819456,
      "step": 20480
    },
    {
      "epoch": 3.3417897055224732,
      "grad_norm": 0.595222532749176,
      "learning_rate": 2.4049897078714e-05,
      "loss": 0.1086,
      "num_input_tokens_seen": 38828992,
      "step": 20485
    },
    {
      "epoch": 3.342605432743291,
      "grad_norm": 0.34276074171066284,
      "learning_rate": 2.404008676749874e-05,
      "loss": 0.1899,
      "num_input_tokens_seen": 38839280,
      "step": 20490
    },
    {
      "epoch": 3.343421159964108,
      "grad_norm": 0.10861239582300186,
      "learning_rate": 2.403027660431418e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 38848768,
      "step": 20495
    },
    {
      "epoch": 3.3442368871849255,
      "grad_norm": 0.11958617717027664,
      "learning_rate": 2.402046659067314e-05,
      "loss": 0.0694,
      "num_input_tokens_seen": 38858736,
      "step": 20500
    },
    {
      "epoch": 3.3450526144057426,
      "grad_norm": 0.3187848627567291,
      "learning_rate": 2.401065672808847e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 38868000,
      "step": 20505
    },
    {
      "epoch": 3.34586834162656,
      "grad_norm": 1.325252890586853,
      "learning_rate": 2.400084701807296e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 38879280,
      "step": 20510
    },
    {
      "epoch": 3.3466840688473773,
      "grad_norm": 0.9593245983123779,
      "learning_rate": 2.39910374621394e-05,
      "loss": 0.1234,
      "num_input_tokens_seen": 38888928,
      "step": 20515
    },
    {
      "epoch": 3.347499796068195,
      "grad_norm": 1.173136591911316,
      "learning_rate": 2.3981228061800544e-05,
      "loss": 0.0879,
      "num_input_tokens_seen": 38898032,
      "step": 20520
    },
    {
      "epoch": 3.348315523289012,
      "grad_norm": 0.2939949035644531,
      "learning_rate": 2.3971418818569115e-05,
      "loss": 0.0578,
      "num_input_tokens_seen": 38907408,
      "step": 20525
    },
    {
      "epoch": 3.3491312505098296,
      "grad_norm": 1.1860663890838623,
      "learning_rate": 2.3961609733957832e-05,
      "loss": 0.216,
      "num_input_tokens_seen": 38916704,
      "step": 20530
    },
    {
      "epoch": 3.3499469777306468,
      "grad_norm": 1.3224918842315674,
      "learning_rate": 2.395180080947939e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 38926144,
      "step": 20535
    },
    {
      "epoch": 3.3507627049514643,
      "grad_norm": 0.4013027548789978,
      "learning_rate": 2.394199204664642e-05,
      "loss": 0.2191,
      "num_input_tokens_seen": 38936448,
      "step": 20540
    },
    {
      "epoch": 3.3515784321722815,
      "grad_norm": 0.4544830024242401,
      "learning_rate": 2.3932183446971583e-05,
      "loss": 0.076,
      "num_input_tokens_seen": 38945824,
      "step": 20545
    },
    {
      "epoch": 3.352394159393099,
      "grad_norm": 1.0129878520965576,
      "learning_rate": 2.3922375011967473e-05,
      "loss": 0.1176,
      "num_input_tokens_seen": 38956000,
      "step": 20550
    },
    {
      "epoch": 3.353209886613916,
      "grad_norm": 0.18221433460712433,
      "learning_rate": 2.3912566743146676e-05,
      "loss": 0.0709,
      "num_input_tokens_seen": 38966576,
      "step": 20555
    },
    {
      "epoch": 3.3540256138347337,
      "grad_norm": 0.11945207417011261,
      "learning_rate": 2.390275864202176e-05,
      "loss": 0.0721,
      "num_input_tokens_seen": 38975840,
      "step": 20560
    },
    {
      "epoch": 3.354841341055551,
      "grad_norm": 0.492468923330307,
      "learning_rate": 2.3892950710105243e-05,
      "loss": 0.301,
      "num_input_tokens_seen": 38985632,
      "step": 20565
    },
    {
      "epoch": 3.3556570682763684,
      "grad_norm": 0.18687844276428223,
      "learning_rate": 2.3883142948909635e-05,
      "loss": 0.1394,
      "num_input_tokens_seen": 38995632,
      "step": 20570
    },
    {
      "epoch": 3.3564727954971856,
      "grad_norm": 0.2167784869670868,
      "learning_rate": 2.3873335359947433e-05,
      "loss": 0.1175,
      "num_input_tokens_seen": 39004544,
      "step": 20575
    },
    {
      "epoch": 3.357288522718003,
      "grad_norm": 1.831156611442566,
      "learning_rate": 2.3863527944731066e-05,
      "loss": 0.1239,
      "num_input_tokens_seen": 39014480,
      "step": 20580
    },
    {
      "epoch": 3.3581042499388203,
      "grad_norm": 0.2293969690799713,
      "learning_rate": 2.385372070477298e-05,
      "loss": 0.0681,
      "num_input_tokens_seen": 39025232,
      "step": 20585
    },
    {
      "epoch": 3.358919977159638,
      "grad_norm": 0.23796580731868744,
      "learning_rate": 2.384391364158556e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 39034768,
      "step": 20590
    },
    {
      "epoch": 3.359735704380455,
      "grad_norm": 1.0344314575195312,
      "learning_rate": 2.3834106756681185e-05,
      "loss": 0.1074,
      "num_input_tokens_seen": 39044288,
      "step": 20595
    },
    {
      "epoch": 3.3605514316012726,
      "grad_norm": 0.029021130874753,
      "learning_rate": 2.3824300051572206e-05,
      "loss": 0.0519,
      "num_input_tokens_seen": 39053472,
      "step": 20600
    },
    {
      "epoch": 3.3605514316012726,
      "eval_loss": 0.14261886477470398,
      "eval_runtime": 133.3091,
      "eval_samples_per_second": 20.441,
      "eval_steps_per_second": 10.224,
      "num_input_tokens_seen": 39053472,
      "step": 20600
    },
    {
      "epoch": 3.36136715882209,
      "grad_norm": 0.4695662260055542,
      "learning_rate": 2.3814493527770923e-05,
      "loss": 0.0803,
      "num_input_tokens_seen": 39062560,
      "step": 20605
    },
    {
      "epoch": 3.3621828860429073,
      "grad_norm": 0.3912522494792938,
      "learning_rate": 2.3804687186789637e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 39072608,
      "step": 20610
    },
    {
      "epoch": 3.3629986132637244,
      "grad_norm": 1.0977520942687988,
      "learning_rate": 2.379488103014062e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 39081072,
      "step": 20615
    },
    {
      "epoch": 3.363814340484542,
      "grad_norm": 0.044887952506542206,
      "learning_rate": 2.3785075059336086e-05,
      "loss": 0.0418,
      "num_input_tokens_seen": 39090992,
      "step": 20620
    },
    {
      "epoch": 3.3646300677053596,
      "grad_norm": 0.5510915517807007,
      "learning_rate": 2.3775269275888248e-05,
      "loss": 0.1222,
      "num_input_tokens_seen": 39101936,
      "step": 20625
    },
    {
      "epoch": 3.3654457949261767,
      "grad_norm": 0.05751759186387062,
      "learning_rate": 2.3765463681309274e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 39110960,
      "step": 20630
    },
    {
      "epoch": 3.366261522146994,
      "grad_norm": 1.0490270853042603,
      "learning_rate": 2.3755658277111313e-05,
      "loss": 0.0861,
      "num_input_tokens_seen": 39120912,
      "step": 20635
    },
    {
      "epoch": 3.3670772493678114,
      "grad_norm": 0.14503681659698486,
      "learning_rate": 2.374585306480649e-05,
      "loss": 0.1394,
      "num_input_tokens_seen": 39131136,
      "step": 20640
    },
    {
      "epoch": 3.367892976588629,
      "grad_norm": 1.3354748487472534,
      "learning_rate": 2.3736048045906877e-05,
      "loss": 0.0737,
      "num_input_tokens_seen": 39139072,
      "step": 20645
    },
    {
      "epoch": 3.368708703809446,
      "grad_norm": 0.8040873408317566,
      "learning_rate": 2.372624322192454e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 39148896,
      "step": 20650
    },
    {
      "epoch": 3.3695244310302637,
      "grad_norm": 0.11357704550027847,
      "learning_rate": 2.3716438594371516e-05,
      "loss": 0.1947,
      "num_input_tokens_seen": 39158816,
      "step": 20655
    },
    {
      "epoch": 3.370340158251081,
      "grad_norm": 0.6213445663452148,
      "learning_rate": 2.3706634164759784e-05,
      "loss": 0.1129,
      "num_input_tokens_seen": 39168400,
      "step": 20660
    },
    {
      "epoch": 3.3711558854718984,
      "grad_norm": 0.06929223239421844,
      "learning_rate": 2.3696829934601323e-05,
      "loss": 0.0621,
      "num_input_tokens_seen": 39177680,
      "step": 20665
    },
    {
      "epoch": 3.3719716126927155,
      "grad_norm": 0.14483006298542023,
      "learning_rate": 2.3687025905408053e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 39187072,
      "step": 20670
    },
    {
      "epoch": 3.372787339913533,
      "grad_norm": 0.28490176796913147,
      "learning_rate": 2.3677222078691886e-05,
      "loss": 0.0634,
      "num_input_tokens_seen": 39197088,
      "step": 20675
    },
    {
      "epoch": 3.37360306713435,
      "grad_norm": 0.07916145771741867,
      "learning_rate": 2.366741845596471e-05,
      "loss": 0.0618,
      "num_input_tokens_seen": 39206384,
      "step": 20680
    },
    {
      "epoch": 3.374418794355168,
      "grad_norm": 1.8784455060958862,
      "learning_rate": 2.3657615038738343e-05,
      "loss": 0.2355,
      "num_input_tokens_seen": 39216672,
      "step": 20685
    },
    {
      "epoch": 3.375234521575985,
      "grad_norm": 0.15725085139274597,
      "learning_rate": 2.3647811828524614e-05,
      "loss": 0.0727,
      "num_input_tokens_seen": 39227008,
      "step": 20690
    },
    {
      "epoch": 3.3760502487968025,
      "grad_norm": 0.6866264939308167,
      "learning_rate": 2.363800882683529e-05,
      "loss": 0.1379,
      "num_input_tokens_seen": 39237568,
      "step": 20695
    },
    {
      "epoch": 3.3768659760176196,
      "grad_norm": 0.08959080278873444,
      "learning_rate": 2.3628206035182125e-05,
      "loss": 0.1849,
      "num_input_tokens_seen": 39246960,
      "step": 20700
    },
    {
      "epoch": 3.377681703238437,
      "grad_norm": 0.34117791056632996,
      "learning_rate": 2.361840345507683e-05,
      "loss": 0.0442,
      "num_input_tokens_seen": 39256672,
      "step": 20705
    },
    {
      "epoch": 3.3784974304592543,
      "grad_norm": 1.019550085067749,
      "learning_rate": 2.3608601088031073e-05,
      "loss": 0.2867,
      "num_input_tokens_seen": 39266272,
      "step": 20710
    },
    {
      "epoch": 3.379313157680072,
      "grad_norm": 1.1611926555633545,
      "learning_rate": 2.3598798935556516e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 39275632,
      "step": 20715
    },
    {
      "epoch": 3.380128884900889,
      "grad_norm": 0.09654998779296875,
      "learning_rate": 2.3588996999164784e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 39284144,
      "step": 20720
    },
    {
      "epoch": 3.3809446121217066,
      "grad_norm": 0.8113727569580078,
      "learning_rate": 2.3579195280367434e-05,
      "loss": 0.1904,
      "num_input_tokens_seen": 39294000,
      "step": 20725
    },
    {
      "epoch": 3.3817603393425237,
      "grad_norm": 0.8828007578849792,
      "learning_rate": 2.356939378067603e-05,
      "loss": 0.0798,
      "num_input_tokens_seen": 39304224,
      "step": 20730
    },
    {
      "epoch": 3.3825760665633413,
      "grad_norm": 0.08257811516523361,
      "learning_rate": 2.3559592501602092e-05,
      "loss": 0.0652,
      "num_input_tokens_seen": 39313888,
      "step": 20735
    },
    {
      "epoch": 3.3833917937841584,
      "grad_norm": 0.7575118541717529,
      "learning_rate": 2.3549791444657076e-05,
      "loss": 0.1785,
      "num_input_tokens_seen": 39322928,
      "step": 20740
    },
    {
      "epoch": 3.384207521004976,
      "grad_norm": 0.4199017882347107,
      "learning_rate": 2.353999061135246e-05,
      "loss": 0.1078,
      "num_input_tokens_seen": 39331728,
      "step": 20745
    },
    {
      "epoch": 3.385023248225793,
      "grad_norm": 1.0717939138412476,
      "learning_rate": 2.3530190003199626e-05,
      "loss": 0.1486,
      "num_input_tokens_seen": 39340192,
      "step": 20750
    },
    {
      "epoch": 3.3858389754466107,
      "grad_norm": 1.6201452016830444,
      "learning_rate": 2.3520389621709965e-05,
      "loss": 0.2151,
      "num_input_tokens_seen": 39350560,
      "step": 20755
    },
    {
      "epoch": 3.386654702667428,
      "grad_norm": 0.6367127895355225,
      "learning_rate": 2.351058946839483e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 39360208,
      "step": 20760
    },
    {
      "epoch": 3.3874704298882454,
      "grad_norm": 0.46853941679000854,
      "learning_rate": 2.350078954476551e-05,
      "loss": 0.0735,
      "num_input_tokens_seen": 39369984,
      "step": 20765
    },
    {
      "epoch": 3.3882861571090626,
      "grad_norm": 0.9841338396072388,
      "learning_rate": 2.3490989852333272e-05,
      "loss": 0.1274,
      "num_input_tokens_seen": 39377872,
      "step": 20770
    },
    {
      "epoch": 3.38910188432988,
      "grad_norm": 0.39976072311401367,
      "learning_rate": 2.3481190392609377e-05,
      "loss": 0.0878,
      "num_input_tokens_seen": 39386880,
      "step": 20775
    },
    {
      "epoch": 3.3899176115506973,
      "grad_norm": 0.4195132255554199,
      "learning_rate": 2.3471391167105e-05,
      "loss": 0.0689,
      "num_input_tokens_seen": 39395040,
      "step": 20780
    },
    {
      "epoch": 3.390733338771515,
      "grad_norm": 0.24525699019432068,
      "learning_rate": 2.3461592177331325e-05,
      "loss": 0.0404,
      "num_input_tokens_seen": 39403504,
      "step": 20785
    },
    {
      "epoch": 3.391549065992332,
      "grad_norm": 0.7520690560340881,
      "learning_rate": 2.345179342479946e-05,
      "loss": 0.1949,
      "num_input_tokens_seen": 39413680,
      "step": 20790
    },
    {
      "epoch": 3.3923647932131495,
      "grad_norm": 0.26951122283935547,
      "learning_rate": 2.3441994911020503e-05,
      "loss": 0.0625,
      "num_input_tokens_seen": 39423216,
      "step": 20795
    },
    {
      "epoch": 3.393180520433967,
      "grad_norm": 1.1863116025924683,
      "learning_rate": 2.3432196637505522e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 39432032,
      "step": 20800
    },
    {
      "epoch": 3.393180520433967,
      "eval_loss": 0.142646923661232,
      "eval_runtime": 133.315,
      "eval_samples_per_second": 20.44,
      "eval_steps_per_second": 10.224,
      "num_input_tokens_seen": 39432032,
      "step": 20800
    },
    {
      "epoch": 3.3939962476547842,
      "grad_norm": 0.8808227777481079,
      "learning_rate": 2.3422398605765515e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 39441056,
      "step": 20805
    },
    {
      "epoch": 3.3948119748756014,
      "grad_norm": 1.4935678243637085,
      "learning_rate": 2.3412600817311462e-05,
      "loss": 0.2381,
      "num_input_tokens_seen": 39450912,
      "step": 20810
    },
    {
      "epoch": 3.395627702096419,
      "grad_norm": 0.21320274472236633,
      "learning_rate": 2.3402803273654326e-05,
      "loss": 0.0689,
      "num_input_tokens_seen": 39459904,
      "step": 20815
    },
    {
      "epoch": 3.3964434293172365,
      "grad_norm": 0.030844796448946,
      "learning_rate": 2.3393005976304983e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 39469488,
      "step": 20820
    },
    {
      "epoch": 3.3972591565380537,
      "grad_norm": 0.14157159626483917,
      "learning_rate": 2.338320892677432e-05,
      "loss": 0.0855,
      "num_input_tokens_seen": 39479552,
      "step": 20825
    },
    {
      "epoch": 3.398074883758871,
      "grad_norm": 0.19544626772403717,
      "learning_rate": 2.3373412126573155e-05,
      "loss": 0.0774,
      "num_input_tokens_seen": 39490016,
      "step": 20830
    },
    {
      "epoch": 3.3988906109796884,
      "grad_norm": 0.3080514073371887,
      "learning_rate": 2.3363615577212285e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 39499008,
      "step": 20835
    },
    {
      "epoch": 3.399706338200506,
      "grad_norm": 0.2579825222492218,
      "learning_rate": 2.3353819280202455e-05,
      "loss": 0.1152,
      "num_input_tokens_seen": 39507296,
      "step": 20840
    },
    {
      "epoch": 3.400522065421323,
      "grad_norm": 0.041604023426771164,
      "learning_rate": 2.334402323705438e-05,
      "loss": 0.0558,
      "num_input_tokens_seen": 39516480,
      "step": 20845
    },
    {
      "epoch": 3.4013377926421406,
      "grad_norm": 0.671633780002594,
      "learning_rate": 2.3334227449278725e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 39526224,
      "step": 20850
    },
    {
      "epoch": 3.4021535198629578,
      "grad_norm": 1.3981677293777466,
      "learning_rate": 2.3324431918386143e-05,
      "loss": 0.2363,
      "num_input_tokens_seen": 39535840,
      "step": 20855
    },
    {
      "epoch": 3.4029692470837754,
      "grad_norm": 1.2423367500305176,
      "learning_rate": 2.3314636645887207e-05,
      "loss": 0.2136,
      "num_input_tokens_seen": 39544304,
      "step": 20860
    },
    {
      "epoch": 3.4037849743045925,
      "grad_norm": 0.9316734075546265,
      "learning_rate": 2.3304841633292487e-05,
      "loss": 0.186,
      "num_input_tokens_seen": 39554368,
      "step": 20865
    },
    {
      "epoch": 3.40460070152541,
      "grad_norm": 0.046709608286619186,
      "learning_rate": 2.329504688211248e-05,
      "loss": 0.0587,
      "num_input_tokens_seen": 39562480,
      "step": 20870
    },
    {
      "epoch": 3.405416428746227,
      "grad_norm": 0.66534024477005,
      "learning_rate": 2.3285252393857677e-05,
      "loss": 0.0517,
      "num_input_tokens_seen": 39572768,
      "step": 20875
    },
    {
      "epoch": 3.4062321559670448,
      "grad_norm": 0.4929805099964142,
      "learning_rate": 2.327545817003851e-05,
      "loss": 0.0801,
      "num_input_tokens_seen": 39581968,
      "step": 20880
    },
    {
      "epoch": 3.407047883187862,
      "grad_norm": 1.5129872560501099,
      "learning_rate": 2.326566421216535e-05,
      "loss": 0.2045,
      "num_input_tokens_seen": 39591424,
      "step": 20885
    },
    {
      "epoch": 3.4078636104086795,
      "grad_norm": 2.3432562351226807,
      "learning_rate": 2.3255870521748565e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 39601360,
      "step": 20890
    },
    {
      "epoch": 3.4086793376294966,
      "grad_norm": 0.2589523196220398,
      "learning_rate": 2.3246077100298474e-05,
      "loss": 0.0995,
      "num_input_tokens_seen": 39610528,
      "step": 20895
    },
    {
      "epoch": 3.409495064850314,
      "grad_norm": 0.09719590097665787,
      "learning_rate": 2.3236283949325328e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 39621712,
      "step": 20900
    },
    {
      "epoch": 3.4103107920711313,
      "grad_norm": 0.058330416679382324,
      "learning_rate": 2.3226491070339368e-05,
      "loss": 0.0429,
      "num_input_tokens_seen": 39631296,
      "step": 20905
    },
    {
      "epoch": 3.411126519291949,
      "grad_norm": 1.558571219444275,
      "learning_rate": 2.3216698464850762e-05,
      "loss": 0.0854,
      "num_input_tokens_seen": 39640144,
      "step": 20910
    },
    {
      "epoch": 3.411942246512766,
      "grad_norm": 0.037663768976926804,
      "learning_rate": 2.320690613436967e-05,
      "loss": 0.073,
      "num_input_tokens_seen": 39650128,
      "step": 20915
    },
    {
      "epoch": 3.4127579737335836,
      "grad_norm": 0.6168060302734375,
      "learning_rate": 2.3197114080406192e-05,
      "loss": 0.1562,
      "num_input_tokens_seen": 39660768,
      "step": 20920
    },
    {
      "epoch": 3.4135737009544007,
      "grad_norm": 0.3522522747516632,
      "learning_rate": 2.3187322304470365e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 39669472,
      "step": 20925
    },
    {
      "epoch": 3.4143894281752183,
      "grad_norm": 2.092515468597412,
      "learning_rate": 2.3177530808072222e-05,
      "loss": 0.2331,
      "num_input_tokens_seen": 39679648,
      "step": 20930
    },
    {
      "epoch": 3.4152051553960354,
      "grad_norm": 0.7454647421836853,
      "learning_rate": 2.316773959272174e-05,
      "loss": 0.0952,
      "num_input_tokens_seen": 39689360,
      "step": 20935
    },
    {
      "epoch": 3.416020882616853,
      "grad_norm": 0.0718853548169136,
      "learning_rate": 2.3157948659928823e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 39698112,
      "step": 20940
    },
    {
      "epoch": 3.41683660983767,
      "grad_norm": 0.5608310699462891,
      "learning_rate": 2.3148158011203388e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 39707616,
      "step": 20945
    },
    {
      "epoch": 3.4176523370584877,
      "grad_norm": 0.31618732213974,
      "learning_rate": 2.3138367648055253e-05,
      "loss": 0.0995,
      "num_input_tokens_seen": 39716448,
      "step": 20950
    },
    {
      "epoch": 3.418468064279305,
      "grad_norm": 0.643110990524292,
      "learning_rate": 2.312857757199422e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 39726512,
      "step": 20955
    },
    {
      "epoch": 3.4192837915001224,
      "grad_norm": 0.8579285144805908,
      "learning_rate": 2.3118787784530048e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 39735440,
      "step": 20960
    },
    {
      "epoch": 3.4200995187209395,
      "grad_norm": 0.22456404566764832,
      "learning_rate": 2.310899828717243e-05,
      "loss": 0.0743,
      "num_input_tokens_seen": 39744960,
      "step": 20965
    },
    {
      "epoch": 3.420915245941757,
      "grad_norm": 0.29030081629753113,
      "learning_rate": 2.309920908143104e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 39755120,
      "step": 20970
    },
    {
      "epoch": 3.4217309731625742,
      "grad_norm": 0.4651258587837219,
      "learning_rate": 2.308942016881551e-05,
      "loss": 0.115,
      "num_input_tokens_seen": 39764496,
      "step": 20975
    },
    {
      "epoch": 3.422546700383392,
      "grad_norm": 0.31591325998306274,
      "learning_rate": 2.307963155083539e-05,
      "loss": 0.0649,
      "num_input_tokens_seen": 39774352,
      "step": 20980
    },
    {
      "epoch": 3.423362427604209,
      "grad_norm": 0.3567427694797516,
      "learning_rate": 2.306984322900022e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 39785072,
      "step": 20985
    },
    {
      "epoch": 3.4241781548250265,
      "grad_norm": 0.2024267613887787,
      "learning_rate": 2.3060055204819482e-05,
      "loss": 0.0784,
      "num_input_tokens_seen": 39794304,
      "step": 20990
    },
    {
      "epoch": 3.424993882045844,
      "grad_norm": 0.4387156665325165,
      "learning_rate": 2.3050267479802604e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 39803888,
      "step": 20995
    },
    {
      "epoch": 3.4258096092666612,
      "grad_norm": 0.29855284094810486,
      "learning_rate": 2.304048005545899e-05,
      "loss": 0.216,
      "num_input_tokens_seen": 39812704,
      "step": 21000
    },
    {
      "epoch": 3.4258096092666612,
      "eval_loss": 0.142809197306633,
      "eval_runtime": 133.2849,
      "eval_samples_per_second": 20.445,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 39812704,
      "step": 21000
    },
    {
      "epoch": 3.4266253364874784,
      "grad_norm": 0.5273786783218384,
      "learning_rate": 2.3030692933297972e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 39822384,
      "step": 21005
    },
    {
      "epoch": 3.427441063708296,
      "grad_norm": 1.0843130350112915,
      "learning_rate": 2.3020906114828843e-05,
      "loss": 0.086,
      "num_input_tokens_seen": 39830976,
      "step": 21010
    },
    {
      "epoch": 3.4282567909291135,
      "grad_norm": 0.16450826823711395,
      "learning_rate": 2.301111960156088e-05,
      "loss": 0.2663,
      "num_input_tokens_seen": 39841264,
      "step": 21015
    },
    {
      "epoch": 3.4290725181499306,
      "grad_norm": 0.858090341091156,
      "learning_rate": 2.300133339500326e-05,
      "loss": 0.0689,
      "num_input_tokens_seen": 39850496,
      "step": 21020
    },
    {
      "epoch": 3.429888245370748,
      "grad_norm": 0.1458652764558792,
      "learning_rate": 2.2991547496665148e-05,
      "loss": 0.2193,
      "num_input_tokens_seen": 39858928,
      "step": 21025
    },
    {
      "epoch": 3.4307039725915653,
      "grad_norm": 0.20151811838150024,
      "learning_rate": 2.298176190805565e-05,
      "loss": 0.0667,
      "num_input_tokens_seen": 39867760,
      "step": 21030
    },
    {
      "epoch": 3.431519699812383,
      "grad_norm": 0.10337120294570923,
      "learning_rate": 2.2971976630683826e-05,
      "loss": 0.0631,
      "num_input_tokens_seen": 39877952,
      "step": 21035
    },
    {
      "epoch": 3.4323354270332,
      "grad_norm": 1.123790979385376,
      "learning_rate": 2.29621916660587e-05,
      "loss": 0.0649,
      "num_input_tokens_seen": 39886208,
      "step": 21040
    },
    {
      "epoch": 3.4331511542540176,
      "grad_norm": 0.09438079595565796,
      "learning_rate": 2.295240701568922e-05,
      "loss": 0.0916,
      "num_input_tokens_seen": 39896464,
      "step": 21045
    },
    {
      "epoch": 3.4339668814748348,
      "grad_norm": 0.1136191263794899,
      "learning_rate": 2.2942622681084312e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 39905872,
      "step": 21050
    },
    {
      "epoch": 3.4347826086956523,
      "grad_norm": 1.4649993181228638,
      "learning_rate": 2.293283866375284e-05,
      "loss": 0.2405,
      "num_input_tokens_seen": 39915552,
      "step": 21055
    },
    {
      "epoch": 3.4355983359164695,
      "grad_norm": 0.8536329865455627,
      "learning_rate": 2.2923054965203627e-05,
      "loss": 0.0815,
      "num_input_tokens_seen": 39925344,
      "step": 21060
    },
    {
      "epoch": 3.436414063137287,
      "grad_norm": 1.2985128164291382,
      "learning_rate": 2.2913271586945443e-05,
      "loss": 0.1863,
      "num_input_tokens_seen": 39934784,
      "step": 21065
    },
    {
      "epoch": 3.437229790358104,
      "grad_norm": 1.7800648212432861,
      "learning_rate": 2.290348853048699e-05,
      "loss": 0.1186,
      "num_input_tokens_seen": 39944320,
      "step": 21070
    },
    {
      "epoch": 3.4380455175789217,
      "grad_norm": 0.1483258754014969,
      "learning_rate": 2.2893705797336956e-05,
      "loss": 0.0379,
      "num_input_tokens_seen": 39952256,
      "step": 21075
    },
    {
      "epoch": 3.438861244799739,
      "grad_norm": 0.5891712307929993,
      "learning_rate": 2.288392338900397e-05,
      "loss": 0.2405,
      "num_input_tokens_seen": 39961872,
      "step": 21080
    },
    {
      "epoch": 3.4396769720205564,
      "grad_norm": 1.2163264751434326,
      "learning_rate": 2.2874141306996576e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 39970848,
      "step": 21085
    },
    {
      "epoch": 3.4404926992413736,
      "grad_norm": 0.3359074592590332,
      "learning_rate": 2.2864359552823312e-05,
      "loss": 0.0938,
      "num_input_tokens_seen": 39979856,
      "step": 21090
    },
    {
      "epoch": 3.441308426462191,
      "grad_norm": 1.859217643737793,
      "learning_rate": 2.2854578127992648e-05,
      "loss": 0.254,
      "num_input_tokens_seen": 39989072,
      "step": 21095
    },
    {
      "epoch": 3.4421241536830083,
      "grad_norm": 1.0488313436508179,
      "learning_rate": 2.2844797034012988e-05,
      "loss": 0.054,
      "num_input_tokens_seen": 39999040,
      "step": 21100
    },
    {
      "epoch": 3.442939880903826,
      "grad_norm": 0.7755401134490967,
      "learning_rate": 2.2835016272392722e-05,
      "loss": 0.0922,
      "num_input_tokens_seen": 40009264,
      "step": 21105
    },
    {
      "epoch": 3.443755608124643,
      "grad_norm": 0.7254660725593567,
      "learning_rate": 2.2825235844640142e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 40018992,
      "step": 21110
    },
    {
      "epoch": 3.4445713353454606,
      "grad_norm": 0.4288468062877655,
      "learning_rate": 2.2815455752263522e-05,
      "loss": 0.1519,
      "num_input_tokens_seen": 40028928,
      "step": 21115
    },
    {
      "epoch": 3.4453870625662777,
      "grad_norm": 0.93163001537323,
      "learning_rate": 2.2805675996771092e-05,
      "loss": 0.0565,
      "num_input_tokens_seen": 40040368,
      "step": 21120
    },
    {
      "epoch": 3.4462027897870953,
      "grad_norm": 0.05097810551524162,
      "learning_rate": 2.2795896579670987e-05,
      "loss": 0.1151,
      "num_input_tokens_seen": 40050640,
      "step": 21125
    },
    {
      "epoch": 3.4470185170079124,
      "grad_norm": 0.23014114797115326,
      "learning_rate": 2.2786117502471337e-05,
      "loss": 0.0478,
      "num_input_tokens_seen": 40058928,
      "step": 21130
    },
    {
      "epoch": 3.44783424422873,
      "grad_norm": 0.46132853627204895,
      "learning_rate": 2.2776338766680185e-05,
      "loss": 0.0823,
      "num_input_tokens_seen": 40069056,
      "step": 21135
    },
    {
      "epoch": 3.448649971449547,
      "grad_norm": 2.19551944732666,
      "learning_rate": 2.2766560373805533e-05,
      "loss": 0.166,
      "num_input_tokens_seen": 40078896,
      "step": 21140
    },
    {
      "epoch": 3.4494656986703647,
      "grad_norm": 0.23873572051525116,
      "learning_rate": 2.2756782325355353e-05,
      "loss": 0.1805,
      "num_input_tokens_seen": 40088080,
      "step": 21145
    },
    {
      "epoch": 3.450281425891182,
      "grad_norm": 1.3755522966384888,
      "learning_rate": 2.2747004622837514e-05,
      "loss": 0.1199,
      "num_input_tokens_seen": 40097264,
      "step": 21150
    },
    {
      "epoch": 3.4510971531119994,
      "grad_norm": 0.6096729040145874,
      "learning_rate": 2.2737227267759878e-05,
      "loss": 0.27,
      "num_input_tokens_seen": 40106448,
      "step": 21155
    },
    {
      "epoch": 3.4519128803328165,
      "grad_norm": 0.08111566305160522,
      "learning_rate": 2.272745026163024e-05,
      "loss": 0.0544,
      "num_input_tokens_seen": 40115712,
      "step": 21160
    },
    {
      "epoch": 3.452728607553634,
      "grad_norm": 1.6935352087020874,
      "learning_rate": 2.271767360595633e-05,
      "loss": 0.3179,
      "num_input_tokens_seen": 40125312,
      "step": 21165
    },
    {
      "epoch": 3.4535443347744517,
      "grad_norm": 0.4171724021434784,
      "learning_rate": 2.270789730224583e-05,
      "loss": 0.0844,
      "num_input_tokens_seen": 40135392,
      "step": 21170
    },
    {
      "epoch": 3.454360061995269,
      "grad_norm": 1.7497929334640503,
      "learning_rate": 2.2698121352006367e-05,
      "loss": 0.2177,
      "num_input_tokens_seen": 40145232,
      "step": 21175
    },
    {
      "epoch": 3.455175789216086,
      "grad_norm": 0.16313616931438446,
      "learning_rate": 2.2688345756745517e-05,
      "loss": 0.0742,
      "num_input_tokens_seen": 40154416,
      "step": 21180
    },
    {
      "epoch": 3.4559915164369035,
      "grad_norm": 0.7998703122138977,
      "learning_rate": 2.267857051797081e-05,
      "loss": 0.0934,
      "num_input_tokens_seen": 40164064,
      "step": 21185
    },
    {
      "epoch": 3.456807243657721,
      "grad_norm": 0.06665677577257156,
      "learning_rate": 2.2668795637189695e-05,
      "loss": 0.0342,
      "num_input_tokens_seen": 40172528,
      "step": 21190
    },
    {
      "epoch": 3.457622970878538,
      "grad_norm": 0.38707244396209717,
      "learning_rate": 2.2659021115909586e-05,
      "loss": 0.1173,
      "num_input_tokens_seen": 40181392,
      "step": 21195
    },
    {
      "epoch": 3.4584386980993553,
      "grad_norm": 0.904718816280365,
      "learning_rate": 2.2649246955637847e-05,
      "loss": 0.0973,
      "num_input_tokens_seen": 40191088,
      "step": 21200
    },
    {
      "epoch": 3.4584386980993553,
      "eval_loss": 0.14348314702510834,
      "eval_runtime": 133.2726,
      "eval_samples_per_second": 20.447,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 40191088,
      "step": 21200
    },
    {
      "epoch": 3.459254425320173,
      "grad_norm": 0.9142191410064697,
      "learning_rate": 2.2639473157881766e-05,
      "loss": 0.1914,
      "num_input_tokens_seen": 40200864,
      "step": 21205
    },
    {
      "epoch": 3.4600701525409905,
      "grad_norm": 1.0877704620361328,
      "learning_rate": 2.2629699724148594e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 40210288,
      "step": 21210
    },
    {
      "epoch": 3.4608858797618076,
      "grad_norm": 1.8961459398269653,
      "learning_rate": 2.26199266559455e-05,
      "loss": 0.3472,
      "num_input_tokens_seen": 40219248,
      "step": 21215
    },
    {
      "epoch": 3.461701606982625,
      "grad_norm": 1.2846941947937012,
      "learning_rate": 2.2610153954779625e-05,
      "loss": 0.0812,
      "num_input_tokens_seen": 40229440,
      "step": 21220
    },
    {
      "epoch": 3.4625173342034423,
      "grad_norm": 0.48322799801826477,
      "learning_rate": 2.2600381622158056e-05,
      "loss": 0.0587,
      "num_input_tokens_seen": 40238384,
      "step": 21225
    },
    {
      "epoch": 3.46333306142426,
      "grad_norm": 0.039828822016716,
      "learning_rate": 2.2590609659587783e-05,
      "loss": 0.1659,
      "num_input_tokens_seen": 40247120,
      "step": 21230
    },
    {
      "epoch": 3.464148788645077,
      "grad_norm": 1.4113539457321167,
      "learning_rate": 2.2580838068575787e-05,
      "loss": 0.1022,
      "num_input_tokens_seen": 40255712,
      "step": 21235
    },
    {
      "epoch": 3.4649645158658946,
      "grad_norm": 0.23140022158622742,
      "learning_rate": 2.257106685062896e-05,
      "loss": 0.0932,
      "num_input_tokens_seen": 40263872,
      "step": 21240
    },
    {
      "epoch": 3.4657802430867117,
      "grad_norm": 1.4414631128311157,
      "learning_rate": 2.256129600725415e-05,
      "loss": 0.1847,
      "num_input_tokens_seen": 40273984,
      "step": 21245
    },
    {
      "epoch": 3.4665959703075293,
      "grad_norm": 0.2373601496219635,
      "learning_rate": 2.2551525539958145e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 40283440,
      "step": 21250
    },
    {
      "epoch": 3.4674116975283464,
      "grad_norm": 0.07158520072698593,
      "learning_rate": 2.2541755450247663e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 40292464,
      "step": 21255
    },
    {
      "epoch": 3.468227424749164,
      "grad_norm": 0.9032070636749268,
      "learning_rate": 2.2531985739629382e-05,
      "loss": 0.1978,
      "num_input_tokens_seen": 40302384,
      "step": 21260
    },
    {
      "epoch": 3.469043151969981,
      "grad_norm": 1.5269562005996704,
      "learning_rate": 2.2522216409609924e-05,
      "loss": 0.1555,
      "num_input_tokens_seen": 40311024,
      "step": 21265
    },
    {
      "epoch": 3.4698588791907987,
      "grad_norm": 1.2815748453140259,
      "learning_rate": 2.2512447461695826e-05,
      "loss": 0.1868,
      "num_input_tokens_seen": 40320672,
      "step": 21270
    },
    {
      "epoch": 3.470674606411616,
      "grad_norm": 0.6191981434822083,
      "learning_rate": 2.2502678897393593e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 40330720,
      "step": 21275
    },
    {
      "epoch": 3.4714903336324334,
      "grad_norm": 0.5570726990699768,
      "learning_rate": 2.2492910718209665e-05,
      "loss": 0.0687,
      "num_input_tokens_seen": 40339824,
      "step": 21280
    },
    {
      "epoch": 3.4723060608532506,
      "grad_norm": 0.18272210657596588,
      "learning_rate": 2.2483142925650398e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 40350720,
      "step": 21285
    },
    {
      "epoch": 3.473121788074068,
      "grad_norm": 0.03909534960985184,
      "learning_rate": 2.247337552122213e-05,
      "loss": 0.062,
      "num_input_tokens_seen": 40358608,
      "step": 21290
    },
    {
      "epoch": 3.4739375152948853,
      "grad_norm": 1.3661092519760132,
      "learning_rate": 2.24636085064311e-05,
      "loss": 0.2157,
      "num_input_tokens_seen": 40368336,
      "step": 21295
    },
    {
      "epoch": 3.474753242515703,
      "grad_norm": 0.9205044507980347,
      "learning_rate": 2.245384188278351e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 40377904,
      "step": 21300
    },
    {
      "epoch": 3.47556896973652,
      "grad_norm": 0.22303161025047302,
      "learning_rate": 2.2444075651785513e-05,
      "loss": 0.0703,
      "num_input_tokens_seen": 40387920,
      "step": 21305
    },
    {
      "epoch": 3.4763846969573375,
      "grad_norm": 0.04154827073216438,
      "learning_rate": 2.243430981494316e-05,
      "loss": 0.1827,
      "num_input_tokens_seen": 40396672,
      "step": 21310
    },
    {
      "epoch": 3.4772004241781547,
      "grad_norm": 1.2906736135482788,
      "learning_rate": 2.2424544373762475e-05,
      "loss": 0.1024,
      "num_input_tokens_seen": 40407200,
      "step": 21315
    },
    {
      "epoch": 3.4780161513989722,
      "grad_norm": 0.1564948558807373,
      "learning_rate": 2.2414779329749418e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 40416112,
      "step": 21320
    },
    {
      "epoch": 3.4788318786197894,
      "grad_norm": 1.30617356300354,
      "learning_rate": 2.2405014684409873e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 40425072,
      "step": 21325
    },
    {
      "epoch": 3.479647605840607,
      "grad_norm": 0.6639071702957153,
      "learning_rate": 2.239525043924968e-05,
      "loss": 0.0826,
      "num_input_tokens_seen": 40434944,
      "step": 21330
    },
    {
      "epoch": 3.480463333061424,
      "grad_norm": 0.11811785399913788,
      "learning_rate": 2.2385486595774592e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 40445120,
      "step": 21335
    },
    {
      "epoch": 3.4812790602822417,
      "grad_norm": 0.22117987275123596,
      "learning_rate": 2.237572315549033e-05,
      "loss": 0.0922,
      "num_input_tokens_seen": 40453696,
      "step": 21340
    },
    {
      "epoch": 3.482094787503059,
      "grad_norm": 0.07026801258325577,
      "learning_rate": 2.2365960119902545e-05,
      "loss": 0.0594,
      "num_input_tokens_seen": 40463536,
      "step": 21345
    },
    {
      "epoch": 3.4829105147238764,
      "grad_norm": 0.0952911302447319,
      "learning_rate": 2.2356197490516806e-05,
      "loss": 0.0921,
      "num_input_tokens_seen": 40474464,
      "step": 21350
    },
    {
      "epoch": 3.4837262419446935,
      "grad_norm": 1.087560772895813,
      "learning_rate": 2.234643526883863e-05,
      "loss": 0.2216,
      "num_input_tokens_seen": 40484384,
      "step": 21355
    },
    {
      "epoch": 3.484541969165511,
      "grad_norm": 1.1092647314071655,
      "learning_rate": 2.2336673456373497e-05,
      "loss": 0.1713,
      "num_input_tokens_seen": 40493088,
      "step": 21360
    },
    {
      "epoch": 3.4853576963863286,
      "grad_norm": 1.5009655952453613,
      "learning_rate": 2.2326912054626772e-05,
      "loss": 0.1202,
      "num_input_tokens_seen": 40501968,
      "step": 21365
    },
    {
      "epoch": 3.4861734236071458,
      "grad_norm": 1.3563356399536133,
      "learning_rate": 2.2317151065103813e-05,
      "loss": 0.1476,
      "num_input_tokens_seen": 40511776,
      "step": 21370
    },
    {
      "epoch": 3.486989150827963,
      "grad_norm": 1.5016984939575195,
      "learning_rate": 2.2307390489309865e-05,
      "loss": 0.2322,
      "num_input_tokens_seen": 40522000,
      "step": 21375
    },
    {
      "epoch": 3.4878048780487805,
      "grad_norm": 0.21776576340198517,
      "learning_rate": 2.2297630328750146e-05,
      "loss": 0.2882,
      "num_input_tokens_seen": 40531440,
      "step": 21380
    },
    {
      "epoch": 3.488620605269598,
      "grad_norm": 1.3608430624008179,
      "learning_rate": 2.228787058492979e-05,
      "loss": 0.1947,
      "num_input_tokens_seen": 40540752,
      "step": 21385
    },
    {
      "epoch": 3.489436332490415,
      "grad_norm": 0.9769023656845093,
      "learning_rate": 2.2278111259353875e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 40550080,
      "step": 21390
    },
    {
      "epoch": 3.4902520597112328,
      "grad_norm": 0.4954586625099182,
      "learning_rate": 2.2268352353527395e-05,
      "loss": 0.1346,
      "num_input_tokens_seen": 40559120,
      "step": 21395
    },
    {
      "epoch": 3.49106778693205,
      "grad_norm": 1.034993290901184,
      "learning_rate": 2.225859386895533e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 40567216,
      "step": 21400
    },
    {
      "epoch": 3.49106778693205,
      "eval_loss": 0.141592338681221,
      "eval_runtime": 133.3277,
      "eval_samples_per_second": 20.438,
      "eval_steps_per_second": 10.223,
      "num_input_tokens_seen": 40567216,
      "step": 21400
    },
    {
      "epoch": 3.4918835141528675,
      "grad_norm": 0.06155350059270859,
      "learning_rate": 2.2248835807142525e-05,
      "loss": 0.0461,
      "num_input_tokens_seen": 40577056,
      "step": 21405
    },
    {
      "epoch": 3.4926992413736846,
      "grad_norm": 0.6827909350395203,
      "learning_rate": 2.2239078169593826e-05,
      "loss": 0.0822,
      "num_input_tokens_seen": 40585776,
      "step": 21410
    },
    {
      "epoch": 3.493514968594502,
      "grad_norm": 0.2219216525554657,
      "learning_rate": 2.222932095781396e-05,
      "loss": 0.1084,
      "num_input_tokens_seen": 40594320,
      "step": 21415
    },
    {
      "epoch": 3.4943306958153193,
      "grad_norm": 0.18010617792606354,
      "learning_rate": 2.221956417330762e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 40602976,
      "step": 21420
    },
    {
      "epoch": 3.495146423036137,
      "grad_norm": 0.39198046922683716,
      "learning_rate": 2.2209807817579438e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 40613200,
      "step": 21425
    },
    {
      "epoch": 3.495962150256954,
      "grad_norm": 0.6669220924377441,
      "learning_rate": 2.220005189213394e-05,
      "loss": 0.25,
      "num_input_tokens_seen": 40622128,
      "step": 21430
    },
    {
      "epoch": 3.4967778774777716,
      "grad_norm": 0.543066143989563,
      "learning_rate": 2.2190296398475624e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 40631936,
      "step": 21435
    },
    {
      "epoch": 3.4975936046985887,
      "grad_norm": 0.5443431735038757,
      "learning_rate": 2.2180541338108926e-05,
      "loss": 0.0988,
      "num_input_tokens_seen": 40641776,
      "step": 21440
    },
    {
      "epoch": 3.4984093319194063,
      "grad_norm": 0.18550504744052887,
      "learning_rate": 2.2170786712538176e-05,
      "loss": 0.1298,
      "num_input_tokens_seen": 40651632,
      "step": 21445
    },
    {
      "epoch": 3.4992250591402234,
      "grad_norm": 0.7516301870346069,
      "learning_rate": 2.216103252326768e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 40661584,
      "step": 21450
    },
    {
      "epoch": 3.500040786361041,
      "grad_norm": 0.23726388812065125,
      "learning_rate": 2.2151278771801635e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 40671120,
      "step": 21455
    },
    {
      "epoch": 3.500856513581858,
      "grad_norm": 0.2664308249950409,
      "learning_rate": 2.21415254596442e-05,
      "loss": 0.1952,
      "num_input_tokens_seen": 40679456,
      "step": 21460
    },
    {
      "epoch": 3.5016722408026757,
      "grad_norm": 0.3316175043582916,
      "learning_rate": 2.213177258829947e-05,
      "loss": 0.0964,
      "num_input_tokens_seen": 40688752,
      "step": 21465
    },
    {
      "epoch": 3.502487968023493,
      "grad_norm": 0.024339454248547554,
      "learning_rate": 2.2122020159271445e-05,
      "loss": 0.1793,
      "num_input_tokens_seen": 40698112,
      "step": 21470
    },
    {
      "epoch": 3.5033036952443104,
      "grad_norm": 0.09717108309268951,
      "learning_rate": 2.2112268174064075e-05,
      "loss": 0.0514,
      "num_input_tokens_seen": 40707664,
      "step": 21475
    },
    {
      "epoch": 3.5041194224651275,
      "grad_norm": 0.7138119339942932,
      "learning_rate": 2.2102516634181253e-05,
      "loss": 0.1241,
      "num_input_tokens_seen": 40716528,
      "step": 21480
    },
    {
      "epoch": 3.504935149685945,
      "grad_norm": 0.21075791120529175,
      "learning_rate": 2.209276554112677e-05,
      "loss": 0.2524,
      "num_input_tokens_seen": 40725200,
      "step": 21485
    },
    {
      "epoch": 3.5057508769067622,
      "grad_norm": 1.332502007484436,
      "learning_rate": 2.2083014896404384e-05,
      "loss": 0.1776,
      "num_input_tokens_seen": 40735200,
      "step": 21490
    },
    {
      "epoch": 3.50656660412758,
      "grad_norm": 0.8701757788658142,
      "learning_rate": 2.207326470151775e-05,
      "loss": 0.0999,
      "num_input_tokens_seen": 40745008,
      "step": 21495
    },
    {
      "epoch": 3.507382331348397,
      "grad_norm": 0.658557116985321,
      "learning_rate": 2.2063514957970477e-05,
      "loss": 0.1066,
      "num_input_tokens_seen": 40755216,
      "step": 21500
    },
    {
      "epoch": 3.5081980585692145,
      "grad_norm": 0.1835586279630661,
      "learning_rate": 2.205376566726611e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 40765392,
      "step": 21505
    },
    {
      "epoch": 3.5090137857900316,
      "grad_norm": 2.2431082725524902,
      "learning_rate": 2.204401683090809e-05,
      "loss": 0.1788,
      "num_input_tokens_seen": 40775392,
      "step": 21510
    },
    {
      "epoch": 3.5098295130108492,
      "grad_norm": 1.8059134483337402,
      "learning_rate": 2.203426845039982e-05,
      "loss": 0.0814,
      "num_input_tokens_seen": 40784400,
      "step": 21515
    },
    {
      "epoch": 3.510645240231667,
      "grad_norm": 0.19315163791179657,
      "learning_rate": 2.202452052724464e-05,
      "loss": 0.1365,
      "num_input_tokens_seen": 40794048,
      "step": 21520
    },
    {
      "epoch": 3.511460967452484,
      "grad_norm": 1.6388014554977417,
      "learning_rate": 2.2014773062945777e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 40803344,
      "step": 21525
    },
    {
      "epoch": 3.512276694673301,
      "grad_norm": 0.6213790774345398,
      "learning_rate": 2.2005026059006427e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 40812688,
      "step": 21530
    },
    {
      "epoch": 3.5130924218941186,
      "grad_norm": 1.0001444816589355,
      "learning_rate": 2.1995279516929695e-05,
      "loss": 0.0949,
      "num_input_tokens_seen": 40822288,
      "step": 21535
    },
    {
      "epoch": 3.513908149114936,
      "grad_norm": 0.10035409033298492,
      "learning_rate": 2.1985533438218613e-05,
      "loss": 0.0433,
      "num_input_tokens_seen": 40832272,
      "step": 21540
    },
    {
      "epoch": 3.5147238763357533,
      "grad_norm": 0.6284580230712891,
      "learning_rate": 2.197578782437617e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 40842704,
      "step": 21545
    },
    {
      "epoch": 3.5155396035565705,
      "grad_norm": 0.564678966999054,
      "learning_rate": 2.196604267690524e-05,
      "loss": 0.1277,
      "num_input_tokens_seen": 40852064,
      "step": 21550
    },
    {
      "epoch": 3.516355330777388,
      "grad_norm": 1.4981147050857544,
      "learning_rate": 2.195629799730865e-05,
      "loss": 0.2286,
      "num_input_tokens_seen": 40861936,
      "step": 21555
    },
    {
      "epoch": 3.5171710579982056,
      "grad_norm": 0.413236528635025,
      "learning_rate": 2.1946553787089173e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 40871056,
      "step": 21560
    },
    {
      "epoch": 3.5179867852190227,
      "grad_norm": 0.5983855128288269,
      "learning_rate": 2.193681004774947e-05,
      "loss": 0.0588,
      "num_input_tokens_seen": 40881088,
      "step": 21565
    },
    {
      "epoch": 3.51880251243984,
      "grad_norm": 0.40063586831092834,
      "learning_rate": 2.1927066780792154e-05,
      "loss": 0.1105,
      "num_input_tokens_seen": 40889360,
      "step": 21570
    },
    {
      "epoch": 3.5196182396606575,
      "grad_norm": 1.0707708597183228,
      "learning_rate": 2.191732398771975e-05,
      "loss": 0.0903,
      "num_input_tokens_seen": 40900048,
      "step": 21575
    },
    {
      "epoch": 3.520433966881475,
      "grad_norm": 0.37542521953582764,
      "learning_rate": 2.1907581670034725e-05,
      "loss": 0.1044,
      "num_input_tokens_seen": 40909552,
      "step": 21580
    },
    {
      "epoch": 3.521249694102292,
      "grad_norm": 0.4489583373069763,
      "learning_rate": 2.189783982923948e-05,
      "loss": 0.1751,
      "num_input_tokens_seen": 40917760,
      "step": 21585
    },
    {
      "epoch": 3.5220654213231093,
      "grad_norm": 0.3840826451778412,
      "learning_rate": 2.1888098466836303e-05,
      "loss": 0.0517,
      "num_input_tokens_seen": 40927936,
      "step": 21590
    },
    {
      "epoch": 3.522881148543927,
      "grad_norm": 0.23190823197364807,
      "learning_rate": 2.1878357584327457e-05,
      "loss": 0.1945,
      "num_input_tokens_seen": 40938016,
      "step": 21595
    },
    {
      "epoch": 3.5236968757647444,
      "grad_norm": 0.8581728339195251,
      "learning_rate": 2.1868617183215103e-05,
      "loss": 0.0711,
      "num_input_tokens_seen": 40947696,
      "step": 21600
    },
    {
      "epoch": 3.5236968757647444,
      "eval_loss": 0.14506670832633972,
      "eval_runtime": 133.275,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 40947696,
      "step": 21600
    },
    {
      "epoch": 3.5245126029855616,
      "grad_norm": 0.028218720108270645,
      "learning_rate": 2.1858877265001327e-05,
      "loss": 0.0301,
      "num_input_tokens_seen": 40956576,
      "step": 21605
    },
    {
      "epoch": 3.525328330206379,
      "grad_norm": 1.2387335300445557,
      "learning_rate": 2.184913783118816e-05,
      "loss": 0.2025,
      "num_input_tokens_seen": 40965040,
      "step": 21610
    },
    {
      "epoch": 3.5261440574271963,
      "grad_norm": 0.1643076092004776,
      "learning_rate": 2.1839398883277522e-05,
      "loss": 0.0848,
      "num_input_tokens_seen": 40974672,
      "step": 21615
    },
    {
      "epoch": 3.526959784648014,
      "grad_norm": 0.4382112920284271,
      "learning_rate": 2.182966042277129e-05,
      "loss": 0.0717,
      "num_input_tokens_seen": 40983024,
      "step": 21620
    },
    {
      "epoch": 3.527775511868831,
      "grad_norm": 0.4888452887535095,
      "learning_rate": 2.181992245117128e-05,
      "loss": 0.0908,
      "num_input_tokens_seen": 40992432,
      "step": 21625
    },
    {
      "epoch": 3.5285912390896486,
      "grad_norm": 0.05386295169591904,
      "learning_rate": 2.181018496997918e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 41002400,
      "step": 21630
    },
    {
      "epoch": 3.5294069663104657,
      "grad_norm": 0.09500690549612045,
      "learning_rate": 2.1800447980696648e-05,
      "loss": 0.1567,
      "num_input_tokens_seen": 41010976,
      "step": 21635
    },
    {
      "epoch": 3.5302226935312833,
      "grad_norm": 1.5546579360961914,
      "learning_rate": 2.1790711484825248e-05,
      "loss": 0.166,
      "num_input_tokens_seen": 41020224,
      "step": 21640
    },
    {
      "epoch": 3.5310384207521004,
      "grad_norm": 0.7963579893112183,
      "learning_rate": 2.178097548386646e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 41029936,
      "step": 21645
    },
    {
      "epoch": 3.531854147972918,
      "grad_norm": 0.617689847946167,
      "learning_rate": 2.1771239979321712e-05,
      "loss": 0.081,
      "num_input_tokens_seen": 41039152,
      "step": 21650
    },
    {
      "epoch": 3.532669875193735,
      "grad_norm": 0.05623701959848404,
      "learning_rate": 2.1761504972692327e-05,
      "loss": 0.0982,
      "num_input_tokens_seen": 41048704,
      "step": 21655
    },
    {
      "epoch": 3.5334856024145527,
      "grad_norm": 0.6957537531852722,
      "learning_rate": 2.1751770465479572e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 41059040,
      "step": 21660
    },
    {
      "epoch": 3.53430132963537,
      "grad_norm": 1.3289217948913574,
      "learning_rate": 2.174203645918464e-05,
      "loss": 0.1242,
      "num_input_tokens_seen": 41068160,
      "step": 21665
    },
    {
      "epoch": 3.5351170568561874,
      "grad_norm": 0.36678260564804077,
      "learning_rate": 2.1732302955308624e-05,
      "loss": 0.1044,
      "num_input_tokens_seen": 41077872,
      "step": 21670
    },
    {
      "epoch": 3.5359327840770045,
      "grad_norm": 0.03769486024975777,
      "learning_rate": 2.172256995535255e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 41088320,
      "step": 21675
    },
    {
      "epoch": 3.536748511297822,
      "grad_norm": 0.060591619461774826,
      "learning_rate": 2.171283746081739e-05,
      "loss": 0.0644,
      "num_input_tokens_seen": 41098608,
      "step": 21680
    },
    {
      "epoch": 3.537564238518639,
      "grad_norm": 2.156155824661255,
      "learning_rate": 2.1703105473203988e-05,
      "loss": 0.2438,
      "num_input_tokens_seen": 41108528,
      "step": 21685
    },
    {
      "epoch": 3.538379965739457,
      "grad_norm": 0.9475821256637573,
      "learning_rate": 2.1693373994013168e-05,
      "loss": 0.2082,
      "num_input_tokens_seen": 41118208,
      "step": 21690
    },
    {
      "epoch": 3.539195692960274,
      "grad_norm": 1.8798831701278687,
      "learning_rate": 2.168364302474562e-05,
      "loss": 0.1655,
      "num_input_tokens_seen": 41127776,
      "step": 21695
    },
    {
      "epoch": 3.5400114201810915,
      "grad_norm": 0.9296677708625793,
      "learning_rate": 2.167391256690199e-05,
      "loss": 0.1532,
      "num_input_tokens_seen": 41138176,
      "step": 21700
    },
    {
      "epoch": 3.5408271474019086,
      "grad_norm": 0.7110586166381836,
      "learning_rate": 2.1664182621982855e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 41147520,
      "step": 21705
    },
    {
      "epoch": 3.541642874622726,
      "grad_norm": 0.5803350806236267,
      "learning_rate": 2.1654453191488673e-05,
      "loss": 0.2897,
      "num_input_tokens_seen": 41156560,
      "step": 21710
    },
    {
      "epoch": 3.5424586018435438,
      "grad_norm": 0.3491998314857483,
      "learning_rate": 2.1644724276919846e-05,
      "loss": 0.0388,
      "num_input_tokens_seen": 41166816,
      "step": 21715
    },
    {
      "epoch": 3.543274329064361,
      "grad_norm": 0.33002087473869324,
      "learning_rate": 2.1634995879776715e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 41176576,
      "step": 21720
    },
    {
      "epoch": 3.544090056285178,
      "grad_norm": 1.774640679359436,
      "learning_rate": 2.162526800155949e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 41184880,
      "step": 21725
    },
    {
      "epoch": 3.5449057835059956,
      "grad_norm": 0.5362977385520935,
      "learning_rate": 2.1615540643768363e-05,
      "loss": 0.102,
      "num_input_tokens_seen": 41195472,
      "step": 21730
    },
    {
      "epoch": 3.545721510726813,
      "grad_norm": 0.6444342136383057,
      "learning_rate": 2.160581380790339e-05,
      "loss": 0.2187,
      "num_input_tokens_seen": 41204752,
      "step": 21735
    },
    {
      "epoch": 3.5465372379476303,
      "grad_norm": 0.8647328615188599,
      "learning_rate": 2.1596087495464586e-05,
      "loss": 0.0899,
      "num_input_tokens_seen": 41213568,
      "step": 21740
    },
    {
      "epoch": 3.5473529651684474,
      "grad_norm": 1.177512288093567,
      "learning_rate": 2.1586361707951866e-05,
      "loss": 0.252,
      "num_input_tokens_seen": 41224032,
      "step": 21745
    },
    {
      "epoch": 3.548168692389265,
      "grad_norm": 0.3681838810443878,
      "learning_rate": 2.157663644686507e-05,
      "loss": 0.0524,
      "num_input_tokens_seen": 41233216,
      "step": 21750
    },
    {
      "epoch": 3.5489844196100826,
      "grad_norm": 0.456143319606781,
      "learning_rate": 2.156691171370396e-05,
      "loss": 0.0961,
      "num_input_tokens_seen": 41242960,
      "step": 21755
    },
    {
      "epoch": 3.5498001468308997,
      "grad_norm": 0.4757087826728821,
      "learning_rate": 2.1557187509968195e-05,
      "loss": 0.0786,
      "num_input_tokens_seen": 41252256,
      "step": 21760
    },
    {
      "epoch": 3.550615874051717,
      "grad_norm": 0.23010407388210297,
      "learning_rate": 2.1547463837157382e-05,
      "loss": 0.0736,
      "num_input_tokens_seen": 41260336,
      "step": 21765
    },
    {
      "epoch": 3.5514316012725344,
      "grad_norm": 0.5038819909095764,
      "learning_rate": 2.1537740696771045e-05,
      "loss": 0.1763,
      "num_input_tokens_seen": 41270464,
      "step": 21770
    },
    {
      "epoch": 3.552247328493352,
      "grad_norm": 0.13080178201198578,
      "learning_rate": 2.1528018090308587e-05,
      "loss": 0.2994,
      "num_input_tokens_seen": 41280432,
      "step": 21775
    },
    {
      "epoch": 3.553063055714169,
      "grad_norm": 1.0143475532531738,
      "learning_rate": 2.151829601926938e-05,
      "loss": 0.2166,
      "num_input_tokens_seen": 41289616,
      "step": 21780
    },
    {
      "epoch": 3.5538787829349863,
      "grad_norm": 2.3216819763183594,
      "learning_rate": 2.1508574485152684e-05,
      "loss": 0.1886,
      "num_input_tokens_seen": 41300272,
      "step": 21785
    },
    {
      "epoch": 3.554694510155804,
      "grad_norm": 0.15660852193832397,
      "learning_rate": 2.1498853489457667e-05,
      "loss": 0.0913,
      "num_input_tokens_seen": 41310352,
      "step": 21790
    },
    {
      "epoch": 3.5555102373766214,
      "grad_norm": 0.2879185974597931,
      "learning_rate": 2.1489133033683455e-05,
      "loss": 0.0518,
      "num_input_tokens_seen": 41320816,
      "step": 21795
    },
    {
      "epoch": 3.5563259645974385,
      "grad_norm": 0.4832475483417511,
      "learning_rate": 2.1479413119329038e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 41330624,
      "step": 21800
    },
    {
      "epoch": 3.5563259645974385,
      "eval_loss": 0.14316152036190033,
      "eval_runtime": 133.2849,
      "eval_samples_per_second": 20.445,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 41330624,
      "step": 21800
    },
    {
      "epoch": 3.557141691818256,
      "grad_norm": 0.17931777238845825,
      "learning_rate": 2.1469693747893355e-05,
      "loss": 0.0957,
      "num_input_tokens_seen": 41340608,
      "step": 21805
    },
    {
      "epoch": 3.5579574190390733,
      "grad_norm": 0.7302353978157043,
      "learning_rate": 2.1459974920875274e-05,
      "loss": 0.09,
      "num_input_tokens_seen": 41350240,
      "step": 21810
    },
    {
      "epoch": 3.558773146259891,
      "grad_norm": 0.48526301980018616,
      "learning_rate": 2.145025663977354e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 41360064,
      "step": 21815
    },
    {
      "epoch": 3.559588873480708,
      "grad_norm": 0.16101285815238953,
      "learning_rate": 2.1440538906086844e-05,
      "loss": 0.1192,
      "num_input_tokens_seen": 41368912,
      "step": 21820
    },
    {
      "epoch": 3.5604046007015255,
      "grad_norm": 0.3441193401813507,
      "learning_rate": 2.1430821721313782e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 41379072,
      "step": 21825
    },
    {
      "epoch": 3.5612203279223427,
      "grad_norm": 2.4835705757141113,
      "learning_rate": 2.142110508695286e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 41389056,
      "step": 21830
    },
    {
      "epoch": 3.5620360551431602,
      "grad_norm": 0.10054531693458557,
      "learning_rate": 2.1411389004502515e-05,
      "loss": 0.0392,
      "num_input_tokens_seen": 41398656,
      "step": 21835
    },
    {
      "epoch": 3.5628517823639774,
      "grad_norm": 1.2371710538864136,
      "learning_rate": 2.140167347546107e-05,
      "loss": 0.1233,
      "num_input_tokens_seen": 41408576,
      "step": 21840
    },
    {
      "epoch": 3.563667509584795,
      "grad_norm": 1.3310399055480957,
      "learning_rate": 2.1391958501326793e-05,
      "loss": 0.0939,
      "num_input_tokens_seen": 41417872,
      "step": 21845
    },
    {
      "epoch": 3.564483236805612,
      "grad_norm": 0.17263711988925934,
      "learning_rate": 2.1382244083597873e-05,
      "loss": 0.2549,
      "num_input_tokens_seen": 41426560,
      "step": 21850
    },
    {
      "epoch": 3.5652989640264297,
      "grad_norm": 1.8669098615646362,
      "learning_rate": 2.137253022377237e-05,
      "loss": 0.1825,
      "num_input_tokens_seen": 41436544,
      "step": 21855
    },
    {
      "epoch": 3.566114691247247,
      "grad_norm": 0.0742611289024353,
      "learning_rate": 2.136281692334829e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 41446704,
      "step": 21860
    },
    {
      "epoch": 3.5669304184680644,
      "grad_norm": 0.5786566138267517,
      "learning_rate": 2.135310418382356e-05,
      "loss": 0.0786,
      "num_input_tokens_seen": 41455872,
      "step": 21865
    },
    {
      "epoch": 3.5677461456888815,
      "grad_norm": 0.2827242910861969,
      "learning_rate": 2.134339200669598e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 41465520,
      "step": 21870
    },
    {
      "epoch": 3.568561872909699,
      "grad_norm": 0.2370273321866989,
      "learning_rate": 2.133368039346331e-05,
      "loss": 0.1215,
      "num_input_tokens_seen": 41474416,
      "step": 21875
    },
    {
      "epoch": 3.569377600130516,
      "grad_norm": 0.5009439587593079,
      "learning_rate": 2.1323969345623195e-05,
      "loss": 0.1875,
      "num_input_tokens_seen": 41485136,
      "step": 21880
    },
    {
      "epoch": 3.5701933273513338,
      "grad_norm": 1.8242380619049072,
      "learning_rate": 2.1314258864673207e-05,
      "loss": 0.0823,
      "num_input_tokens_seen": 41495024,
      "step": 21885
    },
    {
      "epoch": 3.5710090545721513,
      "grad_norm": 0.07081816345453262,
      "learning_rate": 2.130454895211082e-05,
      "loss": 0.0858,
      "num_input_tokens_seen": 41503584,
      "step": 21890
    },
    {
      "epoch": 3.5718247817929685,
      "grad_norm": 0.07878760248422623,
      "learning_rate": 2.129483960943342e-05,
      "loss": 0.1863,
      "num_input_tokens_seen": 41513984,
      "step": 21895
    },
    {
      "epoch": 3.5726405090137856,
      "grad_norm": 0.5085501670837402,
      "learning_rate": 2.128513083813831e-05,
      "loss": 0.0542,
      "num_input_tokens_seen": 41523392,
      "step": 21900
    },
    {
      "epoch": 3.573456236234603,
      "grad_norm": 1.3890854120254517,
      "learning_rate": 2.1275422639722724e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 41532880,
      "step": 21905
    },
    {
      "epoch": 3.5742719634554208,
      "grad_norm": 0.11629372835159302,
      "learning_rate": 2.126571501568376e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 41541952,
      "step": 21910
    },
    {
      "epoch": 3.575087690676238,
      "grad_norm": 1.5107500553131104,
      "learning_rate": 2.1256007967518478e-05,
      "loss": 0.1102,
      "num_input_tokens_seen": 41550352,
      "step": 21915
    },
    {
      "epoch": 3.575903417897055,
      "grad_norm": 0.0871625542640686,
      "learning_rate": 2.124630149672381e-05,
      "loss": 0.0605,
      "num_input_tokens_seen": 41560080,
      "step": 21920
    },
    {
      "epoch": 3.5767191451178726,
      "grad_norm": 2.1954967975616455,
      "learning_rate": 2.1236595604796624e-05,
      "loss": 0.195,
      "num_input_tokens_seen": 41569408,
      "step": 21925
    },
    {
      "epoch": 3.57753487233869,
      "grad_norm": 0.5232017040252686,
      "learning_rate": 2.1226890293233693e-05,
      "loss": 0.1214,
      "num_input_tokens_seen": 41578192,
      "step": 21930
    },
    {
      "epoch": 3.5783505995595073,
      "grad_norm": 0.5867001414299011,
      "learning_rate": 2.1217185563531694e-05,
      "loss": 0.0695,
      "num_input_tokens_seen": 41588096,
      "step": 21935
    },
    {
      "epoch": 3.5791663267803244,
      "grad_norm": 1.3693581819534302,
      "learning_rate": 2.120748141718721e-05,
      "loss": 0.2728,
      "num_input_tokens_seen": 41597680,
      "step": 21940
    },
    {
      "epoch": 3.579982054001142,
      "grad_norm": 0.6799781322479248,
      "learning_rate": 2.1197777855696765e-05,
      "loss": 0.1864,
      "num_input_tokens_seen": 41605712,
      "step": 21945
    },
    {
      "epoch": 3.5807977812219596,
      "grad_norm": 0.3195289075374603,
      "learning_rate": 2.1188074880556746e-05,
      "loss": 0.0469,
      "num_input_tokens_seen": 41614816,
      "step": 21950
    },
    {
      "epoch": 3.5816135084427767,
      "grad_norm": 1.2046583890914917,
      "learning_rate": 2.1178372493263495e-05,
      "loss": 0.2531,
      "num_input_tokens_seen": 41624560,
      "step": 21955
    },
    {
      "epoch": 3.582429235663594,
      "grad_norm": 0.03790891170501709,
      "learning_rate": 2.116867069531322e-05,
      "loss": 0.0215,
      "num_input_tokens_seen": 41634496,
      "step": 21960
    },
    {
      "epoch": 3.5832449628844114,
      "grad_norm": 0.057881806045770645,
      "learning_rate": 2.1158969488202073e-05,
      "loss": 0.1317,
      "num_input_tokens_seen": 41643312,
      "step": 21965
    },
    {
      "epoch": 3.584060690105229,
      "grad_norm": 0.5747739672660828,
      "learning_rate": 2.114926887342611e-05,
      "loss": 0.1022,
      "num_input_tokens_seen": 41653472,
      "step": 21970
    },
    {
      "epoch": 3.584876417326046,
      "grad_norm": 1.5144150257110596,
      "learning_rate": 2.113956885248127e-05,
      "loss": 0.0885,
      "num_input_tokens_seen": 41662608,
      "step": 21975
    },
    {
      "epoch": 3.5856921445468637,
      "grad_norm": 1.6461156606674194,
      "learning_rate": 2.112986942686342e-05,
      "loss": 0.0427,
      "num_input_tokens_seen": 41671776,
      "step": 21980
    },
    {
      "epoch": 3.586507871767681,
      "grad_norm": 1.2626327276229858,
      "learning_rate": 2.112017059806835e-05,
      "loss": 0.101,
      "num_input_tokens_seen": 41680272,
      "step": 21985
    },
    {
      "epoch": 3.5873235989884984,
      "grad_norm": 0.2005404829978943,
      "learning_rate": 2.1110472367591724e-05,
      "loss": 0.1754,
      "num_input_tokens_seen": 41690080,
      "step": 21990
    },
    {
      "epoch": 3.5881393262093155,
      "grad_norm": 0.8241003155708313,
      "learning_rate": 2.1100774736929145e-05,
      "loss": 0.208,
      "num_input_tokens_seen": 41699168,
      "step": 21995
    },
    {
      "epoch": 3.588955053430133,
      "grad_norm": 0.42732834815979004,
      "learning_rate": 2.10910777075761e-05,
      "loss": 0.1808,
      "num_input_tokens_seen": 41708800,
      "step": 22000
    },
    {
      "epoch": 3.588955053430133,
      "eval_loss": 0.14470598101615906,
      "eval_runtime": 133.3861,
      "eval_samples_per_second": 20.429,
      "eval_steps_per_second": 10.218,
      "num_input_tokens_seen": 41708800,
      "step": 22000
    },
    {
      "epoch": 3.5897707806509502,
      "grad_norm": 0.2291581630706787,
      "learning_rate": 2.108138128102799e-05,
      "loss": 0.049,
      "num_input_tokens_seen": 41719280,
      "step": 22005
    },
    {
      "epoch": 3.590586507871768,
      "grad_norm": 0.19753888249397278,
      "learning_rate": 2.107168545878014e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 41727968,
      "step": 22010
    },
    {
      "epoch": 3.591402235092585,
      "grad_norm": 0.846733570098877,
      "learning_rate": 2.106199024232775e-05,
      "loss": 0.0486,
      "num_input_tokens_seen": 41738160,
      "step": 22015
    },
    {
      "epoch": 3.5922179623134025,
      "grad_norm": 0.09710843861103058,
      "learning_rate": 2.105229563316595e-05,
      "loss": 0.0577,
      "num_input_tokens_seen": 41746832,
      "step": 22020
    },
    {
      "epoch": 3.5930336895342196,
      "grad_norm": 0.36567285656929016,
      "learning_rate": 2.1042601632789784e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 41756304,
      "step": 22025
    },
    {
      "epoch": 3.593849416755037,
      "grad_norm": 0.11419189721345901,
      "learning_rate": 2.103290824269417e-05,
      "loss": 0.0974,
      "num_input_tokens_seen": 41765952,
      "step": 22030
    },
    {
      "epoch": 3.5946651439758543,
      "grad_norm": 2.0174059867858887,
      "learning_rate": 2.1023215464373965e-05,
      "loss": 0.1095,
      "num_input_tokens_seen": 41776080,
      "step": 22035
    },
    {
      "epoch": 3.595480871196672,
      "grad_norm": 0.8370333909988403,
      "learning_rate": 2.1013523299323908e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 41786448,
      "step": 22040
    },
    {
      "epoch": 3.596296598417489,
      "grad_norm": 0.8536776900291443,
      "learning_rate": 2.1003831749038654e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 41796848,
      "step": 22045
    },
    {
      "epoch": 3.5971123256383066,
      "grad_norm": 0.5076929330825806,
      "learning_rate": 2.099414081501277e-05,
      "loss": 0.1437,
      "num_input_tokens_seen": 41805520,
      "step": 22050
    },
    {
      "epoch": 3.5979280528591238,
      "grad_norm": 0.2981204390525818,
      "learning_rate": 2.09844504987407e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 41813168,
      "step": 22055
    },
    {
      "epoch": 3.5987437800799413,
      "grad_norm": 1.2045081853866577,
      "learning_rate": 2.097476080171683e-05,
      "loss": 0.2688,
      "num_input_tokens_seen": 41822768,
      "step": 22060
    },
    {
      "epoch": 3.5995595073007585,
      "grad_norm": 1.3055968284606934,
      "learning_rate": 2.0965071725435436e-05,
      "loss": 0.1177,
      "num_input_tokens_seen": 41832624,
      "step": 22065
    },
    {
      "epoch": 3.600375234521576,
      "grad_norm": 0.6108844876289368,
      "learning_rate": 2.0955383271390684e-05,
      "loss": 0.0473,
      "num_input_tokens_seen": 41840624,
      "step": 22070
    },
    {
      "epoch": 3.601190961742393,
      "grad_norm": 1.6068111658096313,
      "learning_rate": 2.094569544107666e-05,
      "loss": 0.2367,
      "num_input_tokens_seen": 41848944,
      "step": 22075
    },
    {
      "epoch": 3.6020066889632107,
      "grad_norm": 0.0717737153172493,
      "learning_rate": 2.093600823598735e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 41858752,
      "step": 22080
    },
    {
      "epoch": 3.6028224161840283,
      "grad_norm": 0.9173979163169861,
      "learning_rate": 2.092632165761663e-05,
      "loss": 0.0915,
      "num_input_tokens_seen": 41869216,
      "step": 22085
    },
    {
      "epoch": 3.6036381434048455,
      "grad_norm": 0.9092733860015869,
      "learning_rate": 2.091663570745832e-05,
      "loss": 0.227,
      "num_input_tokens_seen": 41878464,
      "step": 22090
    },
    {
      "epoch": 3.6044538706256626,
      "grad_norm": 2.748973846435547,
      "learning_rate": 2.0906950387006086e-05,
      "loss": 0.2268,
      "num_input_tokens_seen": 41888528,
      "step": 22095
    },
    {
      "epoch": 3.60526959784648,
      "grad_norm": 0.5954205989837646,
      "learning_rate": 2.0897265697753543e-05,
      "loss": 0.0804,
      "num_input_tokens_seen": 41897824,
      "step": 22100
    },
    {
      "epoch": 3.6060853250672977,
      "grad_norm": 0.13280044496059418,
      "learning_rate": 2.088758164119419e-05,
      "loss": 0.0711,
      "num_input_tokens_seen": 41906640,
      "step": 22105
    },
    {
      "epoch": 3.606901052288115,
      "grad_norm": 0.9148492217063904,
      "learning_rate": 2.0877898218821428e-05,
      "loss": 0.2546,
      "num_input_tokens_seen": 41915728,
      "step": 22110
    },
    {
      "epoch": 3.607716779508932,
      "grad_norm": 0.7707553505897522,
      "learning_rate": 2.0868215432128565e-05,
      "loss": 0.116,
      "num_input_tokens_seen": 41925968,
      "step": 22115
    },
    {
      "epoch": 3.6085325067297496,
      "grad_norm": 0.3528697192668915,
      "learning_rate": 2.0858533282608796e-05,
      "loss": 0.1896,
      "num_input_tokens_seen": 41935504,
      "step": 22120
    },
    {
      "epoch": 3.609348233950567,
      "grad_norm": 0.2803976535797119,
      "learning_rate": 2.084885177175524e-05,
      "loss": 0.1663,
      "num_input_tokens_seen": 41944544,
      "step": 22125
    },
    {
      "epoch": 3.6101639611713843,
      "grad_norm": 0.4202144742012024,
      "learning_rate": 2.0839170901060917e-05,
      "loss": 0.1995,
      "num_input_tokens_seen": 41954880,
      "step": 22130
    },
    {
      "epoch": 3.6109796883922014,
      "grad_norm": 0.9616308212280273,
      "learning_rate": 2.082949067201872e-05,
      "loss": 0.0723,
      "num_input_tokens_seen": 41963808,
      "step": 22135
    },
    {
      "epoch": 3.611795415613019,
      "grad_norm": 1.4191062450408936,
      "learning_rate": 2.0819811086121475e-05,
      "loss": 0.0807,
      "num_input_tokens_seen": 41973120,
      "step": 22140
    },
    {
      "epoch": 3.6126111428338366,
      "grad_norm": 0.665264904499054,
      "learning_rate": 2.08101321448619e-05,
      "loss": 0.0543,
      "num_input_tokens_seen": 41982960,
      "step": 22145
    },
    {
      "epoch": 3.6134268700546537,
      "grad_norm": 0.12061848491430283,
      "learning_rate": 2.080045384973259e-05,
      "loss": 0.2013,
      "num_input_tokens_seen": 41993008,
      "step": 22150
    },
    {
      "epoch": 3.614242597275471,
      "grad_norm": 0.6874961256980896,
      "learning_rate": 2.0790776202226082e-05,
      "loss": 0.1827,
      "num_input_tokens_seen": 42003056,
      "step": 22155
    },
    {
      "epoch": 3.6150583244962884,
      "grad_norm": 0.37441983819007874,
      "learning_rate": 2.078109920383477e-05,
      "loss": 0.105,
      "num_input_tokens_seen": 42013200,
      "step": 22160
    },
    {
      "epoch": 3.615874051717106,
      "grad_norm": 0.8098769187927246,
      "learning_rate": 2.0771422856050978e-05,
      "loss": 0.3746,
      "num_input_tokens_seen": 42022272,
      "step": 22165
    },
    {
      "epoch": 3.616689778937923,
      "grad_norm": 0.36997923254966736,
      "learning_rate": 2.076174716036693e-05,
      "loss": 0.1048,
      "num_input_tokens_seen": 42031376,
      "step": 22170
    },
    {
      "epoch": 3.6175055061587407,
      "grad_norm": 0.5359885692596436,
      "learning_rate": 2.075207211827472e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 42041376,
      "step": 22175
    },
    {
      "epoch": 3.618321233379558,
      "grad_norm": 1.4174479246139526,
      "learning_rate": 2.074239773126638e-05,
      "loss": 0.1212,
      "num_input_tokens_seen": 42051808,
      "step": 22180
    },
    {
      "epoch": 3.6191369606003754,
      "grad_norm": 0.7908551096916199,
      "learning_rate": 2.073272400083382e-05,
      "loss": 0.0662,
      "num_input_tokens_seen": 42061392,
      "step": 22185
    },
    {
      "epoch": 3.6199526878211925,
      "grad_norm": 1.1812525987625122,
      "learning_rate": 2.072305092846883e-05,
      "loss": 0.2115,
      "num_input_tokens_seen": 42070320,
      "step": 22190
    },
    {
      "epoch": 3.62076841504201,
      "grad_norm": 1.023435354232788,
      "learning_rate": 2.0713378515663152e-05,
      "loss": 0.1908,
      "num_input_tokens_seen": 42079280,
      "step": 22195
    },
    {
      "epoch": 3.621584142262827,
      "grad_norm": 0.1729682832956314,
      "learning_rate": 2.070370676390836e-05,
      "loss": 0.0172,
      "num_input_tokens_seen": 42087824,
      "step": 22200
    },
    {
      "epoch": 3.621584142262827,
      "eval_loss": 0.1432579904794693,
      "eval_runtime": 133.3101,
      "eval_samples_per_second": 20.441,
      "eval_steps_per_second": 10.224,
      "num_input_tokens_seen": 42087824,
      "step": 22200
    },
    {
      "epoch": 3.622399869483645,
      "grad_norm": 0.18375761806964874,
      "learning_rate": 2.0694035674695974e-05,
      "loss": 0.0462,
      "num_input_tokens_seen": 42097584,
      "step": 22205
    },
    {
      "epoch": 3.623215596704462,
      "grad_norm": 1.7807406187057495,
      "learning_rate": 2.0684365249517416e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 42107744,
      "step": 22210
    },
    {
      "epoch": 3.6240313239252795,
      "grad_norm": 0.13475212454795837,
      "learning_rate": 2.067469548986396e-05,
      "loss": 0.0629,
      "num_input_tokens_seen": 42116448,
      "step": 22215
    },
    {
      "epoch": 3.6248470511460966,
      "grad_norm": 0.3267293870449066,
      "learning_rate": 2.066502639722681e-05,
      "loss": 0.207,
      "num_input_tokens_seen": 42126144,
      "step": 22220
    },
    {
      "epoch": 3.625662778366914,
      "grad_norm": 1.5954020023345947,
      "learning_rate": 2.065535797309708e-05,
      "loss": 0.2232,
      "num_input_tokens_seen": 42135072,
      "step": 22225
    },
    {
      "epoch": 3.6264785055877313,
      "grad_norm": 0.29721277952194214,
      "learning_rate": 2.0645690218965736e-05,
      "loss": 0.0325,
      "num_input_tokens_seen": 42144752,
      "step": 22230
    },
    {
      "epoch": 3.627294232808549,
      "grad_norm": 1.0660350322723389,
      "learning_rate": 2.063602313632369e-05,
      "loss": 0.3,
      "num_input_tokens_seen": 42154880,
      "step": 22235
    },
    {
      "epoch": 3.628109960029366,
      "grad_norm": 0.05159661918878555,
      "learning_rate": 2.0626356726661704e-05,
      "loss": 0.1695,
      "num_input_tokens_seen": 42164560,
      "step": 22240
    },
    {
      "epoch": 3.6289256872501836,
      "grad_norm": 0.2926480174064636,
      "learning_rate": 2.0616690991470477e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 42174016,
      "step": 22245
    },
    {
      "epoch": 3.6297414144710007,
      "grad_norm": 1.1910234689712524,
      "learning_rate": 2.0607025932240595e-05,
      "loss": 0.2543,
      "num_input_tokens_seen": 42181104,
      "step": 22250
    },
    {
      "epoch": 3.6305571416918183,
      "grad_norm": 1.0140410661697388,
      "learning_rate": 2.059736155046251e-05,
      "loss": 0.124,
      "num_input_tokens_seen": 42189456,
      "step": 22255
    },
    {
      "epoch": 3.631372868912636,
      "grad_norm": 0.3674299120903015,
      "learning_rate": 2.0587697847626603e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 42198912,
      "step": 22260
    },
    {
      "epoch": 3.632188596133453,
      "grad_norm": 0.1802603006362915,
      "learning_rate": 2.057803482522314e-05,
      "loss": 0.069,
      "num_input_tokens_seen": 42208352,
      "step": 22265
    },
    {
      "epoch": 3.63300432335427,
      "grad_norm": 1.2719846963882446,
      "learning_rate": 2.056837248474227e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 42218304,
      "step": 22270
    },
    {
      "epoch": 3.6338200505750877,
      "grad_norm": 0.22887352108955383,
      "learning_rate": 2.0558710827674064e-05,
      "loss": 0.1094,
      "num_input_tokens_seen": 42226928,
      "step": 22275
    },
    {
      "epoch": 3.6346357777959053,
      "grad_norm": 0.8224477171897888,
      "learning_rate": 2.054904985550845e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 42236464,
      "step": 22280
    },
    {
      "epoch": 3.6354515050167224,
      "grad_norm": 0.9922171235084534,
      "learning_rate": 2.0539389569735287e-05,
      "loss": 0.1876,
      "num_input_tokens_seen": 42245280,
      "step": 22285
    },
    {
      "epoch": 3.6362672322375396,
      "grad_norm": 1.4762208461761475,
      "learning_rate": 2.052972997184431e-05,
      "loss": 0.1366,
      "num_input_tokens_seen": 42253840,
      "step": 22290
    },
    {
      "epoch": 3.637082959458357,
      "grad_norm": 2.005229949951172,
      "learning_rate": 2.0520071063325146e-05,
      "loss": 0.1055,
      "num_input_tokens_seen": 42262736,
      "step": 22295
    },
    {
      "epoch": 3.6378986866791747,
      "grad_norm": 0.9499843120574951,
      "learning_rate": 2.051041284566732e-05,
      "loss": 0.2766,
      "num_input_tokens_seen": 42271664,
      "step": 22300
    },
    {
      "epoch": 3.638714413899992,
      "grad_norm": 0.28469961881637573,
      "learning_rate": 2.050075532036026e-05,
      "loss": 0.1159,
      "num_input_tokens_seen": 42281312,
      "step": 22305
    },
    {
      "epoch": 3.639530141120809,
      "grad_norm": 0.25527775287628174,
      "learning_rate": 2.0491098488893264e-05,
      "loss": 0.0687,
      "num_input_tokens_seen": 42290640,
      "step": 22310
    },
    {
      "epoch": 3.6403458683416265,
      "grad_norm": 2.166116952896118,
      "learning_rate": 2.0481442352755546e-05,
      "loss": 0.3105,
      "num_input_tokens_seen": 42300976,
      "step": 22315
    },
    {
      "epoch": 3.641161595562444,
      "grad_norm": 1.010463833808899,
      "learning_rate": 2.0471786913436198e-05,
      "loss": 0.1765,
      "num_input_tokens_seen": 42310896,
      "step": 22320
    },
    {
      "epoch": 3.6419773227832613,
      "grad_norm": 0.19907525181770325,
      "learning_rate": 2.0462132172424218e-05,
      "loss": 0.0481,
      "num_input_tokens_seen": 42320784,
      "step": 22325
    },
    {
      "epoch": 3.6427930500040784,
      "grad_norm": 1.4173423051834106,
      "learning_rate": 2.0452478131208484e-05,
      "loss": 0.367,
      "num_input_tokens_seen": 42331072,
      "step": 22330
    },
    {
      "epoch": 3.643608777224896,
      "grad_norm": 1.1432923078536987,
      "learning_rate": 2.0442824791277765e-05,
      "loss": 0.1983,
      "num_input_tokens_seen": 42340688,
      "step": 22335
    },
    {
      "epoch": 3.6444245044457135,
      "grad_norm": 1.5234376192092896,
      "learning_rate": 2.0433172154120727e-05,
      "loss": 0.2659,
      "num_input_tokens_seen": 42349312,
      "step": 22340
    },
    {
      "epoch": 3.6452402316665307,
      "grad_norm": 1.4853898286819458,
      "learning_rate": 2.0423520221225947e-05,
      "loss": 0.1636,
      "num_input_tokens_seen": 42357200,
      "step": 22345
    },
    {
      "epoch": 3.6460559588873482,
      "grad_norm": 0.8296502828598022,
      "learning_rate": 2.0413868994081848e-05,
      "loss": 0.0865,
      "num_input_tokens_seen": 42366688,
      "step": 22350
    },
    {
      "epoch": 3.6468716861081654,
      "grad_norm": 0.6451224088668823,
      "learning_rate": 2.0404218474176795e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 42375920,
      "step": 22355
    },
    {
      "epoch": 3.647687413328983,
      "grad_norm": 0.3018069863319397,
      "learning_rate": 2.0394568662999002e-05,
      "loss": 0.1433,
      "num_input_tokens_seen": 42385520,
      "step": 22360
    },
    {
      "epoch": 3.6485031405498,
      "grad_norm": 0.33584877848625183,
      "learning_rate": 2.0384919562036593e-05,
      "loss": 0.1733,
      "num_input_tokens_seen": 42395584,
      "step": 22365
    },
    {
      "epoch": 3.6493188677706176,
      "grad_norm": 0.28320279717445374,
      "learning_rate": 2.0375271172777593e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 42404256,
      "step": 22370
    },
    {
      "epoch": 3.6501345949914348,
      "grad_norm": 0.2998692989349365,
      "learning_rate": 2.0365623496709885e-05,
      "loss": 0.0206,
      "num_input_tokens_seen": 42414016,
      "step": 22375
    },
    {
      "epoch": 3.6509503222122524,
      "grad_norm": 0.3969086706638336,
      "learning_rate": 2.0355976535321283e-05,
      "loss": 0.0683,
      "num_input_tokens_seen": 42423840,
      "step": 22380
    },
    {
      "epoch": 3.6517660494330695,
      "grad_norm": 0.9699941277503967,
      "learning_rate": 2.034633029009945e-05,
      "loss": 0.049,
      "num_input_tokens_seen": 42433568,
      "step": 22385
    },
    {
      "epoch": 3.652581776653887,
      "grad_norm": 1.560117244720459,
      "learning_rate": 2.0336684762531972e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 42442976,
      "step": 22390
    },
    {
      "epoch": 3.653397503874704,
      "grad_norm": 0.40815234184265137,
      "learning_rate": 2.032703995410631e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 42451600,
      "step": 22395
    },
    {
      "epoch": 3.6542132310955218,
      "grad_norm": 0.19441862404346466,
      "learning_rate": 2.031739586630981e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 42461936,
      "step": 22400
    },
    {
      "epoch": 3.6542132310955218,
      "eval_loss": 0.14719568192958832,
      "eval_runtime": 133.3438,
      "eval_samples_per_second": 20.436,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 42461936,
      "step": 22400
    },
    {
      "epoch": 3.655028958316339,
      "grad_norm": 0.28003692626953125,
      "learning_rate": 2.0307752500629707e-05,
      "loss": 0.2811,
      "num_input_tokens_seen": 42471968,
      "step": 22405
    },
    {
      "epoch": 3.6558446855371565,
      "grad_norm": 0.21572554111480713,
      "learning_rate": 2.0298109858553144e-05,
      "loss": 0.0972,
      "num_input_tokens_seen": 42480288,
      "step": 22410
    },
    {
      "epoch": 3.6566604127579736,
      "grad_norm": 0.10169428586959839,
      "learning_rate": 2.028846794156712e-05,
      "loss": 0.2831,
      "num_input_tokens_seen": 42489504,
      "step": 22415
    },
    {
      "epoch": 3.657476139978791,
      "grad_norm": 0.10576808452606201,
      "learning_rate": 2.027882675115856e-05,
      "loss": 0.0255,
      "num_input_tokens_seen": 42499440,
      "step": 22420
    },
    {
      "epoch": 3.6582918671996083,
      "grad_norm": 0.32688233256340027,
      "learning_rate": 2.026918628881423e-05,
      "loss": 0.0613,
      "num_input_tokens_seen": 42509088,
      "step": 22425
    },
    {
      "epoch": 3.659107594420426,
      "grad_norm": 0.8960078954696655,
      "learning_rate": 2.0259546556020833e-05,
      "loss": 0.1006,
      "num_input_tokens_seen": 42519328,
      "step": 22430
    },
    {
      "epoch": 3.659923321641243,
      "grad_norm": 0.5269643664360046,
      "learning_rate": 2.024990755426493e-05,
      "loss": 0.2398,
      "num_input_tokens_seen": 42528976,
      "step": 22435
    },
    {
      "epoch": 3.6607390488620606,
      "grad_norm": 0.23161335289478302,
      "learning_rate": 2.0240269285032975e-05,
      "loss": 0.0788,
      "num_input_tokens_seen": 42539424,
      "step": 22440
    },
    {
      "epoch": 3.6615547760828777,
      "grad_norm": 0.6006945967674255,
      "learning_rate": 2.0230631749811306e-05,
      "loss": 0.0734,
      "num_input_tokens_seen": 42548640,
      "step": 22445
    },
    {
      "epoch": 3.6623705033036953,
      "grad_norm": 2.2624244689941406,
      "learning_rate": 2.0220994950086162e-05,
      "loss": 0.271,
      "num_input_tokens_seen": 42557808,
      "step": 22450
    },
    {
      "epoch": 3.663186230524513,
      "grad_norm": 0.656875729560852,
      "learning_rate": 2.021135888734365e-05,
      "loss": 0.0635,
      "num_input_tokens_seen": 42567632,
      "step": 22455
    },
    {
      "epoch": 3.66400195774533,
      "grad_norm": 0.06215217709541321,
      "learning_rate": 2.0201723563069783e-05,
      "loss": 0.0454,
      "num_input_tokens_seen": 42576720,
      "step": 22460
    },
    {
      "epoch": 3.664817684966147,
      "grad_norm": 0.24144211411476135,
      "learning_rate": 2.0192088978750433e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 42586832,
      "step": 22465
    },
    {
      "epoch": 3.6656334121869647,
      "grad_norm": 0.4782620370388031,
      "learning_rate": 2.0182455135871385e-05,
      "loss": 0.0698,
      "num_input_tokens_seen": 42595312,
      "step": 22470
    },
    {
      "epoch": 3.6664491394077823,
      "grad_norm": 0.8218011856079102,
      "learning_rate": 2.0172822035918305e-05,
      "loss": 0.0965,
      "num_input_tokens_seen": 42604400,
      "step": 22475
    },
    {
      "epoch": 3.6672648666285994,
      "grad_norm": 0.26745688915252686,
      "learning_rate": 2.016318968037671e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 42614272,
      "step": 22480
    },
    {
      "epoch": 3.6680805938494165,
      "grad_norm": 0.13224972784519196,
      "learning_rate": 2.015355807073206e-05,
      "loss": 0.0638,
      "num_input_tokens_seen": 42622240,
      "step": 22485
    },
    {
      "epoch": 3.668896321070234,
      "grad_norm": 0.7539871335029602,
      "learning_rate": 2.0143927208469664e-05,
      "loss": 0.1886,
      "num_input_tokens_seen": 42632752,
      "step": 22490
    },
    {
      "epoch": 3.6697120482910517,
      "grad_norm": 0.06657547503709793,
      "learning_rate": 2.0134297095074708e-05,
      "loss": 0.0454,
      "num_input_tokens_seen": 42643168,
      "step": 22495
    },
    {
      "epoch": 3.670527775511869,
      "grad_norm": 0.5301278829574585,
      "learning_rate": 2.0124667732032297e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 42652672,
      "step": 22500
    },
    {
      "epoch": 3.671343502732686,
      "grad_norm": 1.3471862077713013,
      "learning_rate": 2.011503912082738e-05,
      "loss": 0.1936,
      "num_input_tokens_seen": 42661472,
      "step": 22505
    },
    {
      "epoch": 3.6721592299535035,
      "grad_norm": 0.03948565945029259,
      "learning_rate": 2.0105411262944823e-05,
      "loss": 0.1532,
      "num_input_tokens_seen": 42671264,
      "step": 22510
    },
    {
      "epoch": 3.672974957174321,
      "grad_norm": 1.3784326314926147,
      "learning_rate": 2.0095784159869366e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 42680144,
      "step": 22515
    },
    {
      "epoch": 3.6737906843951382,
      "grad_norm": 0.05115339532494545,
      "learning_rate": 2.0086157813085608e-05,
      "loss": 0.0339,
      "num_input_tokens_seen": 42688848,
      "step": 22520
    },
    {
      "epoch": 3.6746064116159554,
      "grad_norm": 0.3764669895172119,
      "learning_rate": 2.0076532224078068e-05,
      "loss": 0.0796,
      "num_input_tokens_seen": 42698672,
      "step": 22525
    },
    {
      "epoch": 3.675422138836773,
      "grad_norm": 0.09814213961362839,
      "learning_rate": 2.0066907394331142e-05,
      "loss": 0.0317,
      "num_input_tokens_seen": 42708720,
      "step": 22530
    },
    {
      "epoch": 3.6762378660575905,
      "grad_norm": 0.8544650673866272,
      "learning_rate": 2.0057283325329077e-05,
      "loss": 0.1682,
      "num_input_tokens_seen": 42718560,
      "step": 22535
    },
    {
      "epoch": 3.6770535932784076,
      "grad_norm": 0.24334833025932312,
      "learning_rate": 2.0047660018556047e-05,
      "loss": 0.0614,
      "num_input_tokens_seen": 42727984,
      "step": 22540
    },
    {
      "epoch": 3.677869320499225,
      "grad_norm": 0.6895797848701477,
      "learning_rate": 2.0038037475496075e-05,
      "loss": 0.2346,
      "num_input_tokens_seen": 42737968,
      "step": 22545
    },
    {
      "epoch": 3.6786850477200423,
      "grad_norm": 0.45802873373031616,
      "learning_rate": 2.0028415697633073e-05,
      "loss": 0.1847,
      "num_input_tokens_seen": 42747328,
      "step": 22550
    },
    {
      "epoch": 3.67950077494086,
      "grad_norm": 1.4126415252685547,
      "learning_rate": 2.0018794686450858e-05,
      "loss": 0.1878,
      "num_input_tokens_seen": 42755168,
      "step": 22555
    },
    {
      "epoch": 3.680316502161677,
      "grad_norm": 0.07305067777633667,
      "learning_rate": 2.0009174443433088e-05,
      "loss": 0.055,
      "num_input_tokens_seen": 42765680,
      "step": 22560
    },
    {
      "epoch": 3.6811322293824946,
      "grad_norm": 1.0465726852416992,
      "learning_rate": 1.999955497006334e-05,
      "loss": 0.0559,
      "num_input_tokens_seen": 42775024,
      "step": 22565
    },
    {
      "epoch": 3.6819479566033118,
      "grad_norm": 0.6952049732208252,
      "learning_rate": 1.9989936267825067e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 42784464,
      "step": 22570
    },
    {
      "epoch": 3.6827636838241293,
      "grad_norm": 1.4067460298538208,
      "learning_rate": 1.9980318338201572e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 42793168,
      "step": 22575
    },
    {
      "epoch": 3.6835794110449465,
      "grad_norm": 0.15011803805828094,
      "learning_rate": 1.997070118267607e-05,
      "loss": 0.0537,
      "num_input_tokens_seen": 42804160,
      "step": 22580
    },
    {
      "epoch": 3.684395138265764,
      "grad_norm": 0.5446372628211975,
      "learning_rate": 1.9961084802731654e-05,
      "loss": 0.0414,
      "num_input_tokens_seen": 42814464,
      "step": 22585
    },
    {
      "epoch": 3.685210865486581,
      "grad_norm": 0.8810356855392456,
      "learning_rate": 1.9951469199851273e-05,
      "loss": 0.1162,
      "num_input_tokens_seen": 42824848,
      "step": 22590
    },
    {
      "epoch": 3.6860265927073987,
      "grad_norm": 0.8190500140190125,
      "learning_rate": 1.99418543755178e-05,
      "loss": 0.1933,
      "num_input_tokens_seen": 42835248,
      "step": 22595
    },
    {
      "epoch": 3.686842319928216,
      "grad_norm": 0.7922072410583496,
      "learning_rate": 1.9932240331213936e-05,
      "loss": 0.1676,
      "num_input_tokens_seen": 42843696,
      "step": 22600
    },
    {
      "epoch": 3.686842319928216,
      "eval_loss": 0.1442652940750122,
      "eval_runtime": 133.2783,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 42843696,
      "step": 22600
    },
    {
      "epoch": 3.6876580471490334,
      "grad_norm": 0.038881514221429825,
      "learning_rate": 1.9922627068422297e-05,
      "loss": 0.0329,
      "num_input_tokens_seen": 42851664,
      "step": 22605
    },
    {
      "epoch": 3.6884737743698506,
      "grad_norm": 0.7583550810813904,
      "learning_rate": 1.991301458862538e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 42862480,
      "step": 22610
    },
    {
      "epoch": 3.689289501590668,
      "grad_norm": 0.5552899241447449,
      "learning_rate": 1.9903402893305536e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 42871584,
      "step": 22615
    },
    {
      "epoch": 3.6901052288114853,
      "grad_norm": 0.2245475798845291,
      "learning_rate": 1.9893791983945016e-05,
      "loss": 0.0826,
      "num_input_tokens_seen": 42882400,
      "step": 22620
    },
    {
      "epoch": 3.690920956032303,
      "grad_norm": 0.5054088234901428,
      "learning_rate": 1.988418186202594e-05,
      "loss": 0.1877,
      "num_input_tokens_seen": 42892064,
      "step": 22625
    },
    {
      "epoch": 3.6917366832531204,
      "grad_norm": 1.35702645778656,
      "learning_rate": 1.98745725290303e-05,
      "loss": 0.3038,
      "num_input_tokens_seen": 42902720,
      "step": 22630
    },
    {
      "epoch": 3.6925524104739376,
      "grad_norm": 0.212171271443367,
      "learning_rate": 1.986496398644e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 42910928,
      "step": 22635
    },
    {
      "epoch": 3.6933681376947547,
      "grad_norm": 0.6981814503669739,
      "learning_rate": 1.9855356235736777e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 42921472,
      "step": 22640
    },
    {
      "epoch": 3.6941838649155723,
      "grad_norm": 1.0658966302871704,
      "learning_rate": 1.9845749278402277e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 42931328,
      "step": 22645
    },
    {
      "epoch": 3.69499959213639,
      "grad_norm": 1.2456375360488892,
      "learning_rate": 1.9836143115918006e-05,
      "loss": 0.209,
      "num_input_tokens_seen": 42939808,
      "step": 22650
    },
    {
      "epoch": 3.695815319357207,
      "grad_norm": 1.184840440750122,
      "learning_rate": 1.9826537749765367e-05,
      "loss": 0.2163,
      "num_input_tokens_seen": 42949136,
      "step": 22655
    },
    {
      "epoch": 3.696631046578024,
      "grad_norm": 1.4914332628250122,
      "learning_rate": 1.9816933181425625e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 42958720,
      "step": 22660
    },
    {
      "epoch": 3.6974467737988417,
      "grad_norm": 0.4355516731739044,
      "learning_rate": 1.9807329412379903e-05,
      "loss": 0.0824,
      "num_input_tokens_seen": 42968560,
      "step": 22665
    },
    {
      "epoch": 3.6982625010196593,
      "grad_norm": 0.22499388456344604,
      "learning_rate": 1.9797726444109247e-05,
      "loss": 0.0207,
      "num_input_tokens_seen": 42978352,
      "step": 22670
    },
    {
      "epoch": 3.6990782282404764,
      "grad_norm": 0.907474935054779,
      "learning_rate": 1.9788124278094557e-05,
      "loss": 0.0599,
      "num_input_tokens_seen": 42987072,
      "step": 22675
    },
    {
      "epoch": 3.6998939554612935,
      "grad_norm": 0.390603244304657,
      "learning_rate": 1.9778522915816594e-05,
      "loss": 0.1051,
      "num_input_tokens_seen": 42996384,
      "step": 22680
    },
    {
      "epoch": 3.700709682682111,
      "grad_norm": 1.1235615015029907,
      "learning_rate": 1.9768922358756014e-05,
      "loss": 0.1677,
      "num_input_tokens_seen": 43006016,
      "step": 22685
    },
    {
      "epoch": 3.7015254099029287,
      "grad_norm": 0.2782011926174164,
      "learning_rate": 1.9759322608393353e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 43014816,
      "step": 22690
    },
    {
      "epoch": 3.702341137123746,
      "grad_norm": 0.9246197938919067,
      "learning_rate": 1.9749723666208992e-05,
      "loss": 0.2186,
      "num_input_tokens_seen": 43024912,
      "step": 22695
    },
    {
      "epoch": 3.703156864344563,
      "grad_norm": 0.04824531450867653,
      "learning_rate": 1.9740125533683235e-05,
      "loss": 0.2737,
      "num_input_tokens_seen": 43034864,
      "step": 22700
    },
    {
      "epoch": 3.7039725915653805,
      "grad_norm": 0.14588680863380432,
      "learning_rate": 1.9730528212296208e-05,
      "loss": 0.2241,
      "num_input_tokens_seen": 43045168,
      "step": 22705
    },
    {
      "epoch": 3.704788318786198,
      "grad_norm": 0.7310459017753601,
      "learning_rate": 1.9720931703527945e-05,
      "loss": 0.0933,
      "num_input_tokens_seen": 43054208,
      "step": 22710
    },
    {
      "epoch": 3.705604046007015,
      "grad_norm": 1.0379981994628906,
      "learning_rate": 1.9711336008858373e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 43063824,
      "step": 22715
    },
    {
      "epoch": 3.7064197732278323,
      "grad_norm": 0.33693307638168335,
      "learning_rate": 1.9701741129767233e-05,
      "loss": 0.1743,
      "num_input_tokens_seen": 43073872,
      "step": 22720
    },
    {
      "epoch": 3.70723550044865,
      "grad_norm": 1.1611669063568115,
      "learning_rate": 1.9692147067734202e-05,
      "loss": 0.3914,
      "num_input_tokens_seen": 43082672,
      "step": 22725
    },
    {
      "epoch": 3.7080512276694675,
      "grad_norm": 0.386863648891449,
      "learning_rate": 1.96825538242388e-05,
      "loss": 0.0972,
      "num_input_tokens_seen": 43092384,
      "step": 22730
    },
    {
      "epoch": 3.7088669548902846,
      "grad_norm": 0.13261811435222626,
      "learning_rate": 1.967296140076041e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 43101408,
      "step": 22735
    },
    {
      "epoch": 3.709682682111102,
      "grad_norm": 1.7315073013305664,
      "learning_rate": 1.966336979877833e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 43111344,
      "step": 22740
    },
    {
      "epoch": 3.7104984093319193,
      "grad_norm": 0.5063236951828003,
      "learning_rate": 1.9653779019771678e-05,
      "loss": 0.1364,
      "num_input_tokens_seen": 43119344,
      "step": 22745
    },
    {
      "epoch": 3.711314136552737,
      "grad_norm": 0.24609622359275818,
      "learning_rate": 1.9644189065219488e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 43129152,
      "step": 22750
    },
    {
      "epoch": 3.712129863773554,
      "grad_norm": 0.053422681987285614,
      "learning_rate": 1.9634599936600655e-05,
      "loss": 0.07,
      "num_input_tokens_seen": 43138080,
      "step": 22755
    },
    {
      "epoch": 3.7129455909943716,
      "grad_norm": 0.10405254364013672,
      "learning_rate": 1.9625011635393935e-05,
      "loss": 0.0831,
      "num_input_tokens_seen": 43146624,
      "step": 22760
    },
    {
      "epoch": 3.7137613182151887,
      "grad_norm": 0.7285462021827698,
      "learning_rate": 1.9615424163077963e-05,
      "loss": 0.0812,
      "num_input_tokens_seen": 43155456,
      "step": 22765
    },
    {
      "epoch": 3.7145770454360063,
      "grad_norm": 0.792094886302948,
      "learning_rate": 1.9605837521131263e-05,
      "loss": 0.2295,
      "num_input_tokens_seen": 43164800,
      "step": 22770
    },
    {
      "epoch": 3.7153927726568234,
      "grad_norm": 0.14230966567993164,
      "learning_rate": 1.9596251711032192e-05,
      "loss": 0.1129,
      "num_input_tokens_seen": 43174960,
      "step": 22775
    },
    {
      "epoch": 3.716208499877641,
      "grad_norm": 0.16100344061851501,
      "learning_rate": 1.958666673425903e-05,
      "loss": 0.0671,
      "num_input_tokens_seen": 43183488,
      "step": 22780
    },
    {
      "epoch": 3.717024227098458,
      "grad_norm": 0.7069581747055054,
      "learning_rate": 1.957708259228987e-05,
      "loss": 0.0384,
      "num_input_tokens_seen": 43194192,
      "step": 22785
    },
    {
      "epoch": 3.7178399543192757,
      "grad_norm": 1.1963592767715454,
      "learning_rate": 1.956749928660273e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 43202528,
      "step": 22790
    },
    {
      "epoch": 3.718655681540093,
      "grad_norm": 0.7193416357040405,
      "learning_rate": 1.955791681867547e-05,
      "loss": 0.2037,
      "num_input_tokens_seen": 43211664,
      "step": 22795
    },
    {
      "epoch": 3.7194714087609104,
      "grad_norm": 1.0497053861618042,
      "learning_rate": 1.9548335189985824e-05,
      "loss": 0.1843,
      "num_input_tokens_seen": 43221120,
      "step": 22800
    },
    {
      "epoch": 3.7194714087609104,
      "eval_loss": 0.14296412467956543,
      "eval_runtime": 133.3428,
      "eval_samples_per_second": 20.436,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 43221120,
      "step": 22800
    },
    {
      "epoch": 3.7202871359817276,
      "grad_norm": 0.0438268668949604,
      "learning_rate": 1.9538754402011396e-05,
      "loss": 0.1106,
      "num_input_tokens_seen": 43231152,
      "step": 22805
    },
    {
      "epoch": 3.721102863202545,
      "grad_norm": 1.029457688331604,
      "learning_rate": 1.952917445622968e-05,
      "loss": 0.287,
      "num_input_tokens_seen": 43240384,
      "step": 22810
    },
    {
      "epoch": 3.7219185904233623,
      "grad_norm": 0.6279735565185547,
      "learning_rate": 1.9519595354118005e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 43249712,
      "step": 22815
    },
    {
      "epoch": 3.72273431764418,
      "grad_norm": 1.1751487255096436,
      "learning_rate": 1.951001709715361e-05,
      "loss": 0.1591,
      "num_input_tokens_seen": 43259136,
      "step": 22820
    },
    {
      "epoch": 3.7235500448649974,
      "grad_norm": 0.17066650092601776,
      "learning_rate": 1.9500439686813556e-05,
      "loss": 0.0484,
      "num_input_tokens_seen": 43269408,
      "step": 22825
    },
    {
      "epoch": 3.7243657720858145,
      "grad_norm": 0.2647336721420288,
      "learning_rate": 1.949086312457482e-05,
      "loss": 0.0568,
      "num_input_tokens_seen": 43280288,
      "step": 22830
    },
    {
      "epoch": 3.7251814993066317,
      "grad_norm": 0.958854615688324,
      "learning_rate": 1.9481287411914223e-05,
      "loss": 0.1109,
      "num_input_tokens_seen": 43289856,
      "step": 22835
    },
    {
      "epoch": 3.7259972265274492,
      "grad_norm": 0.7304872870445251,
      "learning_rate": 1.9471712550308457e-05,
      "loss": 0.1323,
      "num_input_tokens_seen": 43300368,
      "step": 22840
    },
    {
      "epoch": 3.726812953748267,
      "grad_norm": 0.18392875790596008,
      "learning_rate": 1.946213854123409e-05,
      "loss": 0.2024,
      "num_input_tokens_seen": 43309664,
      "step": 22845
    },
    {
      "epoch": 3.727628680969084,
      "grad_norm": 1.0111522674560547,
      "learning_rate": 1.9452565386167554e-05,
      "loss": 0.1053,
      "num_input_tokens_seen": 43319152,
      "step": 22850
    },
    {
      "epoch": 3.728444408189901,
      "grad_norm": 1.8569986820220947,
      "learning_rate": 1.9442993086585142e-05,
      "loss": 0.1976,
      "num_input_tokens_seen": 43328160,
      "step": 22855
    },
    {
      "epoch": 3.7292601354107187,
      "grad_norm": 0.0927785336971283,
      "learning_rate": 1.9433421643963043e-05,
      "loss": 0.1108,
      "num_input_tokens_seen": 43337648,
      "step": 22860
    },
    {
      "epoch": 3.7300758626315362,
      "grad_norm": 0.13145023584365845,
      "learning_rate": 1.942385105977727e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 43347872,
      "step": 22865
    },
    {
      "epoch": 3.7308915898523534,
      "grad_norm": 1.392690896987915,
      "learning_rate": 1.9414281335503743e-05,
      "loss": 0.2471,
      "num_input_tokens_seen": 43357728,
      "step": 22870
    },
    {
      "epoch": 3.7317073170731705,
      "grad_norm": 0.13904500007629395,
      "learning_rate": 1.9404712472618232e-05,
      "loss": 0.0844,
      "num_input_tokens_seen": 43366544,
      "step": 22875
    },
    {
      "epoch": 3.732523044293988,
      "grad_norm": 2.464387893676758,
      "learning_rate": 1.939514447259636e-05,
      "loss": 0.2558,
      "num_input_tokens_seen": 43375824,
      "step": 22880
    },
    {
      "epoch": 3.7333387715148056,
      "grad_norm": 1.1649115085601807,
      "learning_rate": 1.938557733691365e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 43385376,
      "step": 22885
    },
    {
      "epoch": 3.7341544987356228,
      "grad_norm": 0.7213959097862244,
      "learning_rate": 1.9376011067045476e-05,
      "loss": 0.2262,
      "num_input_tokens_seen": 43394320,
      "step": 22890
    },
    {
      "epoch": 3.73497022595644,
      "grad_norm": 0.5462323427200317,
      "learning_rate": 1.9366445664467065e-05,
      "loss": 0.1895,
      "num_input_tokens_seen": 43402688,
      "step": 22895
    },
    {
      "epoch": 3.7357859531772575,
      "grad_norm": 0.19455894827842712,
      "learning_rate": 1.9356881130653533e-05,
      "loss": 0.1668,
      "num_input_tokens_seen": 43411872,
      "step": 22900
    },
    {
      "epoch": 3.736601680398075,
      "grad_norm": 1.2145687341690063,
      "learning_rate": 1.9347317467079846e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 43420944,
      "step": 22905
    },
    {
      "epoch": 3.737417407618892,
      "grad_norm": 0.07079560309648514,
      "learning_rate": 1.9337754675220836e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 43429824,
      "step": 22910
    },
    {
      "epoch": 3.7382331348397098,
      "grad_norm": 0.8979502320289612,
      "learning_rate": 1.9328192756551218e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 43438576,
      "step": 22915
    },
    {
      "epoch": 3.739048862060527,
      "grad_norm": 0.28209570050239563,
      "learning_rate": 1.931863171254555e-05,
      "loss": 0.0372,
      "num_input_tokens_seen": 43448576,
      "step": 22920
    },
    {
      "epoch": 3.7398645892813445,
      "grad_norm": 0.9143474102020264,
      "learning_rate": 1.930907154467826e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 43458224,
      "step": 22925
    },
    {
      "epoch": 3.7406803165021616,
      "grad_norm": 0.36810827255249023,
      "learning_rate": 1.9299512254423673e-05,
      "loss": 0.0985,
      "num_input_tokens_seen": 43465888,
      "step": 22930
    },
    {
      "epoch": 3.741496043722979,
      "grad_norm": 0.5659452676773071,
      "learning_rate": 1.9289953843255914e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 43475216,
      "step": 22935
    },
    {
      "epoch": 3.7423117709437963,
      "grad_norm": 0.9974454045295715,
      "learning_rate": 1.9280396312649048e-05,
      "loss": 0.1007,
      "num_input_tokens_seen": 43484992,
      "step": 22940
    },
    {
      "epoch": 3.743127498164614,
      "grad_norm": 0.5526713728904724,
      "learning_rate": 1.9270839664076936e-05,
      "loss": 0.0495,
      "num_input_tokens_seen": 43493712,
      "step": 22945
    },
    {
      "epoch": 3.743943225385431,
      "grad_norm": 0.46489986777305603,
      "learning_rate": 1.9261283899013345e-05,
      "loss": 0.126,
      "num_input_tokens_seen": 43502880,
      "step": 22950
    },
    {
      "epoch": 3.7447589526062486,
      "grad_norm": 0.26596638560295105,
      "learning_rate": 1.92517290189319e-05,
      "loss": 0.037,
      "num_input_tokens_seen": 43511920,
      "step": 22955
    },
    {
      "epoch": 3.7455746798270657,
      "grad_norm": 0.4020860195159912,
      "learning_rate": 1.924217502530607e-05,
      "loss": 0.1081,
      "num_input_tokens_seen": 43521312,
      "step": 22960
    },
    {
      "epoch": 3.7463904070478833,
      "grad_norm": 0.23657365143299103,
      "learning_rate": 1.9232621919609207e-05,
      "loss": 0.1049,
      "num_input_tokens_seen": 43531680,
      "step": 22965
    },
    {
      "epoch": 3.7472061342687004,
      "grad_norm": 1.0555493831634521,
      "learning_rate": 1.9223069703314534e-05,
      "loss": 0.048,
      "num_input_tokens_seen": 43541008,
      "step": 22970
    },
    {
      "epoch": 3.748021861489518,
      "grad_norm": 0.18933618068695068,
      "learning_rate": 1.92135183778951e-05,
      "loss": 0.0274,
      "num_input_tokens_seen": 43551168,
      "step": 22975
    },
    {
      "epoch": 3.748837588710335,
      "grad_norm": 0.17745551466941833,
      "learning_rate": 1.9203967944823857e-05,
      "loss": 0.1665,
      "num_input_tokens_seen": 43560640,
      "step": 22980
    },
    {
      "epoch": 3.7496533159311527,
      "grad_norm": 0.3755565881729126,
      "learning_rate": 1.9194418405573588e-05,
      "loss": 0.0663,
      "num_input_tokens_seen": 43569472,
      "step": 22985
    },
    {
      "epoch": 3.75046904315197,
      "grad_norm": 0.2985245883464813,
      "learning_rate": 1.9184869761616954e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 43578720,
      "step": 22990
    },
    {
      "epoch": 3.7512847703727874,
      "grad_norm": 1.6237298250198364,
      "learning_rate": 1.9175322014426495e-05,
      "loss": 0.3683,
      "num_input_tokens_seen": 43588112,
      "step": 22995
    },
    {
      "epoch": 3.7521004975936045,
      "grad_norm": 0.744203507900238,
      "learning_rate": 1.9165775165474565e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 43597776,
      "step": 23000
    },
    {
      "epoch": 3.7521004975936045,
      "eval_loss": 0.1420077234506607,
      "eval_runtime": 133.3006,
      "eval_samples_per_second": 20.443,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 43597776,
      "step": 23000
    },
    {
      "epoch": 3.752916224814422,
      "grad_norm": 0.08608266711235046,
      "learning_rate": 1.9156229216233434e-05,
      "loss": 0.0723,
      "num_input_tokens_seen": 43606992,
      "step": 23005
    },
    {
      "epoch": 3.7537319520352392,
      "grad_norm": 0.1558219939470291,
      "learning_rate": 1.9146684168175184e-05,
      "loss": 0.0452,
      "num_input_tokens_seen": 43617072,
      "step": 23010
    },
    {
      "epoch": 3.754547679256057,
      "grad_norm": 0.6023638844490051,
      "learning_rate": 1.9137140022771796e-05,
      "loss": 0.073,
      "num_input_tokens_seen": 43627424,
      "step": 23015
    },
    {
      "epoch": 3.7553634064768744,
      "grad_norm": 1.4207044839859009,
      "learning_rate": 1.9127596781495103e-05,
      "loss": 0.0937,
      "num_input_tokens_seen": 43637728,
      "step": 23020
    },
    {
      "epoch": 3.7561791336976915,
      "grad_norm": 0.10047546029090881,
      "learning_rate": 1.9118054445816767e-05,
      "loss": 0.1124,
      "num_input_tokens_seen": 43646784,
      "step": 23025
    },
    {
      "epoch": 3.7569948609185086,
      "grad_norm": 1.4219563007354736,
      "learning_rate": 1.9108513017208356e-05,
      "loss": 0.1394,
      "num_input_tokens_seen": 43656416,
      "step": 23030
    },
    {
      "epoch": 3.7578105881393262,
      "grad_norm": 0.22592194378376007,
      "learning_rate": 1.9098972497141287e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 43665632,
      "step": 23035
    },
    {
      "epoch": 3.758626315360144,
      "grad_norm": 0.4525085389614105,
      "learning_rate": 1.9089432887086806e-05,
      "loss": 0.1814,
      "num_input_tokens_seen": 43674624,
      "step": 23040
    },
    {
      "epoch": 3.759442042580961,
      "grad_norm": 1.5226702690124512,
      "learning_rate": 1.9079894188516056e-05,
      "loss": 0.2649,
      "num_input_tokens_seen": 43684416,
      "step": 23045
    },
    {
      "epoch": 3.760257769801778,
      "grad_norm": 0.23534975945949554,
      "learning_rate": 1.907035640290002e-05,
      "loss": 0.0927,
      "num_input_tokens_seen": 43692928,
      "step": 23050
    },
    {
      "epoch": 3.7610734970225956,
      "grad_norm": 1.0802618265151978,
      "learning_rate": 1.9060819531709534e-05,
      "loss": 0.1965,
      "num_input_tokens_seen": 43703456,
      "step": 23055
    },
    {
      "epoch": 3.761889224243413,
      "grad_norm": 0.14761096239089966,
      "learning_rate": 1.9051283576415325e-05,
      "loss": 0.0906,
      "num_input_tokens_seen": 43711632,
      "step": 23060
    },
    {
      "epoch": 3.7627049514642303,
      "grad_norm": 1.3662430047988892,
      "learning_rate": 1.904174853848793e-05,
      "loss": 0.2001,
      "num_input_tokens_seen": 43721008,
      "step": 23065
    },
    {
      "epoch": 3.7635206786850475,
      "grad_norm": 0.07849553227424622,
      "learning_rate": 1.903221441939779e-05,
      "loss": 0.0608,
      "num_input_tokens_seen": 43730736,
      "step": 23070
    },
    {
      "epoch": 3.764336405905865,
      "grad_norm": 1.621938943862915,
      "learning_rate": 1.9022681220615194e-05,
      "loss": 0.245,
      "num_input_tokens_seen": 43741072,
      "step": 23075
    },
    {
      "epoch": 3.7651521331266826,
      "grad_norm": 0.17963460087776184,
      "learning_rate": 1.9013148943610255e-05,
      "loss": 0.2103,
      "num_input_tokens_seen": 43751296,
      "step": 23080
    },
    {
      "epoch": 3.7659678603474998,
      "grad_norm": 2.15437388420105,
      "learning_rate": 1.9003617589852998e-05,
      "loss": 0.2078,
      "num_input_tokens_seen": 43760304,
      "step": 23085
    },
    {
      "epoch": 3.766783587568317,
      "grad_norm": 0.38541075587272644,
      "learning_rate": 1.899408716081326e-05,
      "loss": 0.2959,
      "num_input_tokens_seen": 43769808,
      "step": 23090
    },
    {
      "epoch": 3.7675993147891345,
      "grad_norm": 0.9803973436355591,
      "learning_rate": 1.898455765796075e-05,
      "loss": 0.1794,
      "num_input_tokens_seen": 43778528,
      "step": 23095
    },
    {
      "epoch": 3.768415042009952,
      "grad_norm": 0.34253618121147156,
      "learning_rate": 1.8975029082765053e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 43789632,
      "step": 23100
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 0.8924728631973267,
      "learning_rate": 1.8965501436695577e-05,
      "loss": 0.0751,
      "num_input_tokens_seen": 43799344,
      "step": 23105
    },
    {
      "epoch": 3.7700464964515867,
      "grad_norm": 0.5969791412353516,
      "learning_rate": 1.895597472122161e-05,
      "loss": 0.0863,
      "num_input_tokens_seen": 43808656,
      "step": 23110
    },
    {
      "epoch": 3.770862223672404,
      "grad_norm": 0.19809108972549438,
      "learning_rate": 1.894644893781231e-05,
      "loss": 0.1739,
      "num_input_tokens_seen": 43817504,
      "step": 23115
    },
    {
      "epoch": 3.7716779508932214,
      "grad_norm": 0.5675562024116516,
      "learning_rate": 1.893692408793665e-05,
      "loss": 0.0543,
      "num_input_tokens_seen": 43827200,
      "step": 23120
    },
    {
      "epoch": 3.7724936781140386,
      "grad_norm": 0.6486147046089172,
      "learning_rate": 1.8927400173063493e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 43836976,
      "step": 23125
    },
    {
      "epoch": 3.773309405334856,
      "grad_norm": 1.449682593345642,
      "learning_rate": 1.891787719466154e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 43846768,
      "step": 23130
    },
    {
      "epoch": 3.7741251325556733,
      "grad_norm": 0.061908021569252014,
      "learning_rate": 1.8908355154199346e-05,
      "loss": 0.215,
      "num_input_tokens_seen": 43855056,
      "step": 23135
    },
    {
      "epoch": 3.774940859776491,
      "grad_norm": 1.318931221961975,
      "learning_rate": 1.8898834053145357e-05,
      "loss": 0.1878,
      "num_input_tokens_seen": 43864032,
      "step": 23140
    },
    {
      "epoch": 3.775756586997308,
      "grad_norm": 0.6429188251495361,
      "learning_rate": 1.8889313892967813e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 43874160,
      "step": 23145
    },
    {
      "epoch": 3.7765723142181256,
      "grad_norm": 0.8795682191848755,
      "learning_rate": 1.8879794675134863e-05,
      "loss": 0.2069,
      "num_input_tokens_seen": 43883200,
      "step": 23150
    },
    {
      "epoch": 3.7773880414389427,
      "grad_norm": 0.2853083312511444,
      "learning_rate": 1.8870276401114494e-05,
      "loss": 0.0923,
      "num_input_tokens_seen": 43893152,
      "step": 23155
    },
    {
      "epoch": 3.7782037686597603,
      "grad_norm": 0.48923638463020325,
      "learning_rate": 1.886075907237453e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 43903408,
      "step": 23160
    },
    {
      "epoch": 3.7790194958805774,
      "grad_norm": 0.5133200287818909,
      "learning_rate": 1.8851242690382672e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 43913568,
      "step": 23165
    },
    {
      "epoch": 3.779835223101395,
      "grad_norm": 0.959347665309906,
      "learning_rate": 1.884172725660645e-05,
      "loss": 0.0627,
      "num_input_tokens_seen": 43923296,
      "step": 23170
    },
    {
      "epoch": 3.780650950322212,
      "grad_norm": 1.6870814561843872,
      "learning_rate": 1.8832212772513277e-05,
      "loss": 0.1984,
      "num_input_tokens_seen": 43932880,
      "step": 23175
    },
    {
      "epoch": 3.7814666775430297,
      "grad_norm": 0.1177358627319336,
      "learning_rate": 1.8822699239570414e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 43942448,
      "step": 23180
    },
    {
      "epoch": 3.782282404763847,
      "grad_norm": 2.085028886795044,
      "learning_rate": 1.8813186659244943e-05,
      "loss": 0.2168,
      "num_input_tokens_seen": 43951376,
      "step": 23185
    },
    {
      "epoch": 3.7830981319846644,
      "grad_norm": 0.17428331077098846,
      "learning_rate": 1.880367503300385e-05,
      "loss": 0.0591,
      "num_input_tokens_seen": 43960688,
      "step": 23190
    },
    {
      "epoch": 3.783913859205482,
      "grad_norm": 1.4494264125823975,
      "learning_rate": 1.8794164362313927e-05,
      "loss": 0.0951,
      "num_input_tokens_seen": 43969728,
      "step": 23195
    },
    {
      "epoch": 3.784729586426299,
      "grad_norm": 0.1067870706319809,
      "learning_rate": 1.878465464864185e-05,
      "loss": 0.0701,
      "num_input_tokens_seen": 43979312,
      "step": 23200
    },
    {
      "epoch": 3.784729586426299,
      "eval_loss": 0.14252381026744843,
      "eval_runtime": 133.3428,
      "eval_samples_per_second": 20.436,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 43979312,
      "step": 23200
    },
    {
      "epoch": 3.785545313647116,
      "grad_norm": 0.29020756483078003,
      "learning_rate": 1.877514589345414e-05,
      "loss": 0.1654,
      "num_input_tokens_seen": 43989600,
      "step": 23205
    },
    {
      "epoch": 3.786361040867934,
      "grad_norm": 1.412105917930603,
      "learning_rate": 1.876563809821715e-05,
      "loss": 0.1851,
      "num_input_tokens_seen": 43999376,
      "step": 23210
    },
    {
      "epoch": 3.7871767680887514,
      "grad_norm": 0.30276867747306824,
      "learning_rate": 1.8756131264397106e-05,
      "loss": 0.0597,
      "num_input_tokens_seen": 44008400,
      "step": 23215
    },
    {
      "epoch": 3.7879924953095685,
      "grad_norm": 0.12904570996761322,
      "learning_rate": 1.87466253934601e-05,
      "loss": 0.0553,
      "num_input_tokens_seen": 44017520,
      "step": 23220
    },
    {
      "epoch": 3.7888082225303856,
      "grad_norm": 1.6374077796936035,
      "learning_rate": 1.8737120486872033e-05,
      "loss": 0.2538,
      "num_input_tokens_seen": 44027424,
      "step": 23225
    },
    {
      "epoch": 3.789623949751203,
      "grad_norm": 0.030106930062174797,
      "learning_rate": 1.8727616546098696e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 44037312,
      "step": 23230
    },
    {
      "epoch": 3.7904396769720208,
      "grad_norm": 0.23933306336402893,
      "learning_rate": 1.8718113572605716e-05,
      "loss": 0.1146,
      "num_input_tokens_seen": 44047488,
      "step": 23235
    },
    {
      "epoch": 3.791255404192838,
      "grad_norm": 0.7050015330314636,
      "learning_rate": 1.8708611567858554e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 44057232,
      "step": 23240
    },
    {
      "epoch": 3.792071131413655,
      "grad_norm": 0.32146841287612915,
      "learning_rate": 1.8699110533322565e-05,
      "loss": 0.2788,
      "num_input_tokens_seen": 44066704,
      "step": 23245
    },
    {
      "epoch": 3.7928868586344726,
      "grad_norm": 0.17800915241241455,
      "learning_rate": 1.8689610470462897e-05,
      "loss": 0.1086,
      "num_input_tokens_seen": 44076432,
      "step": 23250
    },
    {
      "epoch": 3.79370258585529,
      "grad_norm": 0.051367245614528656,
      "learning_rate": 1.8680111380744604e-05,
      "loss": 0.0613,
      "num_input_tokens_seen": 44084848,
      "step": 23255
    },
    {
      "epoch": 3.7945183130761073,
      "grad_norm": 0.23992693424224854,
      "learning_rate": 1.8670613265632564e-05,
      "loss": 0.055,
      "num_input_tokens_seen": 44094208,
      "step": 23260
    },
    {
      "epoch": 3.7953340402969244,
      "grad_norm": 0.5351288914680481,
      "learning_rate": 1.866111612659149e-05,
      "loss": 0.1127,
      "num_input_tokens_seen": 44103904,
      "step": 23265
    },
    {
      "epoch": 3.796149767517742,
      "grad_norm": 1.5230700969696045,
      "learning_rate": 1.8651619965085967e-05,
      "loss": 0.097,
      "num_input_tokens_seen": 44113760,
      "step": 23270
    },
    {
      "epoch": 3.7969654947385596,
      "grad_norm": 0.6104052662849426,
      "learning_rate": 1.8642124782580433e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 44123312,
      "step": 23275
    },
    {
      "epoch": 3.7977812219593767,
      "grad_norm": 0.07068760693073273,
      "learning_rate": 1.8632630580539144e-05,
      "loss": 0.0519,
      "num_input_tokens_seen": 44131984,
      "step": 23280
    },
    {
      "epoch": 3.7985969491801943,
      "grad_norm": 0.6707145571708679,
      "learning_rate": 1.862313736042625e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 44141504,
      "step": 23285
    },
    {
      "epoch": 3.7994126764010114,
      "grad_norm": 1.0671560764312744,
      "learning_rate": 1.8613645123705703e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 44151120,
      "step": 23290
    },
    {
      "epoch": 3.800228403621829,
      "grad_norm": 0.28607624769210815,
      "learning_rate": 1.8604153871841328e-05,
      "loss": 0.079,
      "num_input_tokens_seen": 44160752,
      "step": 23295
    },
    {
      "epoch": 3.801044130842646,
      "grad_norm": 1.4068570137023926,
      "learning_rate": 1.859466360629682e-05,
      "loss": 0.1777,
      "num_input_tokens_seen": 44169072,
      "step": 23300
    },
    {
      "epoch": 3.8018598580634637,
      "grad_norm": 0.6004501581192017,
      "learning_rate": 1.8585174328535666e-05,
      "loss": 0.0913,
      "num_input_tokens_seen": 44178528,
      "step": 23305
    },
    {
      "epoch": 3.802675585284281,
      "grad_norm": 0.31071197986602783,
      "learning_rate": 1.857568604002124e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 44189536,
      "step": 23310
    },
    {
      "epoch": 3.8034913125050984,
      "grad_norm": 0.305038720369339,
      "learning_rate": 1.8566198742216774e-05,
      "loss": 0.2527,
      "num_input_tokens_seen": 44198144,
      "step": 23315
    },
    {
      "epoch": 3.8043070397259156,
      "grad_norm": 0.03831874579191208,
      "learning_rate": 1.85567124365853e-05,
      "loss": 0.185,
      "num_input_tokens_seen": 44207168,
      "step": 23320
    },
    {
      "epoch": 3.805122766946733,
      "grad_norm": 1.3911106586456299,
      "learning_rate": 1.854722712458975e-05,
      "loss": 0.186,
      "num_input_tokens_seen": 44216816,
      "step": 23325
    },
    {
      "epoch": 3.8059384941675503,
      "grad_norm": 0.1034315899014473,
      "learning_rate": 1.853774280769286e-05,
      "loss": 0.0741,
      "num_input_tokens_seen": 44225632,
      "step": 23330
    },
    {
      "epoch": 3.806754221388368,
      "grad_norm": 0.7822731733322144,
      "learning_rate": 1.852825948735724e-05,
      "loss": 0.0863,
      "num_input_tokens_seen": 44234928,
      "step": 23335
    },
    {
      "epoch": 3.807569948609185,
      "grad_norm": 1.594244122505188,
      "learning_rate": 1.851877716504534e-05,
      "loss": 0.2477,
      "num_input_tokens_seen": 44244224,
      "step": 23340
    },
    {
      "epoch": 3.8083856758300025,
      "grad_norm": 2.1831278800964355,
      "learning_rate": 1.8509295842219448e-05,
      "loss": 0.2483,
      "num_input_tokens_seen": 44253248,
      "step": 23345
    },
    {
      "epoch": 3.8092014030508197,
      "grad_norm": 0.5536495447158813,
      "learning_rate": 1.8499815520341697e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 44261168,
      "step": 23350
    },
    {
      "epoch": 3.8100171302716372,
      "grad_norm": 0.786807119846344,
      "learning_rate": 1.8490336200874094e-05,
      "loss": 0.2429,
      "num_input_tokens_seen": 44270832,
      "step": 23355
    },
    {
      "epoch": 3.8108328574924544,
      "grad_norm": 0.3807106614112854,
      "learning_rate": 1.848085788527844e-05,
      "loss": 0.0365,
      "num_input_tokens_seen": 44280896,
      "step": 23360
    },
    {
      "epoch": 3.811648584713272,
      "grad_norm": 1.6105273962020874,
      "learning_rate": 1.847138057501644e-05,
      "loss": 0.2505,
      "num_input_tokens_seen": 44291280,
      "step": 23365
    },
    {
      "epoch": 3.812464311934089,
      "grad_norm": 1.4780445098876953,
      "learning_rate": 1.8461904271549582e-05,
      "loss": 0.1433,
      "num_input_tokens_seen": 44300400,
      "step": 23370
    },
    {
      "epoch": 3.8132800391549067,
      "grad_norm": 0.3231419622898102,
      "learning_rate": 1.845242897633926e-05,
      "loss": 0.1769,
      "num_input_tokens_seen": 44309488,
      "step": 23375
    },
    {
      "epoch": 3.814095766375724,
      "grad_norm": 0.9976809024810791,
      "learning_rate": 1.844295469084667e-05,
      "loss": 0.0727,
      "num_input_tokens_seen": 44318944,
      "step": 23380
    },
    {
      "epoch": 3.8149114935965414,
      "grad_norm": 0.29282405972480774,
      "learning_rate": 1.843348141653286e-05,
      "loss": 0.0409,
      "num_input_tokens_seen": 44328112,
      "step": 23385
    },
    {
      "epoch": 3.815727220817359,
      "grad_norm": 1.0765663385391235,
      "learning_rate": 1.842400915485874e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 44336960,
      "step": 23390
    },
    {
      "epoch": 3.816542948038176,
      "grad_norm": 0.10613060742616653,
      "learning_rate": 1.8414537907285053e-05,
      "loss": 0.1241,
      "num_input_tokens_seen": 44345600,
      "step": 23395
    },
    {
      "epoch": 3.817358675258993,
      "grad_norm": 0.20921875536441803,
      "learning_rate": 1.840506767527237e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 44354480,
      "step": 23400
    },
    {
      "epoch": 3.817358675258993,
      "eval_loss": 0.1432403177022934,
      "eval_runtime": 133.3852,
      "eval_samples_per_second": 20.43,
      "eval_steps_per_second": 10.219,
      "num_input_tokens_seen": 44354480,
      "step": 23400
    },
    {
      "epoch": 3.8181744024798108,
      "grad_norm": 0.20582039654254913,
      "learning_rate": 1.8395598460281137e-05,
      "loss": 0.0618,
      "num_input_tokens_seen": 44363680,
      "step": 23405
    },
    {
      "epoch": 3.8189901297006283,
      "grad_norm": 0.47556817531585693,
      "learning_rate": 1.838613026377161e-05,
      "loss": 0.0605,
      "num_input_tokens_seen": 44372704,
      "step": 23410
    },
    {
      "epoch": 3.8198058569214455,
      "grad_norm": 0.21163108944892883,
      "learning_rate": 1.8376663087203917e-05,
      "loss": 0.0346,
      "num_input_tokens_seen": 44381104,
      "step": 23415
    },
    {
      "epoch": 3.8206215841422626,
      "grad_norm": 0.4935641586780548,
      "learning_rate": 1.8367196932038014e-05,
      "loss": 0.0545,
      "num_input_tokens_seen": 44388736,
      "step": 23420
    },
    {
      "epoch": 3.82143731136308,
      "grad_norm": 0.40800637006759644,
      "learning_rate": 1.8357731799733686e-05,
      "loss": 0.1114,
      "num_input_tokens_seen": 44398160,
      "step": 23425
    },
    {
      "epoch": 3.8222530385838978,
      "grad_norm": 0.14537736773490906,
      "learning_rate": 1.8348267691750586e-05,
      "loss": 0.12,
      "num_input_tokens_seen": 44408176,
      "step": 23430
    },
    {
      "epoch": 3.823068765804715,
      "grad_norm": 0.20041507482528687,
      "learning_rate": 1.833880460954821e-05,
      "loss": 0.0517,
      "num_input_tokens_seen": 44418496,
      "step": 23435
    },
    {
      "epoch": 3.823884493025532,
      "grad_norm": 0.6338340044021606,
      "learning_rate": 1.8329342554585866e-05,
      "loss": 0.0267,
      "num_input_tokens_seen": 44429584,
      "step": 23440
    },
    {
      "epoch": 3.8247002202463496,
      "grad_norm": 1.4756907224655151,
      "learning_rate": 1.8319881528322735e-05,
      "loss": 0.209,
      "num_input_tokens_seen": 44439568,
      "step": 23445
    },
    {
      "epoch": 3.825515947467167,
      "grad_norm": 0.3698936998844147,
      "learning_rate": 1.8310421532217815e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 44449264,
      "step": 23450
    },
    {
      "epoch": 3.8263316746879843,
      "grad_norm": 0.29431354999542236,
      "learning_rate": 1.8300962567729958e-05,
      "loss": 0.0782,
      "num_input_tokens_seen": 44458368,
      "step": 23455
    },
    {
      "epoch": 3.8271474019088014,
      "grad_norm": 0.11769890040159225,
      "learning_rate": 1.8291504636317866e-05,
      "loss": 0.0923,
      "num_input_tokens_seen": 44467136,
      "step": 23460
    },
    {
      "epoch": 3.827963129129619,
      "grad_norm": 0.16620194911956787,
      "learning_rate": 1.8282047739440055e-05,
      "loss": 0.1468,
      "num_input_tokens_seen": 44477584,
      "step": 23465
    },
    {
      "epoch": 3.8287788563504366,
      "grad_norm": 0.336201936006546,
      "learning_rate": 1.8272591878554903e-05,
      "loss": 0.054,
      "num_input_tokens_seen": 44487392,
      "step": 23470
    },
    {
      "epoch": 3.8295945835712537,
      "grad_norm": 0.09760815650224686,
      "learning_rate": 1.8263137055120638e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 44496528,
      "step": 23475
    },
    {
      "epoch": 3.8304103107920713,
      "grad_norm": 0.934331476688385,
      "learning_rate": 1.8253683270595295e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 44504880,
      "step": 23480
    },
    {
      "epoch": 3.8312260380128884,
      "grad_norm": 0.2996531128883362,
      "learning_rate": 1.824423052643677e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 44514464,
      "step": 23485
    },
    {
      "epoch": 3.832041765233706,
      "grad_norm": 0.3073979616165161,
      "learning_rate": 1.82347788241028e-05,
      "loss": 0.1889,
      "num_input_tokens_seen": 44523792,
      "step": 23490
    },
    {
      "epoch": 3.832857492454523,
      "grad_norm": 1.2948907613754272,
      "learning_rate": 1.8225328165050942e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 44532944,
      "step": 23495
    },
    {
      "epoch": 3.8336732196753407,
      "grad_norm": 1.0033329725265503,
      "learning_rate": 1.821587855073863e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 44542608,
      "step": 23500
    },
    {
      "epoch": 3.834488946896158,
      "grad_norm": 1.0136942863464355,
      "learning_rate": 1.8206429982623086e-05,
      "loss": 0.1031,
      "num_input_tokens_seen": 44552000,
      "step": 23505
    },
    {
      "epoch": 3.8353046741169754,
      "grad_norm": 0.5913650393486023,
      "learning_rate": 1.8196982462161416e-05,
      "loss": 0.1414,
      "num_input_tokens_seen": 44560976,
      "step": 23510
    },
    {
      "epoch": 3.8361204013377925,
      "grad_norm": 0.14520664513111115,
      "learning_rate": 1.818753599081055e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 44570256,
      "step": 23515
    },
    {
      "epoch": 3.83693612855861,
      "grad_norm": 0.4737025797367096,
      "learning_rate": 1.817809057002724e-05,
      "loss": 0.0396,
      "num_input_tokens_seen": 44579968,
      "step": 23520
    },
    {
      "epoch": 3.8377518557794272,
      "grad_norm": 1.5621808767318726,
      "learning_rate": 1.8168646201268096e-05,
      "loss": 0.1919,
      "num_input_tokens_seen": 44587680,
      "step": 23525
    },
    {
      "epoch": 3.838567583000245,
      "grad_norm": 1.9812499284744263,
      "learning_rate": 1.8159202885989557e-05,
      "loss": 0.1211,
      "num_input_tokens_seen": 44596112,
      "step": 23530
    },
    {
      "epoch": 3.839383310221062,
      "grad_norm": 1.574511170387268,
      "learning_rate": 1.814976062564789e-05,
      "loss": 0.0648,
      "num_input_tokens_seen": 44606288,
      "step": 23535
    },
    {
      "epoch": 3.8401990374418795,
      "grad_norm": 0.07771490514278412,
      "learning_rate": 1.8140319421699234e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 44616016,
      "step": 23540
    },
    {
      "epoch": 3.8410147646626966,
      "grad_norm": 0.914222002029419,
      "learning_rate": 1.8130879275599515e-05,
      "loss": 0.1441,
      "num_input_tokens_seen": 44626176,
      "step": 23545
    },
    {
      "epoch": 3.841830491883514,
      "grad_norm": 0.29421085119247437,
      "learning_rate": 1.8121440188804544e-05,
      "loss": 0.1144,
      "num_input_tokens_seen": 44634304,
      "step": 23550
    },
    {
      "epoch": 3.8426462191043314,
      "grad_norm": 0.8232117891311646,
      "learning_rate": 1.811200216276993e-05,
      "loss": 0.2528,
      "num_input_tokens_seen": 44642672,
      "step": 23555
    },
    {
      "epoch": 3.843461946325149,
      "grad_norm": 0.17266520857810974,
      "learning_rate": 1.810256519895115e-05,
      "loss": 0.0768,
      "num_input_tokens_seen": 44651408,
      "step": 23560
    },
    {
      "epoch": 3.8442776735459665,
      "grad_norm": 1.6827950477600098,
      "learning_rate": 1.8093129298803494e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 44660400,
      "step": 23565
    },
    {
      "epoch": 3.8450934007667836,
      "grad_norm": 0.8540980815887451,
      "learning_rate": 1.808369446378209e-05,
      "loss": 0.1344,
      "num_input_tokens_seen": 44669536,
      "step": 23570
    },
    {
      "epoch": 3.8459091279876008,
      "grad_norm": 0.669611394405365,
      "learning_rate": 1.8074260695341914e-05,
      "loss": 0.083,
      "num_input_tokens_seen": 44677536,
      "step": 23575
    },
    {
      "epoch": 3.8467248552084183,
      "grad_norm": 0.21300041675567627,
      "learning_rate": 1.8064827994937782e-05,
      "loss": 0.0302,
      "num_input_tokens_seen": 44687600,
      "step": 23580
    },
    {
      "epoch": 3.847540582429236,
      "grad_norm": 1.2870886325836182,
      "learning_rate": 1.8055396364024317e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 44697248,
      "step": 23585
    },
    {
      "epoch": 3.848356309650053,
      "grad_norm": 0.30095356702804565,
      "learning_rate": 1.804596580405601e-05,
      "loss": 0.0655,
      "num_input_tokens_seen": 44706640,
      "step": 23590
    },
    {
      "epoch": 3.84917203687087,
      "grad_norm": 0.10432151705026627,
      "learning_rate": 1.8036536316487174e-05,
      "loss": 0.0253,
      "num_input_tokens_seen": 44717520,
      "step": 23595
    },
    {
      "epoch": 3.8499877640916877,
      "grad_norm": 0.7624309659004211,
      "learning_rate": 1.802710790277193e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 44727696,
      "step": 23600
    },
    {
      "epoch": 3.8499877640916877,
      "eval_loss": 0.14403435587882996,
      "eval_runtime": 133.3447,
      "eval_samples_per_second": 20.436,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 44727696,
      "step": 23600
    },
    {
      "epoch": 3.8508034913125053,
      "grad_norm": 0.9227971434593201,
      "learning_rate": 1.801768056436429e-05,
      "loss": 0.0954,
      "num_input_tokens_seen": 44736704,
      "step": 23605
    },
    {
      "epoch": 3.8516192185333225,
      "grad_norm": 0.8835411667823792,
      "learning_rate": 1.8008254302718035e-05,
      "loss": 0.1476,
      "num_input_tokens_seen": 44747200,
      "step": 23610
    },
    {
      "epoch": 3.8524349457541396,
      "grad_norm": 0.08242368698120117,
      "learning_rate": 1.7998829119286837e-05,
      "loss": 0.0838,
      "num_input_tokens_seen": 44756272,
      "step": 23615
    },
    {
      "epoch": 3.853250672974957,
      "grad_norm": 0.29915153980255127,
      "learning_rate": 1.798940501552418e-05,
      "loss": 0.0818,
      "num_input_tokens_seen": 44766800,
      "step": 23620
    },
    {
      "epoch": 3.8540664001957747,
      "grad_norm": 1.6980770826339722,
      "learning_rate": 1.797998199288336e-05,
      "loss": 0.1366,
      "num_input_tokens_seen": 44774992,
      "step": 23625
    },
    {
      "epoch": 3.854882127416592,
      "grad_norm": 0.03220108523964882,
      "learning_rate": 1.7970560052817543e-05,
      "loss": 0.2109,
      "num_input_tokens_seen": 44785104,
      "step": 23630
    },
    {
      "epoch": 3.855697854637409,
      "grad_norm": 1.1892790794372559,
      "learning_rate": 1.7961139196779702e-05,
      "loss": 0.1639,
      "num_input_tokens_seen": 44794320,
      "step": 23635
    },
    {
      "epoch": 3.8565135818582266,
      "grad_norm": 0.07272622734308243,
      "learning_rate": 1.7951719426222647e-05,
      "loss": 0.0881,
      "num_input_tokens_seen": 44804160,
      "step": 23640
    },
    {
      "epoch": 3.857329309079044,
      "grad_norm": 0.77216637134552,
      "learning_rate": 1.794230074259904e-05,
      "loss": 0.1323,
      "num_input_tokens_seen": 44812752,
      "step": 23645
    },
    {
      "epoch": 3.8581450362998613,
      "grad_norm": 0.2794467508792877,
      "learning_rate": 1.7932883147361336e-05,
      "loss": 0.1281,
      "num_input_tokens_seen": 44822848,
      "step": 23650
    },
    {
      "epoch": 3.858960763520679,
      "grad_norm": 0.2212969809770584,
      "learning_rate": 1.7923466641961865e-05,
      "loss": 0.0664,
      "num_input_tokens_seen": 44832224,
      "step": 23655
    },
    {
      "epoch": 3.859776490741496,
      "grad_norm": 0.10536817461252213,
      "learning_rate": 1.791405122785278e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 44841120,
      "step": 23660
    },
    {
      "epoch": 3.8605922179623136,
      "grad_norm": 3.0593016147613525,
      "learning_rate": 1.7904636906486037e-05,
      "loss": 0.0941,
      "num_input_tokens_seen": 44851424,
      "step": 23665
    },
    {
      "epoch": 3.8614079451831307,
      "grad_norm": 0.3065524995326996,
      "learning_rate": 1.7895223679313448e-05,
      "loss": 0.1053,
      "num_input_tokens_seen": 44861360,
      "step": 23670
    },
    {
      "epoch": 3.8622236724039483,
      "grad_norm": 0.1761532872915268,
      "learning_rate": 1.7885811547786653e-05,
      "loss": 0.04,
      "num_input_tokens_seen": 44871616,
      "step": 23675
    },
    {
      "epoch": 3.8630393996247654,
      "grad_norm": 0.1854708194732666,
      "learning_rate": 1.7876400513357115e-05,
      "loss": 0.0972,
      "num_input_tokens_seen": 44881488,
      "step": 23680
    },
    {
      "epoch": 3.863855126845583,
      "grad_norm": 0.2723979353904724,
      "learning_rate": 1.7866990577476146e-05,
      "loss": 0.0478,
      "num_input_tokens_seen": 44891104,
      "step": 23685
    },
    {
      "epoch": 3.8646708540664,
      "grad_norm": 0.6512144207954407,
      "learning_rate": 1.7857581741594863e-05,
      "loss": 0.2217,
      "num_input_tokens_seen": 44900704,
      "step": 23690
    },
    {
      "epoch": 3.8654865812872177,
      "grad_norm": 0.6314682364463806,
      "learning_rate": 1.7848174007164237e-05,
      "loss": 0.0709,
      "num_input_tokens_seen": 44909968,
      "step": 23695
    },
    {
      "epoch": 3.866302308508035,
      "grad_norm": 0.12235802412033081,
      "learning_rate": 1.7838767375635052e-05,
      "loss": 0.051,
      "num_input_tokens_seen": 44919584,
      "step": 23700
    },
    {
      "epoch": 3.8671180357288524,
      "grad_norm": 0.4022485017776489,
      "learning_rate": 1.782936184845793e-05,
      "loss": 0.1071,
      "num_input_tokens_seen": 44928736,
      "step": 23705
    },
    {
      "epoch": 3.8679337629496695,
      "grad_norm": 0.34885093569755554,
      "learning_rate": 1.7819957427083334e-05,
      "loss": 0.105,
      "num_input_tokens_seen": 44938304,
      "step": 23710
    },
    {
      "epoch": 3.868749490170487,
      "grad_norm": 0.13120128214359283,
      "learning_rate": 1.7810554112961516e-05,
      "loss": 0.0675,
      "num_input_tokens_seen": 44947616,
      "step": 23715
    },
    {
      "epoch": 3.869565217391304,
      "grad_norm": 0.3778609335422516,
      "learning_rate": 1.7801151907542607e-05,
      "loss": 0.0982,
      "num_input_tokens_seen": 44956560,
      "step": 23720
    },
    {
      "epoch": 3.870380944612122,
      "grad_norm": 0.20410768687725067,
      "learning_rate": 1.7791750812276547e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 44965376,
      "step": 23725
    },
    {
      "epoch": 3.871196671832939,
      "grad_norm": 1.313470482826233,
      "learning_rate": 1.778235082861309e-05,
      "loss": 0.348,
      "num_input_tokens_seen": 44975344,
      "step": 23730
    },
    {
      "epoch": 3.8720123990537565,
      "grad_norm": 0.37485235929489136,
      "learning_rate": 1.777295195800184e-05,
      "loss": 0.2286,
      "num_input_tokens_seen": 44984464,
      "step": 23735
    },
    {
      "epoch": 3.8728281262745736,
      "grad_norm": 1.4188157320022583,
      "learning_rate": 1.7763554201892215e-05,
      "loss": 0.0955,
      "num_input_tokens_seen": 44994048,
      "step": 23740
    },
    {
      "epoch": 3.873643853495391,
      "grad_norm": 0.226064532995224,
      "learning_rate": 1.7754157561733476e-05,
      "loss": 0.0314,
      "num_input_tokens_seen": 45003616,
      "step": 23745
    },
    {
      "epoch": 3.8744595807162083,
      "grad_norm": 0.7209379076957703,
      "learning_rate": 1.7744762038974702e-05,
      "loss": 0.1938,
      "num_input_tokens_seen": 45013152,
      "step": 23750
    },
    {
      "epoch": 3.875275307937026,
      "grad_norm": 1.5982897281646729,
      "learning_rate": 1.7735367635064788e-05,
      "loss": 0.2963,
      "num_input_tokens_seen": 45023616,
      "step": 23755
    },
    {
      "epoch": 3.8760910351578435,
      "grad_norm": 0.9570164084434509,
      "learning_rate": 1.7725974351452474e-05,
      "loss": 0.1081,
      "num_input_tokens_seen": 45032576,
      "step": 23760
    },
    {
      "epoch": 3.8769067623786606,
      "grad_norm": 0.1740606427192688,
      "learning_rate": 1.771658218958634e-05,
      "loss": 0.0488,
      "num_input_tokens_seen": 45041312,
      "step": 23765
    },
    {
      "epoch": 3.8777224895994777,
      "grad_norm": 1.2199435234069824,
      "learning_rate": 1.770719115091475e-05,
      "loss": 0.126,
      "num_input_tokens_seen": 45051040,
      "step": 23770
    },
    {
      "epoch": 3.8785382168202953,
      "grad_norm": 0.04503897950053215,
      "learning_rate": 1.7697801236885935e-05,
      "loss": 0.0598,
      "num_input_tokens_seen": 45061088,
      "step": 23775
    },
    {
      "epoch": 3.879353944041113,
      "grad_norm": 1.709801435470581,
      "learning_rate": 1.7688412448947944e-05,
      "loss": 0.188,
      "num_input_tokens_seen": 45070912,
      "step": 23780
    },
    {
      "epoch": 3.88016967126193,
      "grad_norm": 0.050722669810056686,
      "learning_rate": 1.767902478854862e-05,
      "loss": 0.0592,
      "num_input_tokens_seen": 45080368,
      "step": 23785
    },
    {
      "epoch": 3.880985398482747,
      "grad_norm": 0.6015008091926575,
      "learning_rate": 1.766963825713569e-05,
      "loss": 0.0972,
      "num_input_tokens_seen": 45089392,
      "step": 23790
    },
    {
      "epoch": 3.8818011257035647,
      "grad_norm": 0.05452757328748703,
      "learning_rate": 1.766025285615665e-05,
      "loss": 0.159,
      "num_input_tokens_seen": 45098688,
      "step": 23795
    },
    {
      "epoch": 3.8826168529243823,
      "grad_norm": 0.22327622771263123,
      "learning_rate": 1.7650868587058854e-05,
      "loss": 0.2241,
      "num_input_tokens_seen": 45108608,
      "step": 23800
    },
    {
      "epoch": 3.8826168529243823,
      "eval_loss": 0.14651009440422058,
      "eval_runtime": 133.3244,
      "eval_samples_per_second": 20.439,
      "eval_steps_per_second": 10.223,
      "num_input_tokens_seen": 45108608,
      "step": 23800
    },
    {
      "epoch": 3.8834325801451994,
      "grad_norm": 1.5633008480072021,
      "learning_rate": 1.7641485451289484e-05,
      "loss": 0.2875,
      "num_input_tokens_seen": 45117984,
      "step": 23805
    },
    {
      "epoch": 3.8842483073660166,
      "grad_norm": 0.5796490907669067,
      "learning_rate": 1.7632103450295534e-05,
      "loss": 0.068,
      "num_input_tokens_seen": 45125056,
      "step": 23810
    },
    {
      "epoch": 3.885064034586834,
      "grad_norm": 0.3379683792591095,
      "learning_rate": 1.762272258552381e-05,
      "loss": 0.2308,
      "num_input_tokens_seen": 45135680,
      "step": 23815
    },
    {
      "epoch": 3.8858797618076517,
      "grad_norm": 0.378333181142807,
      "learning_rate": 1.7613342858420988e-05,
      "loss": 0.2383,
      "num_input_tokens_seen": 45145664,
      "step": 23820
    },
    {
      "epoch": 3.886695489028469,
      "grad_norm": 0.12549462914466858,
      "learning_rate": 1.760396427043351e-05,
      "loss": 0.087,
      "num_input_tokens_seen": 45154832,
      "step": 23825
    },
    {
      "epoch": 3.887511216249286,
      "grad_norm": 0.47780442237854004,
      "learning_rate": 1.7594586823007696e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 45164768,
      "step": 23830
    },
    {
      "epoch": 3.8883269434701035,
      "grad_norm": 1.7365716695785522,
      "learning_rate": 1.7585210517589646e-05,
      "loss": 0.2807,
      "num_input_tokens_seen": 45174096,
      "step": 23835
    },
    {
      "epoch": 3.889142670690921,
      "grad_norm": 1.3999332189559937,
      "learning_rate": 1.7575835355625314e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 45183680,
      "step": 23840
    },
    {
      "epoch": 3.8899583979117383,
      "grad_norm": 0.43278786540031433,
      "learning_rate": 1.756646133856048e-05,
      "loss": 0.0739,
      "num_input_tokens_seen": 45193392,
      "step": 23845
    },
    {
      "epoch": 3.890774125132556,
      "grad_norm": 0.9828565120697021,
      "learning_rate": 1.7557088467840714e-05,
      "loss": 0.3101,
      "num_input_tokens_seen": 45201056,
      "step": 23850
    },
    {
      "epoch": 3.891589852353373,
      "grad_norm": 1.5416227579116821,
      "learning_rate": 1.7547716744911438e-05,
      "loss": 0.1398,
      "num_input_tokens_seen": 45209392,
      "step": 23855
    },
    {
      "epoch": 3.8924055795741905,
      "grad_norm": 0.07727719098329544,
      "learning_rate": 1.7538346171217902e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 45219584,
      "step": 23860
    },
    {
      "epoch": 3.8932213067950077,
      "grad_norm": 1.0006595849990845,
      "learning_rate": 1.7528976748205146e-05,
      "loss": 0.0803,
      "num_input_tokens_seen": 45228864,
      "step": 23865
    },
    {
      "epoch": 3.8940370340158252,
      "grad_norm": 1.3783223628997803,
      "learning_rate": 1.751960847731807e-05,
      "loss": 0.0892,
      "num_input_tokens_seen": 45238176,
      "step": 23870
    },
    {
      "epoch": 3.8948527612366424,
      "grad_norm": 0.631540834903717,
      "learning_rate": 1.7510241360001362e-05,
      "loss": 0.1197,
      "num_input_tokens_seen": 45248352,
      "step": 23875
    },
    {
      "epoch": 3.89566848845746,
      "grad_norm": 0.4657268226146698,
      "learning_rate": 1.7500875397699562e-05,
      "loss": 0.2091,
      "num_input_tokens_seen": 45256352,
      "step": 23880
    },
    {
      "epoch": 3.896484215678277,
      "grad_norm": 0.547208309173584,
      "learning_rate": 1.7491510591857015e-05,
      "loss": 0.1263,
      "num_input_tokens_seen": 45265840,
      "step": 23885
    },
    {
      "epoch": 3.8972999428990946,
      "grad_norm": 0.42833057045936584,
      "learning_rate": 1.7482146943917896e-05,
      "loss": 0.1864,
      "num_input_tokens_seen": 45275104,
      "step": 23890
    },
    {
      "epoch": 3.898115670119912,
      "grad_norm": 0.14282232522964478,
      "learning_rate": 1.7472784455326185e-05,
      "loss": 0.0926,
      "num_input_tokens_seen": 45285600,
      "step": 23895
    },
    {
      "epoch": 3.8989313973407294,
      "grad_norm": 0.6811082363128662,
      "learning_rate": 1.746342312752572e-05,
      "loss": 0.0881,
      "num_input_tokens_seen": 45294848,
      "step": 23900
    },
    {
      "epoch": 3.8997471245615465,
      "grad_norm": 0.5607857704162598,
      "learning_rate": 1.74540629619601e-05,
      "loss": 0.1248,
      "num_input_tokens_seen": 45303648,
      "step": 23905
    },
    {
      "epoch": 3.900562851782364,
      "grad_norm": 1.976834774017334,
      "learning_rate": 1.7444703960072815e-05,
      "loss": 0.3093,
      "num_input_tokens_seen": 45311568,
      "step": 23910
    },
    {
      "epoch": 3.901378579003181,
      "grad_norm": 1.2636839151382446,
      "learning_rate": 1.7435346123307118e-05,
      "loss": 0.091,
      "num_input_tokens_seen": 45320896,
      "step": 23915
    },
    {
      "epoch": 3.9021943062239988,
      "grad_norm": 0.4377771317958832,
      "learning_rate": 1.742598945310611e-05,
      "loss": 0.0386,
      "num_input_tokens_seen": 45330304,
      "step": 23920
    },
    {
      "epoch": 3.903010033444816,
      "grad_norm": 0.5158237814903259,
      "learning_rate": 1.741663395091272e-05,
      "loss": 0.0461,
      "num_input_tokens_seen": 45339008,
      "step": 23925
    },
    {
      "epoch": 3.9038257606656335,
      "grad_norm": 1.8904434442520142,
      "learning_rate": 1.7407279618169657e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 45348768,
      "step": 23930
    },
    {
      "epoch": 3.904641487886451,
      "grad_norm": 0.7527264356613159,
      "learning_rate": 1.73979264563195e-05,
      "loss": 0.2259,
      "num_input_tokens_seen": 45357904,
      "step": 23935
    },
    {
      "epoch": 3.905457215107268,
      "grad_norm": 1.5014207363128662,
      "learning_rate": 1.7388574466804625e-05,
      "loss": 0.1076,
      "num_input_tokens_seen": 45368192,
      "step": 23940
    },
    {
      "epoch": 3.9062729423280853,
      "grad_norm": 0.41816869378089905,
      "learning_rate": 1.7379223651067207e-05,
      "loss": 0.2121,
      "num_input_tokens_seen": 45378240,
      "step": 23945
    },
    {
      "epoch": 3.907088669548903,
      "grad_norm": 0.19481196999549866,
      "learning_rate": 1.736987401054928e-05,
      "loss": 0.0812,
      "num_input_tokens_seen": 45387184,
      "step": 23950
    },
    {
      "epoch": 3.9079043967697205,
      "grad_norm": 0.6517174243927002,
      "learning_rate": 1.736052554669266e-05,
      "loss": 0.2059,
      "num_input_tokens_seen": 45396368,
      "step": 23955
    },
    {
      "epoch": 3.9087201239905376,
      "grad_norm": 0.9399107098579407,
      "learning_rate": 1.7351178260939007e-05,
      "loss": 0.119,
      "num_input_tokens_seen": 45406192,
      "step": 23960
    },
    {
      "epoch": 3.9095358512113547,
      "grad_norm": 0.0881383866071701,
      "learning_rate": 1.7341832154729794e-05,
      "loss": 0.1285,
      "num_input_tokens_seen": 45414848,
      "step": 23965
    },
    {
      "epoch": 3.9103515784321723,
      "grad_norm": 0.06766019761562347,
      "learning_rate": 1.7332487229506286e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 45424944,
      "step": 23970
    },
    {
      "epoch": 3.91116730565299,
      "grad_norm": 0.20119895040988922,
      "learning_rate": 1.732314348670961e-05,
      "loss": 0.1169,
      "num_input_tokens_seen": 45435840,
      "step": 23975
    },
    {
      "epoch": 3.911983032873807,
      "grad_norm": 0.2357986718416214,
      "learning_rate": 1.7313800927780686e-05,
      "loss": 0.057,
      "num_input_tokens_seen": 45446352,
      "step": 23980
    },
    {
      "epoch": 3.912798760094624,
      "grad_norm": 0.6124269366264343,
      "learning_rate": 1.7304459554160245e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 45455632,
      "step": 23985
    },
    {
      "epoch": 3.9136144873154417,
      "grad_norm": 0.1299319714307785,
      "learning_rate": 1.7295119367288853e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 45464864,
      "step": 23990
    },
    {
      "epoch": 3.9144302145362593,
      "grad_norm": 0.14840571582317352,
      "learning_rate": 1.728578036860688e-05,
      "loss": 0.0657,
      "num_input_tokens_seen": 45473360,
      "step": 23995
    },
    {
      "epoch": 3.9152459417570764,
      "grad_norm": 1.5311857461929321,
      "learning_rate": 1.7276442559554513e-05,
      "loss": 0.1722,
      "num_input_tokens_seen": 45482928,
      "step": 24000
    },
    {
      "epoch": 3.9152459417570764,
      "eval_loss": 0.1463126689195633,
      "eval_runtime": 133.2299,
      "eval_samples_per_second": 20.453,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 45482928,
      "step": 24000
    },
    {
      "epoch": 3.9160616689778935,
      "grad_norm": 0.17903926968574524,
      "learning_rate": 1.726710594157177e-05,
      "loss": 0.0258,
      "num_input_tokens_seen": 45491600,
      "step": 24005
    },
    {
      "epoch": 3.916877396198711,
      "grad_norm": 0.8616857528686523,
      "learning_rate": 1.725777051609846e-05,
      "loss": 0.058,
      "num_input_tokens_seen": 45501840,
      "step": 24010
    },
    {
      "epoch": 3.9176931234195287,
      "grad_norm": 0.6374239325523376,
      "learning_rate": 1.7248436284574228e-05,
      "loss": 0.1153,
      "num_input_tokens_seen": 45509264,
      "step": 24015
    },
    {
      "epoch": 3.918508850640346,
      "grad_norm": 0.09208217263221741,
      "learning_rate": 1.723910324843855e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 45518288,
      "step": 24020
    },
    {
      "epoch": 3.919324577861163,
      "grad_norm": 1.421075463294983,
      "learning_rate": 1.722977140913067e-05,
      "loss": 0.1918,
      "num_input_tokens_seen": 45528096,
      "step": 24025
    },
    {
      "epoch": 3.9201403050819805,
      "grad_norm": 1.6920884847640991,
      "learning_rate": 1.7220440768089688e-05,
      "loss": 0.283,
      "num_input_tokens_seen": 45537776,
      "step": 24030
    },
    {
      "epoch": 3.920956032302798,
      "grad_norm": 1.6927238702774048,
      "learning_rate": 1.7211111326754505e-05,
      "loss": 0.1859,
      "num_input_tokens_seen": 45546864,
      "step": 24035
    },
    {
      "epoch": 3.9217717595236152,
      "grad_norm": 1.5196208953857422,
      "learning_rate": 1.720178308656383e-05,
      "loss": 0.3694,
      "num_input_tokens_seen": 45556848,
      "step": 24040
    },
    {
      "epoch": 3.922587486744433,
      "grad_norm": 0.5890340209007263,
      "learning_rate": 1.719245604895621e-05,
      "loss": 0.0798,
      "num_input_tokens_seen": 45565344,
      "step": 24045
    },
    {
      "epoch": 3.92340321396525,
      "grad_norm": 0.9344059824943542,
      "learning_rate": 1.7183130215369972e-05,
      "loss": 0.2636,
      "num_input_tokens_seen": 45575216,
      "step": 24050
    },
    {
      "epoch": 3.9242189411860675,
      "grad_norm": 0.3141564428806305,
      "learning_rate": 1.7173805587243292e-05,
      "loss": 0.041,
      "num_input_tokens_seen": 45584848,
      "step": 24055
    },
    {
      "epoch": 3.9250346684068846,
      "grad_norm": 0.0773845985531807,
      "learning_rate": 1.7164482166014147e-05,
      "loss": 0.0327,
      "num_input_tokens_seen": 45594192,
      "step": 24060
    },
    {
      "epoch": 3.925850395627702,
      "grad_norm": 0.5115588903427124,
      "learning_rate": 1.7155159953120313e-05,
      "loss": 0.0404,
      "num_input_tokens_seen": 45603728,
      "step": 24065
    },
    {
      "epoch": 3.9266661228485193,
      "grad_norm": 0.6603475213050842,
      "learning_rate": 1.714583894999941e-05,
      "loss": 0.1706,
      "num_input_tokens_seen": 45613088,
      "step": 24070
    },
    {
      "epoch": 3.927481850069337,
      "grad_norm": 0.05363493785262108,
      "learning_rate": 1.7136519158088826e-05,
      "loss": 0.0529,
      "num_input_tokens_seen": 45622448,
      "step": 24075
    },
    {
      "epoch": 3.928297577290154,
      "grad_norm": 0.13595661520957947,
      "learning_rate": 1.712720057882581e-05,
      "loss": 0.0487,
      "num_input_tokens_seen": 45632752,
      "step": 24080
    },
    {
      "epoch": 3.9291133045109716,
      "grad_norm": 1.8308213949203491,
      "learning_rate": 1.7117883213647413e-05,
      "loss": 0.1175,
      "num_input_tokens_seen": 45642272,
      "step": 24085
    },
    {
      "epoch": 3.9299290317317888,
      "grad_norm": 1.0430011749267578,
      "learning_rate": 1.710856706399046e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 45651888,
      "step": 24090
    },
    {
      "epoch": 3.9307447589526063,
      "grad_norm": 0.7498059272766113,
      "learning_rate": 1.7099252131291648e-05,
      "loss": 0.0866,
      "num_input_tokens_seen": 45661952,
      "step": 24095
    },
    {
      "epoch": 3.9315604861734235,
      "grad_norm": 0.7484574913978577,
      "learning_rate": 1.708993841698744e-05,
      "loss": 0.0581,
      "num_input_tokens_seen": 45672176,
      "step": 24100
    },
    {
      "epoch": 3.932376213394241,
      "grad_norm": 0.12331788241863251,
      "learning_rate": 1.7080625922514132e-05,
      "loss": 0.1233,
      "num_input_tokens_seen": 45682080,
      "step": 24105
    },
    {
      "epoch": 3.933191940615058,
      "grad_norm": 0.0631118193268776,
      "learning_rate": 1.7071314649307836e-05,
      "loss": 0.239,
      "num_input_tokens_seen": 45692256,
      "step": 24110
    },
    {
      "epoch": 3.9340076678358757,
      "grad_norm": 0.0629170760512352,
      "learning_rate": 1.7062004598804448e-05,
      "loss": 0.2789,
      "num_input_tokens_seen": 45702512,
      "step": 24115
    },
    {
      "epoch": 3.934823395056693,
      "grad_norm": 0.1754053235054016,
      "learning_rate": 1.7052695772439702e-05,
      "loss": 0.0968,
      "num_input_tokens_seen": 45712432,
      "step": 24120
    },
    {
      "epoch": 3.9356391222775104,
      "grad_norm": 0.28923991322517395,
      "learning_rate": 1.7043388171649154e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 45721920,
      "step": 24125
    },
    {
      "epoch": 3.936454849498328,
      "grad_norm": 0.7430757284164429,
      "learning_rate": 1.7034081797868127e-05,
      "loss": 0.1223,
      "num_input_tokens_seen": 45732160,
      "step": 24130
    },
    {
      "epoch": 3.937270576719145,
      "grad_norm": 0.36168864369392395,
      "learning_rate": 1.70247766525318e-05,
      "loss": 0.2589,
      "num_input_tokens_seen": 45742096,
      "step": 24135
    },
    {
      "epoch": 3.9380863039399623,
      "grad_norm": 0.17467881739139557,
      "learning_rate": 1.701547273707514e-05,
      "loss": 0.1091,
      "num_input_tokens_seen": 45750176,
      "step": 24140
    },
    {
      "epoch": 3.93890203116078,
      "grad_norm": 0.3524967133998871,
      "learning_rate": 1.7006170052932916e-05,
      "loss": 0.169,
      "num_input_tokens_seen": 45759744,
      "step": 24145
    },
    {
      "epoch": 3.9397177583815974,
      "grad_norm": 0.11192929744720459,
      "learning_rate": 1.6996868601539735e-05,
      "loss": 0.0608,
      "num_input_tokens_seen": 45768896,
      "step": 24150
    },
    {
      "epoch": 3.9405334856024146,
      "grad_norm": 0.1477346271276474,
      "learning_rate": 1.6987568384329977e-05,
      "loss": 0.0369,
      "num_input_tokens_seen": 45777184,
      "step": 24155
    },
    {
      "epoch": 3.9413492128232317,
      "grad_norm": 1.1905351877212524,
      "learning_rate": 1.6978269402737866e-05,
      "loss": 0.2106,
      "num_input_tokens_seen": 45786928,
      "step": 24160
    },
    {
      "epoch": 3.9421649400440493,
      "grad_norm": 0.26549267768859863,
      "learning_rate": 1.696897165819743e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 45797680,
      "step": 24165
    },
    {
      "epoch": 3.942980667264867,
      "grad_norm": 1.086202621459961,
      "learning_rate": 1.6959675152142487e-05,
      "loss": 0.0907,
      "num_input_tokens_seen": 45806384,
      "step": 24170
    },
    {
      "epoch": 3.943796394485684,
      "grad_norm": 0.9333745837211609,
      "learning_rate": 1.6950379886006667e-05,
      "loss": 0.0984,
      "num_input_tokens_seen": 45816576,
      "step": 24175
    },
    {
      "epoch": 3.944612121706501,
      "grad_norm": 1.4222408533096313,
      "learning_rate": 1.6941085861223438e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 45826192,
      "step": 24180
    },
    {
      "epoch": 3.9454278489273187,
      "grad_norm": 1.3284674882888794,
      "learning_rate": 1.6931793079226034e-05,
      "loss": 0.1804,
      "num_input_tokens_seen": 45835856,
      "step": 24185
    },
    {
      "epoch": 3.9462435761481363,
      "grad_norm": 0.7483528256416321,
      "learning_rate": 1.692250154144754e-05,
      "loss": 0.2056,
      "num_input_tokens_seen": 45845488,
      "step": 24190
    },
    {
      "epoch": 3.9470593033689534,
      "grad_norm": 1.4891209602355957,
      "learning_rate": 1.6913211249320807e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 45854112,
      "step": 24195
    },
    {
      "epoch": 3.9478750305897705,
      "grad_norm": 0.07119978219270706,
      "learning_rate": 1.6903922204278522e-05,
      "loss": 0.1048,
      "num_input_tokens_seen": 45861584,
      "step": 24200
    },
    {
      "epoch": 3.9478750305897705,
      "eval_loss": 0.14294619858264923,
      "eval_runtime": 133.378,
      "eval_samples_per_second": 20.431,
      "eval_steps_per_second": 10.219,
      "num_input_tokens_seen": 45861584,
      "step": 24200
    },
    {
      "epoch": 3.948690757810588,
      "grad_norm": 0.23623786866664886,
      "learning_rate": 1.6894634407753186e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 45872688,
      "step": 24205
    },
    {
      "epoch": 3.9495064850314057,
      "grad_norm": 0.034930840134620667,
      "learning_rate": 1.6885347861177077e-05,
      "loss": 0.0458,
      "num_input_tokens_seen": 45882272,
      "step": 24210
    },
    {
      "epoch": 3.950322212252223,
      "grad_norm": 1.2381242513656616,
      "learning_rate": 1.6876062565982298e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 45892672,
      "step": 24215
    },
    {
      "epoch": 3.9511379394730404,
      "grad_norm": 1.503767490386963,
      "learning_rate": 1.6866778523600774e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 45902592,
      "step": 24220
    },
    {
      "epoch": 3.9519536666938575,
      "grad_norm": 1.173865795135498,
      "learning_rate": 1.6857495735464195e-05,
      "loss": 0.1143,
      "num_input_tokens_seen": 45911904,
      "step": 24225
    },
    {
      "epoch": 3.952769393914675,
      "grad_norm": 1.2285101413726807,
      "learning_rate": 1.6848214203004115e-05,
      "loss": 0.2796,
      "num_input_tokens_seen": 45921664,
      "step": 24230
    },
    {
      "epoch": 3.953585121135492,
      "grad_norm": 0.15157805383205414,
      "learning_rate": 1.6838933927651835e-05,
      "loss": 0.0559,
      "num_input_tokens_seen": 45930896,
      "step": 24235
    },
    {
      "epoch": 3.95440084835631,
      "grad_norm": 1.1754801273345947,
      "learning_rate": 1.6829654910838506e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 45940272,
      "step": 24240
    },
    {
      "epoch": 3.955216575577127,
      "grad_norm": 0.04234116151928902,
      "learning_rate": 1.6820377153995065e-05,
      "loss": 0.2322,
      "num_input_tokens_seen": 45950256,
      "step": 24245
    },
    {
      "epoch": 3.9560323027979445,
      "grad_norm": 0.2774607837200165,
      "learning_rate": 1.681110065855226e-05,
      "loss": 0.4454,
      "num_input_tokens_seen": 45959424,
      "step": 24250
    },
    {
      "epoch": 3.9568480300187616,
      "grad_norm": 0.21822351217269897,
      "learning_rate": 1.6801825425940642e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 45968304,
      "step": 24255
    },
    {
      "epoch": 3.957663757239579,
      "grad_norm": 0.1422329992055893,
      "learning_rate": 1.679255145759056e-05,
      "loss": 0.0857,
      "num_input_tokens_seen": 45977280,
      "step": 24260
    },
    {
      "epoch": 3.9584794844603963,
      "grad_norm": 0.9140601754188538,
      "learning_rate": 1.6783278754932187e-05,
      "loss": 0.1738,
      "num_input_tokens_seen": 45987168,
      "step": 24265
    },
    {
      "epoch": 3.959295211681214,
      "grad_norm": 0.7140496969223022,
      "learning_rate": 1.6774007319395496e-05,
      "loss": 0.1908,
      "num_input_tokens_seen": 45996272,
      "step": 24270
    },
    {
      "epoch": 3.960110938902031,
      "grad_norm": 0.26017481088638306,
      "learning_rate": 1.6764737152410243e-05,
      "loss": 0.0791,
      "num_input_tokens_seen": 46005328,
      "step": 24275
    },
    {
      "epoch": 3.9609266661228486,
      "grad_norm": 1.1400198936462402,
      "learning_rate": 1.6755468255406016e-05,
      "loss": 0.4422,
      "num_input_tokens_seen": 46014704,
      "step": 24280
    },
    {
      "epoch": 3.9617423933436657,
      "grad_norm": 1.1746058464050293,
      "learning_rate": 1.674620062981219e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 46023600,
      "step": 24285
    },
    {
      "epoch": 3.9625581205644833,
      "grad_norm": 1.6324880123138428,
      "learning_rate": 1.6736934277057947e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 46032592,
      "step": 24290
    },
    {
      "epoch": 3.9633738477853004,
      "grad_norm": 0.2572663724422455,
      "learning_rate": 1.6727669198572286e-05,
      "loss": 0.1334,
      "num_input_tokens_seen": 46042432,
      "step": 24295
    },
    {
      "epoch": 3.964189575006118,
      "grad_norm": 0.051436133682727814,
      "learning_rate": 1.6718405395783984e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 46052048,
      "step": 24300
    },
    {
      "epoch": 3.965005302226935,
      "grad_norm": 0.5302239060401917,
      "learning_rate": 1.6709142870121643e-05,
      "loss": 0.2493,
      "num_input_tokens_seen": 46060704,
      "step": 24305
    },
    {
      "epoch": 3.9658210294477527,
      "grad_norm": 0.8814501762390137,
      "learning_rate": 1.669988162301367e-05,
      "loss": 0.2384,
      "num_input_tokens_seen": 46071120,
      "step": 24310
    },
    {
      "epoch": 3.96663675666857,
      "grad_norm": 0.8939771056175232,
      "learning_rate": 1.6690621655888243e-05,
      "loss": 0.1164,
      "num_input_tokens_seen": 46081376,
      "step": 24315
    },
    {
      "epoch": 3.9674524838893874,
      "grad_norm": 0.2437012940645218,
      "learning_rate": 1.6681362970173386e-05,
      "loss": 0.0546,
      "num_input_tokens_seen": 46090192,
      "step": 24320
    },
    {
      "epoch": 3.968268211110205,
      "grad_norm": 0.24367326498031616,
      "learning_rate": 1.6672105567296904e-05,
      "loss": 0.3063,
      "num_input_tokens_seen": 46097552,
      "step": 24325
    },
    {
      "epoch": 3.969083938331022,
      "grad_norm": 0.1636185646057129,
      "learning_rate": 1.666284944868639e-05,
      "loss": 0.0568,
      "num_input_tokens_seen": 46108784,
      "step": 24330
    },
    {
      "epoch": 3.9698996655518393,
      "grad_norm": 0.19311486184597015,
      "learning_rate": 1.665359461576927e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 46118464,
      "step": 24335
    },
    {
      "epoch": 3.970715392772657,
      "grad_norm": 0.23652775585651398,
      "learning_rate": 1.6644341069972736e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 46127952,
      "step": 24340
    },
    {
      "epoch": 3.9715311199934744,
      "grad_norm": 1.1688898801803589,
      "learning_rate": 1.6635088812723813e-05,
      "loss": 0.3605,
      "num_input_tokens_seen": 46137504,
      "step": 24345
    },
    {
      "epoch": 3.9723468472142915,
      "grad_norm": 0.6755690574645996,
      "learning_rate": 1.6625837845449328e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 46147120,
      "step": 24350
    },
    {
      "epoch": 3.9731625744351087,
      "grad_norm": 1.426267147064209,
      "learning_rate": 1.6616588169575874e-05,
      "loss": 0.2175,
      "num_input_tokens_seen": 46156512,
      "step": 24355
    },
    {
      "epoch": 3.9739783016559262,
      "grad_norm": 0.519043505191803,
      "learning_rate": 1.6607339786529878e-05,
      "loss": 0.2633,
      "num_input_tokens_seen": 46166768,
      "step": 24360
    },
    {
      "epoch": 3.974794028876744,
      "grad_norm": 0.08866547793149948,
      "learning_rate": 1.659809269773756e-05,
      "loss": 0.0339,
      "num_input_tokens_seen": 46175744,
      "step": 24365
    },
    {
      "epoch": 3.975609756097561,
      "grad_norm": 0.6839200854301453,
      "learning_rate": 1.658884690462493e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 46185440,
      "step": 24370
    },
    {
      "epoch": 3.976425483318378,
      "grad_norm": 1.249464511871338,
      "learning_rate": 1.6579602408617813e-05,
      "loss": 0.2322,
      "num_input_tokens_seen": 46195936,
      "step": 24375
    },
    {
      "epoch": 3.9772412105391957,
      "grad_norm": 0.04690387845039368,
      "learning_rate": 1.657035921114181e-05,
      "loss": 0.1126,
      "num_input_tokens_seen": 46205152,
      "step": 24380
    },
    {
      "epoch": 3.9780569377600132,
      "grad_norm": 0.06581050157546997,
      "learning_rate": 1.656111731362236e-05,
      "loss": 0.0554,
      "num_input_tokens_seen": 46213888,
      "step": 24385
    },
    {
      "epoch": 3.9788726649808304,
      "grad_norm": 1.4638489484786987,
      "learning_rate": 1.6551876717484666e-05,
      "loss": 0.1686,
      "num_input_tokens_seen": 46224000,
      "step": 24390
    },
    {
      "epoch": 3.9796883922016475,
      "grad_norm": 0.33208030462265015,
      "learning_rate": 1.6542637424153752e-05,
      "loss": 0.1178,
      "num_input_tokens_seen": 46234064,
      "step": 24395
    },
    {
      "epoch": 3.980504119422465,
      "grad_norm": 1.1202435493469238,
      "learning_rate": 1.6533399435054418e-05,
      "loss": 0.2229,
      "num_input_tokens_seen": 46243072,
      "step": 24400
    },
    {
      "epoch": 3.980504119422465,
      "eval_loss": 0.14083987474441528,
      "eval_runtime": 133.3487,
      "eval_samples_per_second": 20.435,
      "eval_steps_per_second": 10.221,
      "num_input_tokens_seen": 46243072,
      "step": 24400
    },
    {
      "epoch": 3.9813198466432826,
      "grad_norm": 1.2226130962371826,
      "learning_rate": 1.6524162751611304e-05,
      "loss": 0.3032,
      "num_input_tokens_seen": 46253840,
      "step": 24405
    },
    {
      "epoch": 3.9821355738640998,
      "grad_norm": 0.09409832954406738,
      "learning_rate": 1.6514927375248796e-05,
      "loss": 0.0654,
      "num_input_tokens_seen": 46262208,
      "step": 24410
    },
    {
      "epoch": 3.9829513010849174,
      "grad_norm": 0.07743280380964279,
      "learning_rate": 1.6505693307391127e-05,
      "loss": 0.0591,
      "num_input_tokens_seen": 46271504,
      "step": 24415
    },
    {
      "epoch": 3.9837670283057345,
      "grad_norm": 0.07317022979259491,
      "learning_rate": 1.6496460549462288e-05,
      "loss": 0.0955,
      "num_input_tokens_seen": 46281264,
      "step": 24420
    },
    {
      "epoch": 3.984582755526552,
      "grad_norm": 1.6951392889022827,
      "learning_rate": 1.6487229102886097e-05,
      "loss": 0.4306,
      "num_input_tokens_seen": 46290544,
      "step": 24425
    },
    {
      "epoch": 3.985398482747369,
      "grad_norm": 0.20988045632839203,
      "learning_rate": 1.6477998969086155e-05,
      "loss": 0.0895,
      "num_input_tokens_seen": 46299600,
      "step": 24430
    },
    {
      "epoch": 3.9862142099681868,
      "grad_norm": 0.8303440809249878,
      "learning_rate": 1.646877014948587e-05,
      "loss": 0.0789,
      "num_input_tokens_seen": 46308928,
      "step": 24435
    },
    {
      "epoch": 3.987029937189004,
      "grad_norm": 1.1700952053070068,
      "learning_rate": 1.6459542645508433e-05,
      "loss": 0.2421,
      "num_input_tokens_seen": 46318896,
      "step": 24440
    },
    {
      "epoch": 3.9878456644098215,
      "grad_norm": 0.150459885597229,
      "learning_rate": 1.6450316458576852e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 46328048,
      "step": 24445
    },
    {
      "epoch": 3.9886613916306386,
      "grad_norm": 0.5393175482749939,
      "learning_rate": 1.6441091590113912e-05,
      "loss": 0.0764,
      "num_input_tokens_seen": 46336864,
      "step": 24450
    },
    {
      "epoch": 3.989477118851456,
      "grad_norm": 1.1270545721054077,
      "learning_rate": 1.6431868041542213e-05,
      "loss": 0.1215,
      "num_input_tokens_seen": 46345920,
      "step": 24455
    },
    {
      "epoch": 3.9902928460722733,
      "grad_norm": 0.09895604848861694,
      "learning_rate": 1.6422645814284123e-05,
      "loss": 0.1536,
      "num_input_tokens_seen": 46355056,
      "step": 24460
    },
    {
      "epoch": 3.991108573293091,
      "grad_norm": 0.41770806908607483,
      "learning_rate": 1.6413424909761846e-05,
      "loss": 0.1217,
      "num_input_tokens_seen": 46364608,
      "step": 24465
    },
    {
      "epoch": 3.991924300513908,
      "grad_norm": 0.16567206382751465,
      "learning_rate": 1.640420532939736e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 46374368,
      "step": 24470
    },
    {
      "epoch": 3.9927400277347256,
      "grad_norm": 0.7062143683433533,
      "learning_rate": 1.639498707461242e-05,
      "loss": 0.0487,
      "num_input_tokens_seen": 46383728,
      "step": 24475
    },
    {
      "epoch": 3.9935557549555427,
      "grad_norm": 1.2180252075195312,
      "learning_rate": 1.6385770146828614e-05,
      "loss": 0.1789,
      "num_input_tokens_seen": 46393472,
      "step": 24480
    },
    {
      "epoch": 3.9943714821763603,
      "grad_norm": 0.557160496711731,
      "learning_rate": 1.637655454746731e-05,
      "loss": 0.0501,
      "num_input_tokens_seen": 46401776,
      "step": 24485
    },
    {
      "epoch": 3.9951872093971774,
      "grad_norm": 0.3173367381095886,
      "learning_rate": 1.6367340277949658e-05,
      "loss": 0.072,
      "num_input_tokens_seen": 46411216,
      "step": 24490
    },
    {
      "epoch": 3.996002936617995,
      "grad_norm": 0.06954721361398697,
      "learning_rate": 1.635812733969663e-05,
      "loss": 0.0714,
      "num_input_tokens_seen": 46421072,
      "step": 24495
    },
    {
      "epoch": 3.9968186638388126,
      "grad_norm": 0.07269421964883804,
      "learning_rate": 1.634891573412896e-05,
      "loss": 0.1037,
      "num_input_tokens_seen": 46431232,
      "step": 24500
    },
    {
      "epoch": 3.9976343910596297,
      "grad_norm": 0.9772508144378662,
      "learning_rate": 1.6339705462667196e-05,
      "loss": 0.1869,
      "num_input_tokens_seen": 46440464,
      "step": 24505
    },
    {
      "epoch": 3.998450118280447,
      "grad_norm": 0.4081915616989136,
      "learning_rate": 1.633049652673169e-05,
      "loss": 0.0804,
      "num_input_tokens_seen": 46449488,
      "step": 24510
    },
    {
      "epoch": 3.9992658455012644,
      "grad_norm": 0.0799042209982872,
      "learning_rate": 1.632128892774256e-05,
      "loss": 0.0645,
      "num_input_tokens_seen": 46459168,
      "step": 24515
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.04242286831140518,
      "learning_rate": 1.6312082667119737e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 46467712,
      "step": 24520
    },
    {
      "epoch": 4.000815727220817,
      "grad_norm": 1.128596544265747,
      "learning_rate": 1.630287774628296e-05,
      "loss": 0.1893,
      "num_input_tokens_seen": 46478256,
      "step": 24525
    },
    {
      "epoch": 4.001631454441635,
      "grad_norm": 0.3515690565109253,
      "learning_rate": 1.6293674166651718e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 46488416,
      "step": 24530
    },
    {
      "epoch": 4.002447181662452,
      "grad_norm": 0.29743823409080505,
      "learning_rate": 1.6284471929645338e-05,
      "loss": 0.1757,
      "num_input_tokens_seen": 46495952,
      "step": 24535
    },
    {
      "epoch": 4.003262908883269,
      "grad_norm": 0.7678806781768799,
      "learning_rate": 1.627527103668291e-05,
      "loss": 0.1046,
      "num_input_tokens_seen": 46503760,
      "step": 24540
    },
    {
      "epoch": 4.0040786361040865,
      "grad_norm": 0.0642191618680954,
      "learning_rate": 1.6266071489183327e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 46513904,
      "step": 24545
    },
    {
      "epoch": 4.004894363324905,
      "grad_norm": 1.2621361017227173,
      "learning_rate": 1.6256873288565283e-05,
      "loss": 0.1063,
      "num_input_tokens_seen": 46523760,
      "step": 24550
    },
    {
      "epoch": 4.005710090545722,
      "grad_norm": 0.3033718764781952,
      "learning_rate": 1.6247676436247245e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 46533056,
      "step": 24555
    },
    {
      "epoch": 4.006525817766539,
      "grad_norm": 0.9582658410072327,
      "learning_rate": 1.6238480933647486e-05,
      "loss": 0.203,
      "num_input_tokens_seen": 46541552,
      "step": 24560
    },
    {
      "epoch": 4.007341544987356,
      "grad_norm": 0.062141235917806625,
      "learning_rate": 1.6229286782184083e-05,
      "loss": 0.0732,
      "num_input_tokens_seen": 46551024,
      "step": 24565
    },
    {
      "epoch": 4.008157272208174,
      "grad_norm": 1.6290743350982666,
      "learning_rate": 1.622009398327487e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 46561648,
      "step": 24570
    },
    {
      "epoch": 4.008972999428991,
      "grad_norm": 0.9028600454330444,
      "learning_rate": 1.6210902538337502e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 46571776,
      "step": 24575
    },
    {
      "epoch": 4.009788726649808,
      "grad_norm": 0.693562924861908,
      "learning_rate": 1.6201712448789413e-05,
      "loss": 0.0859,
      "num_input_tokens_seen": 46581248,
      "step": 24580
    },
    {
      "epoch": 4.010604453870625,
      "grad_norm": 0.8933011889457703,
      "learning_rate": 1.6192523716047827e-05,
      "loss": 0.0826,
      "num_input_tokens_seen": 46591248,
      "step": 24585
    },
    {
      "epoch": 4.011420181091443,
      "grad_norm": 0.8258218765258789,
      "learning_rate": 1.6183336341529776e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 46600160,
      "step": 24590
    },
    {
      "epoch": 4.0122359083122605,
      "grad_norm": 0.14995385706424713,
      "learning_rate": 1.6174150326652047e-05,
      "loss": 0.0499,
      "num_input_tokens_seen": 46609344,
      "step": 24595
    },
    {
      "epoch": 4.013051635533078,
      "grad_norm": 0.35230734944343567,
      "learning_rate": 1.6164965672831256e-05,
      "loss": 0.1949,
      "num_input_tokens_seen": 46619680,
      "step": 24600
    },
    {
      "epoch": 4.013051635533078,
      "eval_loss": 0.1428045928478241,
      "eval_runtime": 133.0539,
      "eval_samples_per_second": 20.48,
      "eval_steps_per_second": 10.244,
      "num_input_tokens_seen": 46619680,
      "step": 24600
    },
    {
      "epoch": 4.013867362753895,
      "grad_norm": 0.6473497748374939,
      "learning_rate": 1.6155782381483784e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 46630064,
      "step": 24605
    },
    {
      "epoch": 4.014683089974713,
      "grad_norm": 0.7169033288955688,
      "learning_rate": 1.6146600454025813e-05,
      "loss": 0.1519,
      "num_input_tokens_seen": 46639104,
      "step": 24610
    },
    {
      "epoch": 4.01549881719553,
      "grad_norm": 0.12051587551832199,
      "learning_rate": 1.6137419891873317e-05,
      "loss": 0.0939,
      "num_input_tokens_seen": 46648752,
      "step": 24615
    },
    {
      "epoch": 4.016314544416347,
      "grad_norm": 0.1277441382408142,
      "learning_rate": 1.6128240696442038e-05,
      "loss": 0.0607,
      "num_input_tokens_seen": 46657712,
      "step": 24620
    },
    {
      "epoch": 4.017130271637164,
      "grad_norm": 0.7385458946228027,
      "learning_rate": 1.611906286914753e-05,
      "loss": 0.0713,
      "num_input_tokens_seen": 46666992,
      "step": 24625
    },
    {
      "epoch": 4.017945998857982,
      "grad_norm": 0.8264437317848206,
      "learning_rate": 1.6109886411405144e-05,
      "loss": 0.1424,
      "num_input_tokens_seen": 46676704,
      "step": 24630
    },
    {
      "epoch": 4.018761726078799,
      "grad_norm": 0.48692426085472107,
      "learning_rate": 1.6100711324629985e-05,
      "loss": 0.2103,
      "num_input_tokens_seen": 46687504,
      "step": 24635
    },
    {
      "epoch": 4.0195774532996165,
      "grad_norm": 0.06458544731140137,
      "learning_rate": 1.609153761023698e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 46697712,
      "step": 24640
    },
    {
      "epoch": 4.020393180520434,
      "grad_norm": 0.4408118724822998,
      "learning_rate": 1.608236526964083e-05,
      "loss": 0.3292,
      "num_input_tokens_seen": 46707408,
      "step": 24645
    },
    {
      "epoch": 4.021208907741252,
      "grad_norm": 0.08912907540798187,
      "learning_rate": 1.607319430425601e-05,
      "loss": 0.0834,
      "num_input_tokens_seen": 46716320,
      "step": 24650
    },
    {
      "epoch": 4.022024634962069,
      "grad_norm": 0.3137606978416443,
      "learning_rate": 1.606402471549682e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 46725104,
      "step": 24655
    },
    {
      "epoch": 4.022840362182886,
      "grad_norm": 0.12610679864883423,
      "learning_rate": 1.6054856504777312e-05,
      "loss": 0.0625,
      "num_input_tokens_seen": 46734592,
      "step": 24660
    },
    {
      "epoch": 4.023656089403703,
      "grad_norm": 0.2380906045436859,
      "learning_rate": 1.6045689673511334e-05,
      "loss": 0.0633,
      "num_input_tokens_seen": 46744480,
      "step": 24665
    },
    {
      "epoch": 4.024471816624521,
      "grad_norm": 0.10097064077854156,
      "learning_rate": 1.6036524223112548e-05,
      "loss": 0.1139,
      "num_input_tokens_seen": 46754496,
      "step": 24670
    },
    {
      "epoch": 4.025287543845338,
      "grad_norm": 0.37763845920562744,
      "learning_rate": 1.602736015499436e-05,
      "loss": 0.119,
      "num_input_tokens_seen": 46764208,
      "step": 24675
    },
    {
      "epoch": 4.026103271066155,
      "grad_norm": 1.185520052909851,
      "learning_rate": 1.601819747057e-05,
      "loss": 0.1786,
      "num_input_tokens_seen": 46773664,
      "step": 24680
    },
    {
      "epoch": 4.026918998286972,
      "grad_norm": 0.025994090363383293,
      "learning_rate": 1.6009036171252465e-05,
      "loss": 0.1012,
      "num_input_tokens_seen": 46784160,
      "step": 24685
    },
    {
      "epoch": 4.02773472550779,
      "grad_norm": 0.3036910891532898,
      "learning_rate": 1.599987625845453e-05,
      "loss": 0.0739,
      "num_input_tokens_seen": 46795296,
      "step": 24690
    },
    {
      "epoch": 4.028550452728608,
      "grad_norm": 0.14437945187091827,
      "learning_rate": 1.599071773358879e-05,
      "loss": 0.0402,
      "num_input_tokens_seen": 46804816,
      "step": 24695
    },
    {
      "epoch": 4.029366179949425,
      "grad_norm": 0.4299236834049225,
      "learning_rate": 1.598156059806758e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 46814768,
      "step": 24700
    },
    {
      "epoch": 4.030181907170242,
      "grad_norm": 0.5989664196968079,
      "learning_rate": 1.5972404853303062e-05,
      "loss": 0.1469,
      "num_input_tokens_seen": 46824448,
      "step": 24705
    },
    {
      "epoch": 4.03099763439106,
      "grad_norm": 0.22936755418777466,
      "learning_rate": 1.5963250500707172e-05,
      "loss": 0.098,
      "num_input_tokens_seen": 46833776,
      "step": 24710
    },
    {
      "epoch": 4.031813361611877,
      "grad_norm": 0.053924866020679474,
      "learning_rate": 1.5954097541691612e-05,
      "loss": 0.0482,
      "num_input_tokens_seen": 46842592,
      "step": 24715
    },
    {
      "epoch": 4.032629088832694,
      "grad_norm": 0.12239731103181839,
      "learning_rate": 1.5944945977667884e-05,
      "loss": 0.099,
      "num_input_tokens_seen": 46851952,
      "step": 24720
    },
    {
      "epoch": 4.033444816053512,
      "grad_norm": 0.2682511806488037,
      "learning_rate": 1.593579581004729e-05,
      "loss": 0.086,
      "num_input_tokens_seen": 46860912,
      "step": 24725
    },
    {
      "epoch": 4.034260543274329,
      "grad_norm": 0.6771997809410095,
      "learning_rate": 1.592664704024088e-05,
      "loss": 0.2438,
      "num_input_tokens_seen": 46869552,
      "step": 24730
    },
    {
      "epoch": 4.035076270495146,
      "grad_norm": 0.44267112016677856,
      "learning_rate": 1.591749966965953e-05,
      "loss": 0.0719,
      "num_input_tokens_seen": 46880208,
      "step": 24735
    },
    {
      "epoch": 4.0358919977159635,
      "grad_norm": 0.9167541265487671,
      "learning_rate": 1.5908353699713856e-05,
      "loss": 0.1763,
      "num_input_tokens_seen": 46890032,
      "step": 24740
    },
    {
      "epoch": 4.0367077249367815,
      "grad_norm": 0.38916152715682983,
      "learning_rate": 1.5899209131814298e-05,
      "loss": 0.0758,
      "num_input_tokens_seen": 46899056,
      "step": 24745
    },
    {
      "epoch": 4.037523452157599,
      "grad_norm": 1.8152936697006226,
      "learning_rate": 1.5890065967371067e-05,
      "loss": 0.2354,
      "num_input_tokens_seen": 46909760,
      "step": 24750
    },
    {
      "epoch": 4.038339179378416,
      "grad_norm": 0.6917181015014648,
      "learning_rate": 1.5880924207794144e-05,
      "loss": 0.236,
      "num_input_tokens_seen": 46917664,
      "step": 24755
    },
    {
      "epoch": 4.039154906599233,
      "grad_norm": 1.1647766828536987,
      "learning_rate": 1.5871783854493298e-05,
      "loss": 0.2446,
      "num_input_tokens_seen": 46928608,
      "step": 24760
    },
    {
      "epoch": 4.039970633820051,
      "grad_norm": 0.9389591217041016,
      "learning_rate": 1.5862644908878106e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 46938480,
      "step": 24765
    },
    {
      "epoch": 4.040786361040868,
      "grad_norm": 0.47413626313209534,
      "learning_rate": 1.5853507372357885e-05,
      "loss": 0.0901,
      "num_input_tokens_seen": 46949392,
      "step": 24770
    },
    {
      "epoch": 4.041602088261685,
      "grad_norm": 0.44526949524879456,
      "learning_rate": 1.5844371246341776e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 46958240,
      "step": 24775
    },
    {
      "epoch": 4.042417815482502,
      "grad_norm": 1.2317938804626465,
      "learning_rate": 1.5835236532238674e-05,
      "loss": 0.2373,
      "num_input_tokens_seen": 46967392,
      "step": 24780
    },
    {
      "epoch": 4.04323354270332,
      "grad_norm": 0.4842938184738159,
      "learning_rate": 1.582610323145727e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 46976992,
      "step": 24785
    },
    {
      "epoch": 4.0440492699241375,
      "grad_norm": 0.19059501588344574,
      "learning_rate": 1.5816971345406035e-05,
      "loss": 0.2012,
      "num_input_tokens_seen": 46986992,
      "step": 24790
    },
    {
      "epoch": 4.044864997144955,
      "grad_norm": 0.27915021777153015,
      "learning_rate": 1.5807840875493225e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 46997120,
      "step": 24795
    },
    {
      "epoch": 4.045680724365772,
      "grad_norm": 0.24044522643089294,
      "learning_rate": 1.5798711823126854e-05,
      "loss": 0.1122,
      "num_input_tokens_seen": 47007360,
      "step": 24800
    },
    {
      "epoch": 4.045680724365772,
      "eval_loss": 0.14090557396411896,
      "eval_runtime": 133.2357,
      "eval_samples_per_second": 20.452,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 47007360,
      "step": 24800
    },
    {
      "epoch": 4.04649645158659,
      "grad_norm": 0.1917208433151245,
      "learning_rate": 1.578958418971477e-05,
      "loss": 0.0252,
      "num_input_tokens_seen": 47015968,
      "step": 24805
    },
    {
      "epoch": 4.047312178807407,
      "grad_norm": 0.2729335427284241,
      "learning_rate": 1.578045797666453e-05,
      "loss": 0.073,
      "num_input_tokens_seen": 47025392,
      "step": 24810
    },
    {
      "epoch": 4.048127906028224,
      "grad_norm": 0.4139059782028198,
      "learning_rate": 1.5771333185383548e-05,
      "loss": 0.0653,
      "num_input_tokens_seen": 47035200,
      "step": 24815
    },
    {
      "epoch": 4.048943633249041,
      "grad_norm": 0.6655532121658325,
      "learning_rate": 1.576220981727895e-05,
      "loss": 0.1122,
      "num_input_tokens_seen": 47045792,
      "step": 24820
    },
    {
      "epoch": 4.049759360469859,
      "grad_norm": 0.3127569556236267,
      "learning_rate": 1.575308787375769e-05,
      "loss": 0.0506,
      "num_input_tokens_seen": 47054064,
      "step": 24825
    },
    {
      "epoch": 4.050575087690676,
      "grad_norm": 0.07558508217334747,
      "learning_rate": 1.5743967356226492e-05,
      "loss": 0.1109,
      "num_input_tokens_seen": 47064352,
      "step": 24830
    },
    {
      "epoch": 4.051390814911493,
      "grad_norm": 1.0580655336380005,
      "learning_rate": 1.5734848266091835e-05,
      "loss": 0.0862,
      "num_input_tokens_seen": 47075232,
      "step": 24835
    },
    {
      "epoch": 4.052206542132311,
      "grad_norm": 1.6781868934631348,
      "learning_rate": 1.572573060476001e-05,
      "loss": 0.2809,
      "num_input_tokens_seen": 47084544,
      "step": 24840
    },
    {
      "epoch": 4.053022269353129,
      "grad_norm": 2.1620733737945557,
      "learning_rate": 1.5716614373637085e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 47093888,
      "step": 24845
    },
    {
      "epoch": 4.053837996573946,
      "grad_norm": 0.3356335461139679,
      "learning_rate": 1.570749957412887e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 47102720,
      "step": 24850
    },
    {
      "epoch": 4.054653723794763,
      "grad_norm": 0.1644306182861328,
      "learning_rate": 1.5698386207641013e-05,
      "loss": 0.1762,
      "num_input_tokens_seen": 47112240,
      "step": 24855
    },
    {
      "epoch": 4.05546945101558,
      "grad_norm": 0.31814202666282654,
      "learning_rate": 1.5689274275578884e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 47122704,
      "step": 24860
    },
    {
      "epoch": 4.056285178236398,
      "grad_norm": 0.3317910432815552,
      "learning_rate": 1.5680163779347667e-05,
      "loss": 0.0876,
      "num_input_tokens_seen": 47132592,
      "step": 24865
    },
    {
      "epoch": 4.057100905457215,
      "grad_norm": 0.8034263849258423,
      "learning_rate": 1.5671054720352327e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 47142000,
      "step": 24870
    },
    {
      "epoch": 4.057916632678032,
      "grad_norm": 0.7805517315864563,
      "learning_rate": 1.566194709999757e-05,
      "loss": 0.073,
      "num_input_tokens_seen": 47152464,
      "step": 24875
    },
    {
      "epoch": 4.058732359898849,
      "grad_norm": 0.112892746925354,
      "learning_rate": 1.5652840919687933e-05,
      "loss": 0.0913,
      "num_input_tokens_seen": 47161552,
      "step": 24880
    },
    {
      "epoch": 4.059548087119667,
      "grad_norm": 0.08022848516702652,
      "learning_rate": 1.5643736180827676e-05,
      "loss": 0.066,
      "num_input_tokens_seen": 47170368,
      "step": 24885
    },
    {
      "epoch": 4.0603638143404845,
      "grad_norm": 0.8446530103683472,
      "learning_rate": 1.5634632884820878e-05,
      "loss": 0.1002,
      "num_input_tokens_seen": 47179872,
      "step": 24890
    },
    {
      "epoch": 4.061179541561302,
      "grad_norm": 0.25228413939476013,
      "learning_rate": 1.5625531033071395e-05,
      "loss": 0.0711,
      "num_input_tokens_seen": 47189808,
      "step": 24895
    },
    {
      "epoch": 4.06199526878212,
      "grad_norm": 1.1873282194137573,
      "learning_rate": 1.5616430626982828e-05,
      "loss": 0.2196,
      "num_input_tokens_seen": 47199712,
      "step": 24900
    },
    {
      "epoch": 4.062810996002937,
      "grad_norm": 0.17626123130321503,
      "learning_rate": 1.5607331667958575e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 47208688,
      "step": 24905
    },
    {
      "epoch": 4.063626723223754,
      "grad_norm": 0.1010182797908783,
      "learning_rate": 1.5598234157401824e-05,
      "loss": 0.0613,
      "num_input_tokens_seen": 47218144,
      "step": 24910
    },
    {
      "epoch": 4.064442450444571,
      "grad_norm": 0.26460501551628113,
      "learning_rate": 1.5589138096715503e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 47227232,
      "step": 24915
    },
    {
      "epoch": 4.065258177665389,
      "grad_norm": 0.27996212244033813,
      "learning_rate": 1.5580043487302365e-05,
      "loss": 0.1089,
      "num_input_tokens_seen": 47236320,
      "step": 24920
    },
    {
      "epoch": 4.066073904886206,
      "grad_norm": 0.7929208874702454,
      "learning_rate": 1.5570950330564888e-05,
      "loss": 0.0848,
      "num_input_tokens_seen": 47244784,
      "step": 24925
    },
    {
      "epoch": 4.066889632107023,
      "grad_norm": 0.13338510692119598,
      "learning_rate": 1.5561858627905367e-05,
      "loss": 0.0967,
      "num_input_tokens_seen": 47255168,
      "step": 24930
    },
    {
      "epoch": 4.0677053593278405,
      "grad_norm": 0.48111197352409363,
      "learning_rate": 1.5552768380725857e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 47265136,
      "step": 24935
    },
    {
      "epoch": 4.0685210865486585,
      "grad_norm": 0.4275049567222595,
      "learning_rate": 1.5543679590428183e-05,
      "loss": 0.0535,
      "num_input_tokens_seen": 47274016,
      "step": 24940
    },
    {
      "epoch": 4.069336813769476,
      "grad_norm": 0.45669853687286377,
      "learning_rate": 1.5534592258413943e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 47283440,
      "step": 24945
    },
    {
      "epoch": 4.070152540990293,
      "grad_norm": 1.2530990839004517,
      "learning_rate": 1.5525506386084538e-05,
      "loss": 0.2149,
      "num_input_tokens_seen": 47292576,
      "step": 24950
    },
    {
      "epoch": 4.07096826821111,
      "grad_norm": 0.9829491972923279,
      "learning_rate": 1.55164219748411e-05,
      "loss": 0.2945,
      "num_input_tokens_seen": 47303168,
      "step": 24955
    },
    {
      "epoch": 4.071783995431928,
      "grad_norm": 0.5310719013214111,
      "learning_rate": 1.550733902608459e-05,
      "loss": 0.0436,
      "num_input_tokens_seen": 47312656,
      "step": 24960
    },
    {
      "epoch": 4.072599722652745,
      "grad_norm": 0.22735655307769775,
      "learning_rate": 1.549825754121568e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 47322016,
      "step": 24965
    },
    {
      "epoch": 4.073415449873562,
      "grad_norm": 0.7490960955619812,
      "learning_rate": 1.5489177521634864e-05,
      "loss": 0.1691,
      "num_input_tokens_seen": 47331792,
      "step": 24970
    },
    {
      "epoch": 4.074231177094379,
      "grad_norm": 0.29775822162628174,
      "learning_rate": 1.5480098968742402e-05,
      "loss": 0.1908,
      "num_input_tokens_seen": 47341312,
      "step": 24975
    },
    {
      "epoch": 4.075046904315197,
      "grad_norm": 0.15743322670459747,
      "learning_rate": 1.5471021883938304e-05,
      "loss": 0.0681,
      "num_input_tokens_seen": 47351040,
      "step": 24980
    },
    {
      "epoch": 4.0758626315360145,
      "grad_norm": 1.3147575855255127,
      "learning_rate": 1.546194626862238e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 47361648,
      "step": 24985
    },
    {
      "epoch": 4.076678358756832,
      "grad_norm": 1.3344224691390991,
      "learning_rate": 1.5452872124194216e-05,
      "loss": 0.0665,
      "num_input_tokens_seen": 47371968,
      "step": 24990
    },
    {
      "epoch": 4.077494085977649,
      "grad_norm": 0.5246835350990295,
      "learning_rate": 1.5443799452053136e-05,
      "loss": 0.1593,
      "num_input_tokens_seen": 47381968,
      "step": 24995
    },
    {
      "epoch": 4.078309813198467,
      "grad_norm": 0.4176199734210968,
      "learning_rate": 1.543472825359828e-05,
      "loss": 0.1824,
      "num_input_tokens_seen": 47391600,
      "step": 25000
    },
    {
      "epoch": 4.078309813198467,
      "eval_loss": 0.1419147402048111,
      "eval_runtime": 133.3185,
      "eval_samples_per_second": 20.44,
      "eval_steps_per_second": 10.224,
      "num_input_tokens_seen": 47391600,
      "step": 25000
    },
    {
      "epoch": 4.079125540419284,
      "grad_norm": 1.8115026950836182,
      "learning_rate": 1.5425658530228522e-05,
      "loss": 0.21,
      "num_input_tokens_seen": 47401120,
      "step": 25005
    },
    {
      "epoch": 4.079941267640101,
      "grad_norm": 0.1416134387254715,
      "learning_rate": 1.5416590283342546e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 47410432,
      "step": 25010
    },
    {
      "epoch": 4.080756994860918,
      "grad_norm": 0.586406409740448,
      "learning_rate": 1.5407523514338783e-05,
      "loss": 0.0435,
      "num_input_tokens_seen": 47420336,
      "step": 25015
    },
    {
      "epoch": 4.081572722081736,
      "grad_norm": 0.9284112453460693,
      "learning_rate": 1.539845822461543e-05,
      "loss": 0.1158,
      "num_input_tokens_seen": 47429744,
      "step": 25020
    },
    {
      "epoch": 4.082388449302553,
      "grad_norm": 0.32081860303878784,
      "learning_rate": 1.538939441557048e-05,
      "loss": 0.1768,
      "num_input_tokens_seen": 47438400,
      "step": 25025
    },
    {
      "epoch": 4.08320417652337,
      "grad_norm": 0.13701562583446503,
      "learning_rate": 1.5380332088601696e-05,
      "loss": 0.2953,
      "num_input_tokens_seen": 47447216,
      "step": 25030
    },
    {
      "epoch": 4.0840199037441876,
      "grad_norm": 0.710601806640625,
      "learning_rate": 1.537127124510658e-05,
      "loss": 0.0576,
      "num_input_tokens_seen": 47457840,
      "step": 25035
    },
    {
      "epoch": 4.084835630965006,
      "grad_norm": 1.5856952667236328,
      "learning_rate": 1.5362211886482457e-05,
      "loss": 0.2557,
      "num_input_tokens_seen": 47467280,
      "step": 25040
    },
    {
      "epoch": 4.085651358185823,
      "grad_norm": 0.1923631876707077,
      "learning_rate": 1.5353154014126363e-05,
      "loss": 0.2882,
      "num_input_tokens_seen": 47476656,
      "step": 25045
    },
    {
      "epoch": 4.08646708540664,
      "grad_norm": 0.13887476921081543,
      "learning_rate": 1.534409762943515e-05,
      "loss": 0.096,
      "num_input_tokens_seen": 47485568,
      "step": 25050
    },
    {
      "epoch": 4.087282812627457,
      "grad_norm": 0.23718944191932678,
      "learning_rate": 1.5335042733805438e-05,
      "loss": 0.0557,
      "num_input_tokens_seen": 47494832,
      "step": 25055
    },
    {
      "epoch": 4.088098539848275,
      "grad_norm": 0.1724029928445816,
      "learning_rate": 1.532598932863358e-05,
      "loss": 0.0882,
      "num_input_tokens_seen": 47504416,
      "step": 25060
    },
    {
      "epoch": 4.088914267069092,
      "grad_norm": 0.31518763303756714,
      "learning_rate": 1.531693741531574e-05,
      "loss": 0.0449,
      "num_input_tokens_seen": 47514544,
      "step": 25065
    },
    {
      "epoch": 4.089729994289909,
      "grad_norm": 0.23478154838085175,
      "learning_rate": 1.5307886995247844e-05,
      "loss": 0.0898,
      "num_input_tokens_seen": 47522992,
      "step": 25070
    },
    {
      "epoch": 4.090545721510727,
      "grad_norm": 0.08410820364952087,
      "learning_rate": 1.529883806982557e-05,
      "loss": 0.0452,
      "num_input_tokens_seen": 47532256,
      "step": 25075
    },
    {
      "epoch": 4.091361448731544,
      "grad_norm": 0.07886224240064621,
      "learning_rate": 1.5289790640444376e-05,
      "loss": 0.0613,
      "num_input_tokens_seen": 47543424,
      "step": 25080
    },
    {
      "epoch": 4.0921771759523615,
      "grad_norm": 0.306345134973526,
      "learning_rate": 1.5280744708499494e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 47553104,
      "step": 25085
    },
    {
      "epoch": 4.092992903173179,
      "grad_norm": 0.4837210476398468,
      "learning_rate": 1.527170027538591e-05,
      "loss": 0.1071,
      "num_input_tokens_seen": 47562704,
      "step": 25090
    },
    {
      "epoch": 4.093808630393997,
      "grad_norm": 0.1159185841679573,
      "learning_rate": 1.5262657342498407e-05,
      "loss": 0.12,
      "num_input_tokens_seen": 47572928,
      "step": 25095
    },
    {
      "epoch": 4.094624357614814,
      "grad_norm": 1.2706447839736938,
      "learning_rate": 1.52536159112315e-05,
      "loss": 0.1335,
      "num_input_tokens_seen": 47581712,
      "step": 25100
    },
    {
      "epoch": 4.095440084835631,
      "grad_norm": 0.2653568387031555,
      "learning_rate": 1.5244575982979497e-05,
      "loss": 0.0766,
      "num_input_tokens_seen": 47590640,
      "step": 25105
    },
    {
      "epoch": 4.096255812056448,
      "grad_norm": 0.17762015759944916,
      "learning_rate": 1.5235537559136487e-05,
      "loss": 0.2041,
      "num_input_tokens_seen": 47599376,
      "step": 25110
    },
    {
      "epoch": 4.097071539277266,
      "grad_norm": 0.0676494687795639,
      "learning_rate": 1.5226500641096286e-05,
      "loss": 0.0778,
      "num_input_tokens_seen": 47607952,
      "step": 25115
    },
    {
      "epoch": 4.097887266498083,
      "grad_norm": 0.09675753116607666,
      "learning_rate": 1.5217465230252509e-05,
      "loss": 0.0597,
      "num_input_tokens_seen": 47618736,
      "step": 25120
    },
    {
      "epoch": 4.0987029937189,
      "grad_norm": 0.8748628497123718,
      "learning_rate": 1.5208431327998523e-05,
      "loss": 0.1205,
      "num_input_tokens_seen": 47627312,
      "step": 25125
    },
    {
      "epoch": 4.0995187209397175,
      "grad_norm": 0.2289469838142395,
      "learning_rate": 1.5199398935727477e-05,
      "loss": 0.1232,
      "num_input_tokens_seen": 47637808,
      "step": 25130
    },
    {
      "epoch": 4.1003344481605355,
      "grad_norm": 0.5234981179237366,
      "learning_rate": 1.5190368054832282e-05,
      "loss": 0.1146,
      "num_input_tokens_seen": 47647360,
      "step": 25135
    },
    {
      "epoch": 4.101150175381353,
      "grad_norm": 0.10720696300268173,
      "learning_rate": 1.5181338686705601e-05,
      "loss": 0.0486,
      "num_input_tokens_seen": 47657312,
      "step": 25140
    },
    {
      "epoch": 4.10196590260217,
      "grad_norm": 0.1931077092885971,
      "learning_rate": 1.5172310832739889e-05,
      "loss": 0.1413,
      "num_input_tokens_seen": 47666592,
      "step": 25145
    },
    {
      "epoch": 4.102781629822987,
      "grad_norm": 0.947246789932251,
      "learning_rate": 1.5163284494327346e-05,
      "loss": 0.1114,
      "num_input_tokens_seen": 47676096,
      "step": 25150
    },
    {
      "epoch": 4.103597357043805,
      "grad_norm": 0.02874087542295456,
      "learning_rate": 1.5154259672859952e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 47684928,
      "step": 25155
    },
    {
      "epoch": 4.104413084264622,
      "grad_norm": 0.6851835250854492,
      "learning_rate": 1.5145236369729452e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 47693872,
      "step": 25160
    },
    {
      "epoch": 4.105228811485439,
      "grad_norm": 0.4639970660209656,
      "learning_rate": 1.5136214586327335e-05,
      "loss": 0.0768,
      "num_input_tokens_seen": 47702560,
      "step": 25165
    },
    {
      "epoch": 4.106044538706256,
      "grad_norm": 1.51515793800354,
      "learning_rate": 1.5127194324044885e-05,
      "loss": 0.2104,
      "num_input_tokens_seen": 47711248,
      "step": 25170
    },
    {
      "epoch": 4.106860265927074,
      "grad_norm": 1.1315288543701172,
      "learning_rate": 1.5118175584273148e-05,
      "loss": 0.0912,
      "num_input_tokens_seen": 47719968,
      "step": 25175
    },
    {
      "epoch": 4.1076759931478914,
      "grad_norm": 0.7592498064041138,
      "learning_rate": 1.5109158368402909e-05,
      "loss": 0.2188,
      "num_input_tokens_seen": 47729712,
      "step": 25180
    },
    {
      "epoch": 4.108491720368709,
      "grad_norm": 0.5723746418952942,
      "learning_rate": 1.5100142677824753e-05,
      "loss": 0.1851,
      "num_input_tokens_seen": 47739632,
      "step": 25185
    },
    {
      "epoch": 4.109307447589526,
      "grad_norm": 0.6981586217880249,
      "learning_rate": 1.509112851392901e-05,
      "loss": 0.1138,
      "num_input_tokens_seen": 47748576,
      "step": 25190
    },
    {
      "epoch": 4.110123174810344,
      "grad_norm": 0.6043882966041565,
      "learning_rate": 1.5082115878105763e-05,
      "loss": 0.2616,
      "num_input_tokens_seen": 47758384,
      "step": 25195
    },
    {
      "epoch": 4.110938902031161,
      "grad_norm": 0.3440313935279846,
      "learning_rate": 1.5073104771744892e-05,
      "loss": 0.0632,
      "num_input_tokens_seen": 47768320,
      "step": 25200
    },
    {
      "epoch": 4.110938902031161,
      "eval_loss": 0.14145614206790924,
      "eval_runtime": 133.2907,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 47768320,
      "step": 25200
    },
    {
      "epoch": 4.111754629251978,
      "grad_norm": 2.493425130844116,
      "learning_rate": 1.5064095196236006e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 47776336,
      "step": 25205
    },
    {
      "epoch": 4.112570356472795,
      "grad_norm": 0.7994717955589294,
      "learning_rate": 1.50550871529685e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 47786576,
      "step": 25210
    },
    {
      "epoch": 4.113386083693613,
      "grad_norm": 0.38653749227523804,
      "learning_rate": 1.5046080643331546e-05,
      "loss": 0.0819,
      "num_input_tokens_seen": 47794544,
      "step": 25215
    },
    {
      "epoch": 4.11420181091443,
      "grad_norm": 0.3826112747192383,
      "learning_rate": 1.5037075668714028e-05,
      "loss": 0.0779,
      "num_input_tokens_seen": 47805456,
      "step": 25220
    },
    {
      "epoch": 4.115017538135247,
      "grad_norm": 0.04743867367506027,
      "learning_rate": 1.5028072230504656e-05,
      "loss": 0.0421,
      "num_input_tokens_seen": 47814288,
      "step": 25225
    },
    {
      "epoch": 4.1158332653560645,
      "grad_norm": 0.42388463020324707,
      "learning_rate": 1.5019070330091861e-05,
      "loss": 0.0742,
      "num_input_tokens_seen": 47823648,
      "step": 25230
    },
    {
      "epoch": 4.1166489925768825,
      "grad_norm": 0.03745110705494881,
      "learning_rate": 1.5010069968863843e-05,
      "loss": 0.1092,
      "num_input_tokens_seen": 47833440,
      "step": 25235
    },
    {
      "epoch": 4.1174647197977,
      "grad_norm": 0.36183539032936096,
      "learning_rate": 1.5001071148208584e-05,
      "loss": 0.0779,
      "num_input_tokens_seen": 47843424,
      "step": 25240
    },
    {
      "epoch": 4.118280447018517,
      "grad_norm": 1.7957255840301514,
      "learning_rate": 1.49920738695138e-05,
      "loss": 0.1888,
      "num_input_tokens_seen": 47853264,
      "step": 25245
    },
    {
      "epoch": 4.119096174239334,
      "grad_norm": 0.33418750762939453,
      "learning_rate": 1.4983078134166995e-05,
      "loss": 0.1754,
      "num_input_tokens_seen": 47863312,
      "step": 25250
    },
    {
      "epoch": 4.119911901460152,
      "grad_norm": 0.3555379807949066,
      "learning_rate": 1.4974083943555428e-05,
      "loss": 0.031,
      "num_input_tokens_seen": 47873392,
      "step": 25255
    },
    {
      "epoch": 4.120727628680969,
      "grad_norm": 0.25739026069641113,
      "learning_rate": 1.496509129906611e-05,
      "loss": 0.1911,
      "num_input_tokens_seen": 47883728,
      "step": 25260
    },
    {
      "epoch": 4.121543355901786,
      "grad_norm": 0.22221940755844116,
      "learning_rate": 1.4956100202085809e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 47892032,
      "step": 25265
    },
    {
      "epoch": 4.122359083122603,
      "grad_norm": 0.8228124380111694,
      "learning_rate": 1.4947110654001093e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 47902080,
      "step": 25270
    },
    {
      "epoch": 4.123174810343421,
      "grad_norm": 0.15955013036727905,
      "learning_rate": 1.4938122656198234e-05,
      "loss": 0.1733,
      "num_input_tokens_seen": 47910784,
      "step": 25275
    },
    {
      "epoch": 4.1239905375642385,
      "grad_norm": 0.5657938122749329,
      "learning_rate": 1.4929136210063316e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 47919824,
      "step": 25280
    },
    {
      "epoch": 4.124806264785056,
      "grad_norm": 0.0747467502951622,
      "learning_rate": 1.4920151316982146e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 47928720,
      "step": 25285
    },
    {
      "epoch": 4.125621992005874,
      "grad_norm": 0.18703092634677887,
      "learning_rate": 1.4911167978340312e-05,
      "loss": 0.1038,
      "num_input_tokens_seen": 47937904,
      "step": 25290
    },
    {
      "epoch": 4.126437719226691,
      "grad_norm": 0.632620096206665,
      "learning_rate": 1.4902186195523166e-05,
      "loss": 0.0541,
      "num_input_tokens_seen": 47947984,
      "step": 25295
    },
    {
      "epoch": 4.127253446447508,
      "grad_norm": 1.8714189529418945,
      "learning_rate": 1.4893205969915805e-05,
      "loss": 0.1106,
      "num_input_tokens_seen": 47957792,
      "step": 25300
    },
    {
      "epoch": 4.128069173668325,
      "grad_norm": 0.1496637612581253,
      "learning_rate": 1.4884227302903086e-05,
      "loss": 0.1598,
      "num_input_tokens_seen": 47967696,
      "step": 25305
    },
    {
      "epoch": 4.128884900889143,
      "grad_norm": 0.09659504145383835,
      "learning_rate": 1.4875250195869653e-05,
      "loss": 0.061,
      "num_input_tokens_seen": 47975520,
      "step": 25310
    },
    {
      "epoch": 4.12970062810996,
      "grad_norm": 1.784403681755066,
      "learning_rate": 1.4866274650199862e-05,
      "loss": 0.2269,
      "num_input_tokens_seen": 47984624,
      "step": 25315
    },
    {
      "epoch": 4.130516355330777,
      "grad_norm": 0.30822572112083435,
      "learning_rate": 1.485730066727788e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 47994512,
      "step": 25320
    },
    {
      "epoch": 4.1313320825515945,
      "grad_norm": 0.07316015660762787,
      "learning_rate": 1.4848328248487586e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 48003200,
      "step": 25325
    },
    {
      "epoch": 4.1321478097724125,
      "grad_norm": 0.2221480756998062,
      "learning_rate": 1.4839357395212656e-05,
      "loss": 0.0881,
      "num_input_tokens_seen": 48012128,
      "step": 25330
    },
    {
      "epoch": 4.13296353699323,
      "grad_norm": 0.30372536182403564,
      "learning_rate": 1.4830388108836502e-05,
      "loss": 0.1797,
      "num_input_tokens_seen": 48021680,
      "step": 25335
    },
    {
      "epoch": 4.133779264214047,
      "grad_norm": 0.1212795078754425,
      "learning_rate": 1.4821420390742299e-05,
      "loss": 0.0298,
      "num_input_tokens_seen": 48031360,
      "step": 25340
    },
    {
      "epoch": 4.134594991434864,
      "grad_norm": 0.4595811367034912,
      "learning_rate": 1.4812454242312979e-05,
      "loss": 0.1002,
      "num_input_tokens_seen": 48040416,
      "step": 25345
    },
    {
      "epoch": 4.135410718655682,
      "grad_norm": 0.15096566081047058,
      "learning_rate": 1.4803489664931253e-05,
      "loss": 0.0204,
      "num_input_tokens_seen": 48049408,
      "step": 25350
    },
    {
      "epoch": 4.136226445876499,
      "grad_norm": 0.21900542080402374,
      "learning_rate": 1.4794526659979544e-05,
      "loss": 0.2027,
      "num_input_tokens_seen": 48058176,
      "step": 25355
    },
    {
      "epoch": 4.137042173097316,
      "grad_norm": 1.157753586769104,
      "learning_rate": 1.4785565228840086e-05,
      "loss": 0.2071,
      "num_input_tokens_seen": 48066992,
      "step": 25360
    },
    {
      "epoch": 4.137857900318133,
      "grad_norm": 0.12582314014434814,
      "learning_rate": 1.4776605372894819e-05,
      "loss": 0.1622,
      "num_input_tokens_seen": 48076864,
      "step": 25365
    },
    {
      "epoch": 4.138673627538951,
      "grad_norm": 0.7628117203712463,
      "learning_rate": 1.4767647093525488e-05,
      "loss": 0.1059,
      "num_input_tokens_seen": 48086496,
      "step": 25370
    },
    {
      "epoch": 4.139489354759768,
      "grad_norm": 1.3084897994995117,
      "learning_rate": 1.4758690392113566e-05,
      "loss": 0.2429,
      "num_input_tokens_seen": 48095840,
      "step": 25375
    },
    {
      "epoch": 4.1403050819805856,
      "grad_norm": 1.1198846101760864,
      "learning_rate": 1.4749735270040276e-05,
      "loss": 0.157,
      "num_input_tokens_seen": 48105264,
      "step": 25380
    },
    {
      "epoch": 4.141120809201403,
      "grad_norm": 0.02896707132458687,
      "learning_rate": 1.4740781728686623e-05,
      "loss": 0.0749,
      "num_input_tokens_seen": 48115088,
      "step": 25385
    },
    {
      "epoch": 4.141936536422221,
      "grad_norm": 0.3019353151321411,
      "learning_rate": 1.4731829769433358e-05,
      "loss": 0.1519,
      "num_input_tokens_seen": 48123456,
      "step": 25390
    },
    {
      "epoch": 4.142752263643038,
      "grad_norm": 0.46587586402893066,
      "learning_rate": 1.4722879393660976e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 48133488,
      "step": 25395
    },
    {
      "epoch": 4.143567990863855,
      "grad_norm": 0.4185059666633606,
      "learning_rate": 1.4713930602749748e-05,
      "loss": 0.1647,
      "num_input_tokens_seen": 48143424,
      "step": 25400
    },
    {
      "epoch": 4.143567990863855,
      "eval_loss": 0.1448117196559906,
      "eval_runtime": 133.2889,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 48143424,
      "step": 25400
    },
    {
      "epoch": 4.144383718084672,
      "grad_norm": 1.9235844612121582,
      "learning_rate": 1.470498339807968e-05,
      "loss": 0.2398,
      "num_input_tokens_seen": 48152160,
      "step": 25405
    },
    {
      "epoch": 4.14519944530549,
      "grad_norm": 1.241614580154419,
      "learning_rate": 1.4696037781030542e-05,
      "loss": 0.2131,
      "num_input_tokens_seen": 48161312,
      "step": 25410
    },
    {
      "epoch": 4.146015172526307,
      "grad_norm": 0.4720090925693512,
      "learning_rate": 1.4687093752981876e-05,
      "loss": 0.2171,
      "num_input_tokens_seen": 48171008,
      "step": 25415
    },
    {
      "epoch": 4.146830899747124,
      "grad_norm": 0.12637141346931458,
      "learning_rate": 1.4678151315312943e-05,
      "loss": 0.0272,
      "num_input_tokens_seen": 48180368,
      "step": 25420
    },
    {
      "epoch": 4.1476466269679415,
      "grad_norm": 0.1231917142868042,
      "learning_rate": 1.4669210469402789e-05,
      "loss": 0.0458,
      "num_input_tokens_seen": 48190144,
      "step": 25425
    },
    {
      "epoch": 4.1484623541887595,
      "grad_norm": 0.36392414569854736,
      "learning_rate": 1.4660271216630218e-05,
      "loss": 0.1193,
      "num_input_tokens_seen": 48200672,
      "step": 25430
    },
    {
      "epoch": 4.149278081409577,
      "grad_norm": 0.3716631531715393,
      "learning_rate": 1.4651333558373748e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 48210464,
      "step": 25435
    },
    {
      "epoch": 4.150093808630394,
      "grad_norm": 0.18386605381965637,
      "learning_rate": 1.4642397496011707e-05,
      "loss": 0.0957,
      "num_input_tokens_seen": 48218944,
      "step": 25440
    },
    {
      "epoch": 4.150909535851211,
      "grad_norm": 0.23937924206256866,
      "learning_rate": 1.4633463030922129e-05,
      "loss": 0.1172,
      "num_input_tokens_seen": 48227888,
      "step": 25445
    },
    {
      "epoch": 4.151725263072029,
      "grad_norm": 0.5342777967453003,
      "learning_rate": 1.462453016448282e-05,
      "loss": 0.0482,
      "num_input_tokens_seen": 48237472,
      "step": 25450
    },
    {
      "epoch": 4.152540990292846,
      "grad_norm": 0.07495041191577911,
      "learning_rate": 1.4615598898071354e-05,
      "loss": 0.0998,
      "num_input_tokens_seen": 48247872,
      "step": 25455
    },
    {
      "epoch": 4.153356717513663,
      "grad_norm": 1.3037307262420654,
      "learning_rate": 1.4606669233065026e-05,
      "loss": 0.1807,
      "num_input_tokens_seen": 48257008,
      "step": 25460
    },
    {
      "epoch": 4.154172444734481,
      "grad_norm": 0.1795211136341095,
      "learning_rate": 1.4597741170840914e-05,
      "loss": 0.0348,
      "num_input_tokens_seen": 48266048,
      "step": 25465
    },
    {
      "epoch": 4.154988171955298,
      "grad_norm": 0.8986180424690247,
      "learning_rate": 1.4588814712775853e-05,
      "loss": 0.0921,
      "num_input_tokens_seen": 48276048,
      "step": 25470
    },
    {
      "epoch": 4.1558038991761155,
      "grad_norm": 0.42890679836273193,
      "learning_rate": 1.4579889860246382e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 48286432,
      "step": 25475
    },
    {
      "epoch": 4.156619626396933,
      "grad_norm": 0.29744651913642883,
      "learning_rate": 1.457096661462885e-05,
      "loss": 0.0327,
      "num_input_tokens_seen": 48296944,
      "step": 25480
    },
    {
      "epoch": 4.157435353617751,
      "grad_norm": 0.15923798084259033,
      "learning_rate": 1.4562044977299322e-05,
      "loss": 0.2067,
      "num_input_tokens_seen": 48306064,
      "step": 25485
    },
    {
      "epoch": 4.158251080838568,
      "grad_norm": 0.019180743023753166,
      "learning_rate": 1.4553124949633623e-05,
      "loss": 0.1128,
      "num_input_tokens_seen": 48315696,
      "step": 25490
    },
    {
      "epoch": 4.159066808059385,
      "grad_norm": 0.771841287612915,
      "learning_rate": 1.4544206533007354e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 48323808,
      "step": 25495
    },
    {
      "epoch": 4.159882535280202,
      "grad_norm": 1.1438711881637573,
      "learning_rate": 1.4535289728795821e-05,
      "loss": 0.1934,
      "num_input_tokens_seen": 48333632,
      "step": 25500
    },
    {
      "epoch": 4.16069826250102,
      "grad_norm": 0.23338839411735535,
      "learning_rate": 1.4526374538374132e-05,
      "loss": 0.0774,
      "num_input_tokens_seen": 48343888,
      "step": 25505
    },
    {
      "epoch": 4.161513989721837,
      "grad_norm": 1.0641096830368042,
      "learning_rate": 1.4517460963117097e-05,
      "loss": 0.1158,
      "num_input_tokens_seen": 48353104,
      "step": 25510
    },
    {
      "epoch": 4.162329716942654,
      "grad_norm": 1.2206650972366333,
      "learning_rate": 1.4508549004399314e-05,
      "loss": 0.1877,
      "num_input_tokens_seen": 48362304,
      "step": 25515
    },
    {
      "epoch": 4.163145444163471,
      "grad_norm": 0.6187209486961365,
      "learning_rate": 1.449963866359513e-05,
      "loss": 0.1701,
      "num_input_tokens_seen": 48371824,
      "step": 25520
    },
    {
      "epoch": 4.1639611713842895,
      "grad_norm": 1.3500025272369385,
      "learning_rate": 1.4490729942078607e-05,
      "loss": 0.1055,
      "num_input_tokens_seen": 48381808,
      "step": 25525
    },
    {
      "epoch": 4.164776898605107,
      "grad_norm": 0.04440561309456825,
      "learning_rate": 1.4481822841223608e-05,
      "loss": 0.2223,
      "num_input_tokens_seen": 48391296,
      "step": 25530
    },
    {
      "epoch": 4.165592625825924,
      "grad_norm": 0.10795129090547562,
      "learning_rate": 1.4472917362403704e-05,
      "loss": 0.2445,
      "num_input_tokens_seen": 48400656,
      "step": 25535
    },
    {
      "epoch": 4.166408353046741,
      "grad_norm": 0.4508681893348694,
      "learning_rate": 1.4464013506992224e-05,
      "loss": 0.1434,
      "num_input_tokens_seen": 48409744,
      "step": 25540
    },
    {
      "epoch": 4.167224080267559,
      "grad_norm": 0.8611025214195251,
      "learning_rate": 1.4455111276362277e-05,
      "loss": 0.204,
      "num_input_tokens_seen": 48418608,
      "step": 25545
    },
    {
      "epoch": 4.168039807488376,
      "grad_norm": 0.7020213603973389,
      "learning_rate": 1.4446210671886676e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 48427104,
      "step": 25550
    },
    {
      "epoch": 4.168855534709193,
      "grad_norm": 0.3756517469882965,
      "learning_rate": 1.4437311694938015e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 48437584,
      "step": 25555
    },
    {
      "epoch": 4.16967126193001,
      "grad_norm": 2.328500509262085,
      "learning_rate": 1.442841434688864e-05,
      "loss": 0.1807,
      "num_input_tokens_seen": 48447664,
      "step": 25560
    },
    {
      "epoch": 4.170486989150828,
      "grad_norm": 0.7998062968254089,
      "learning_rate": 1.4419518629110615e-05,
      "loss": 0.1999,
      "num_input_tokens_seen": 48458288,
      "step": 25565
    },
    {
      "epoch": 4.171302716371645,
      "grad_norm": 0.6007382869720459,
      "learning_rate": 1.4410624542975778e-05,
      "loss": 0.0244,
      "num_input_tokens_seen": 48466096,
      "step": 25570
    },
    {
      "epoch": 4.1721184435924625,
      "grad_norm": 0.7679567933082581,
      "learning_rate": 1.4401732089855724e-05,
      "loss": 0.0735,
      "num_input_tokens_seen": 48475888,
      "step": 25575
    },
    {
      "epoch": 4.17293417081328,
      "grad_norm": 0.3971518874168396,
      "learning_rate": 1.4392841271121754e-05,
      "loss": 0.0601,
      "num_input_tokens_seen": 48485648,
      "step": 25580
    },
    {
      "epoch": 4.173749898034098,
      "grad_norm": 1.0626888275146484,
      "learning_rate": 1.438395208814497e-05,
      "loss": 0.1955,
      "num_input_tokens_seen": 48494864,
      "step": 25585
    },
    {
      "epoch": 4.174565625254915,
      "grad_norm": 0.10934080928564072,
      "learning_rate": 1.4375064542296174e-05,
      "loss": 0.0902,
      "num_input_tokens_seen": 48503824,
      "step": 25590
    },
    {
      "epoch": 4.175381352475732,
      "grad_norm": 1.7301937341690063,
      "learning_rate": 1.4366178634945946e-05,
      "loss": 0.1777,
      "num_input_tokens_seen": 48514432,
      "step": 25595
    },
    {
      "epoch": 4.176197079696549,
      "grad_norm": 0.5016588568687439,
      "learning_rate": 1.4357294367464616e-05,
      "loss": 0.0713,
      "num_input_tokens_seen": 48524368,
      "step": 25600
    },
    {
      "epoch": 4.176197079696549,
      "eval_loss": 0.14253157377243042,
      "eval_runtime": 133.3492,
      "eval_samples_per_second": 20.435,
      "eval_steps_per_second": 10.221,
      "num_input_tokens_seen": 48524368,
      "step": 25600
    },
    {
      "epoch": 4.177012806917367,
      "grad_norm": 0.8619951009750366,
      "learning_rate": 1.434841174122224e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 48533120,
      "step": 25605
    },
    {
      "epoch": 4.177828534138184,
      "grad_norm": 0.3166031241416931,
      "learning_rate": 1.4339530757588615e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 48542848,
      "step": 25610
    },
    {
      "epoch": 4.178644261359001,
      "grad_norm": 1.4346449375152588,
      "learning_rate": 1.433065141793333e-05,
      "loss": 0.1144,
      "num_input_tokens_seen": 48552192,
      "step": 25615
    },
    {
      "epoch": 4.1794599885798185,
      "grad_norm": 1.4396593570709229,
      "learning_rate": 1.4321773723625665e-05,
      "loss": 0.0827,
      "num_input_tokens_seen": 48561792,
      "step": 25620
    },
    {
      "epoch": 4.1802757158006365,
      "grad_norm": 0.10527782887220383,
      "learning_rate": 1.4312897676034693e-05,
      "loss": 0.1939,
      "num_input_tokens_seen": 48571376,
      "step": 25625
    },
    {
      "epoch": 4.181091443021454,
      "grad_norm": 1.50812566280365,
      "learning_rate": 1.4304023276529188e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 48581984,
      "step": 25630
    },
    {
      "epoch": 4.181907170242271,
      "grad_norm": 0.23217974603176117,
      "learning_rate": 1.4295150526477712e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 48591072,
      "step": 25635
    },
    {
      "epoch": 4.182722897463089,
      "grad_norm": 0.2367866486310959,
      "learning_rate": 1.4286279427248562e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 48599776,
      "step": 25640
    },
    {
      "epoch": 4.183538624683906,
      "grad_norm": 0.3578239977359772,
      "learning_rate": 1.4277409980209747e-05,
      "loss": 0.0512,
      "num_input_tokens_seen": 48608960,
      "step": 25645
    },
    {
      "epoch": 4.184354351904723,
      "grad_norm": 0.08148783445358276,
      "learning_rate": 1.4268542186729061e-05,
      "loss": 0.2457,
      "num_input_tokens_seen": 48617696,
      "step": 25650
    },
    {
      "epoch": 4.18517007912554,
      "grad_norm": 0.962177574634552,
      "learning_rate": 1.4259676048174043e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 48627008,
      "step": 25655
    },
    {
      "epoch": 4.185985806346358,
      "grad_norm": 0.46808305382728577,
      "learning_rate": 1.4250811565911937e-05,
      "loss": 0.0424,
      "num_input_tokens_seen": 48636096,
      "step": 25660
    },
    {
      "epoch": 4.186801533567175,
      "grad_norm": 0.13779428601264954,
      "learning_rate": 1.4241948741309782e-05,
      "loss": 0.1875,
      "num_input_tokens_seen": 48645376,
      "step": 25665
    },
    {
      "epoch": 4.1876172607879925,
      "grad_norm": 0.6043750643730164,
      "learning_rate": 1.4233087575734317e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 48653632,
      "step": 25670
    },
    {
      "epoch": 4.18843298800881,
      "grad_norm": 0.5838006138801575,
      "learning_rate": 1.422422807055206e-05,
      "loss": 0.0297,
      "num_input_tokens_seen": 48664544,
      "step": 25675
    },
    {
      "epoch": 4.189248715229628,
      "grad_norm": 1.4528696537017822,
      "learning_rate": 1.4215370227129243e-05,
      "loss": 0.224,
      "num_input_tokens_seen": 48673744,
      "step": 25680
    },
    {
      "epoch": 4.190064442450445,
      "grad_norm": 1.2936978340148926,
      "learning_rate": 1.4206514046831876e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 48684112,
      "step": 25685
    },
    {
      "epoch": 4.190880169671262,
      "grad_norm": 0.13449108600616455,
      "learning_rate": 1.419765953102567e-05,
      "loss": 0.0898,
      "num_input_tokens_seen": 48693376,
      "step": 25690
    },
    {
      "epoch": 4.191695896892079,
      "grad_norm": 0.24693527817726135,
      "learning_rate": 1.4188806681076125e-05,
      "loss": 0.0632,
      "num_input_tokens_seen": 48702832,
      "step": 25695
    },
    {
      "epoch": 4.192511624112897,
      "grad_norm": 0.06368295848369598,
      "learning_rate": 1.4179955498348443e-05,
      "loss": 0.061,
      "num_input_tokens_seen": 48711488,
      "step": 25700
    },
    {
      "epoch": 4.193327351333714,
      "grad_norm": 0.33494746685028076,
      "learning_rate": 1.4171105984207605e-05,
      "loss": 0.0764,
      "num_input_tokens_seen": 48720992,
      "step": 25705
    },
    {
      "epoch": 4.194143078554531,
      "grad_norm": 0.2988407015800476,
      "learning_rate": 1.4162258140018304e-05,
      "loss": 0.0598,
      "num_input_tokens_seen": 48731520,
      "step": 25710
    },
    {
      "epoch": 4.194958805775348,
      "grad_norm": 1.0163177251815796,
      "learning_rate": 1.4153411967144986e-05,
      "loss": 0.2835,
      "num_input_tokens_seen": 48741472,
      "step": 25715
    },
    {
      "epoch": 4.195774532996166,
      "grad_norm": 0.954037606716156,
      "learning_rate": 1.4144567466951864e-05,
      "loss": 0.0995,
      "num_input_tokens_seen": 48750640,
      "step": 25720
    },
    {
      "epoch": 4.196590260216984,
      "grad_norm": 0.846411406993866,
      "learning_rate": 1.4135724640802844e-05,
      "loss": 0.0683,
      "num_input_tokens_seen": 48759792,
      "step": 25725
    },
    {
      "epoch": 4.197405987437801,
      "grad_norm": 2.244664430618286,
      "learning_rate": 1.4126883490061615e-05,
      "loss": 0.2649,
      "num_input_tokens_seen": 48769952,
      "step": 25730
    },
    {
      "epoch": 4.198221714658618,
      "grad_norm": 0.07447097450494766,
      "learning_rate": 1.4118044016091603e-05,
      "loss": 0.0779,
      "num_input_tokens_seen": 48779808,
      "step": 25735
    },
    {
      "epoch": 4.199037441879436,
      "grad_norm": 0.08979686349630356,
      "learning_rate": 1.410920622025594e-05,
      "loss": 0.0487,
      "num_input_tokens_seen": 48788176,
      "step": 25740
    },
    {
      "epoch": 4.199853169100253,
      "grad_norm": 0.14223256707191467,
      "learning_rate": 1.4100370103917554e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 48796976,
      "step": 25745
    },
    {
      "epoch": 4.20066889632107,
      "grad_norm": 0.5401244163513184,
      "learning_rate": 1.409153566843907e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 48806496,
      "step": 25750
    },
    {
      "epoch": 4.201484623541887,
      "grad_norm": 0.49269381165504456,
      "learning_rate": 1.408270291518286e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 48816464,
      "step": 25755
    },
    {
      "epoch": 4.202300350762705,
      "grad_norm": 0.07035249471664429,
      "learning_rate": 1.407387184551107e-05,
      "loss": 0.0517,
      "num_input_tokens_seen": 48826208,
      "step": 25760
    },
    {
      "epoch": 4.203116077983522,
      "grad_norm": 0.48995262384414673,
      "learning_rate": 1.4065042460785532e-05,
      "loss": 0.3135,
      "num_input_tokens_seen": 48834848,
      "step": 25765
    },
    {
      "epoch": 4.2039318052043395,
      "grad_norm": 0.22148214280605316,
      "learning_rate": 1.405621476236787e-05,
      "loss": 0.0934,
      "num_input_tokens_seen": 48843184,
      "step": 25770
    },
    {
      "epoch": 4.204747532425157,
      "grad_norm": 1.8118451833724976,
      "learning_rate": 1.4047388751619423e-05,
      "loss": 0.0562,
      "num_input_tokens_seen": 48853488,
      "step": 25775
    },
    {
      "epoch": 4.205563259645975,
      "grad_norm": 0.9988796710968018,
      "learning_rate": 1.4038564429901264e-05,
      "loss": 0.1768,
      "num_input_tokens_seen": 48863248,
      "step": 25780
    },
    {
      "epoch": 4.206378986866792,
      "grad_norm": 0.7533751130104065,
      "learning_rate": 1.4029741798574227e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 48872224,
      "step": 25785
    },
    {
      "epoch": 4.207194714087609,
      "grad_norm": 1.4707852602005005,
      "learning_rate": 1.402092085899886e-05,
      "loss": 0.0907,
      "num_input_tokens_seen": 48881312,
      "step": 25790
    },
    {
      "epoch": 4.208010441308426,
      "grad_norm": 0.7922506332397461,
      "learning_rate": 1.4012101612535464e-05,
      "loss": 0.0774,
      "num_input_tokens_seen": 48890608,
      "step": 25795
    },
    {
      "epoch": 4.208826168529244,
      "grad_norm": 0.21085822582244873,
      "learning_rate": 1.4003284060544092e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 48899856,
      "step": 25800
    },
    {
      "epoch": 4.208826168529244,
      "eval_loss": 0.1409369707107544,
      "eval_runtime": 133.2774,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 48899856,
      "step": 25800
    },
    {
      "epoch": 4.209641895750061,
      "grad_norm": 0.18672145903110504,
      "learning_rate": 1.3994468204384504e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 48909040,
      "step": 25805
    },
    {
      "epoch": 4.210457622970878,
      "grad_norm": 0.5488470196723938,
      "learning_rate": 1.398565404541622e-05,
      "loss": 0.0978,
      "num_input_tokens_seen": 48918816,
      "step": 25810
    },
    {
      "epoch": 4.211273350191696,
      "grad_norm": 1.5300683975219727,
      "learning_rate": 1.3976841584998513e-05,
      "loss": 0.1273,
      "num_input_tokens_seen": 48928816,
      "step": 25815
    },
    {
      "epoch": 4.2120890774125135,
      "grad_norm": 0.48963579535484314,
      "learning_rate": 1.3968030824490352e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 48937760,
      "step": 25820
    },
    {
      "epoch": 4.212904804633331,
      "grad_norm": 1.0482734441757202,
      "learning_rate": 1.3959221765250469e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 48948208,
      "step": 25825
    },
    {
      "epoch": 4.213720531854148,
      "grad_norm": 1.2274601459503174,
      "learning_rate": 1.3950414408637343e-05,
      "loss": 0.1024,
      "num_input_tokens_seen": 48956656,
      "step": 25830
    },
    {
      "epoch": 4.214536259074965,
      "grad_norm": 0.9338482022285461,
      "learning_rate": 1.3941608756009166e-05,
      "loss": 0.1528,
      "num_input_tokens_seen": 48965360,
      "step": 25835
    },
    {
      "epoch": 4.215351986295783,
      "grad_norm": 0.3489474654197693,
      "learning_rate": 1.3932804808723898e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 48975568,
      "step": 25840
    },
    {
      "epoch": 4.2161677135166,
      "grad_norm": 0.14495794475078583,
      "learning_rate": 1.3924002568139194e-05,
      "loss": 0.1053,
      "num_input_tokens_seen": 48984880,
      "step": 25845
    },
    {
      "epoch": 4.216983440737417,
      "grad_norm": 0.14745032787322998,
      "learning_rate": 1.3915202035612485e-05,
      "loss": 0.0387,
      "num_input_tokens_seen": 48993872,
      "step": 25850
    },
    {
      "epoch": 4.217799167958235,
      "grad_norm": 1.4763966798782349,
      "learning_rate": 1.3906403212500935e-05,
      "loss": 0.1008,
      "num_input_tokens_seen": 49003584,
      "step": 25855
    },
    {
      "epoch": 4.218614895179052,
      "grad_norm": 0.15160030126571655,
      "learning_rate": 1.3897606100161409e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 49013456,
      "step": 25860
    },
    {
      "epoch": 4.219430622399869,
      "grad_norm": 0.17578990757465363,
      "learning_rate": 1.388881069995055e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 49024288,
      "step": 25865
    },
    {
      "epoch": 4.220246349620687,
      "grad_norm": 2.5806870460510254,
      "learning_rate": 1.3880017013224708e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 49034080,
      "step": 25870
    },
    {
      "epoch": 4.221062076841505,
      "grad_norm": 0.26026690006256104,
      "learning_rate": 1.3871225041339984e-05,
      "loss": 0.12,
      "num_input_tokens_seen": 49043856,
      "step": 25875
    },
    {
      "epoch": 4.221877804062322,
      "grad_norm": 1.7439746856689453,
      "learning_rate": 1.386243478565222e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 49054272,
      "step": 25880
    },
    {
      "epoch": 4.222693531283139,
      "grad_norm": 0.4573206603527069,
      "learning_rate": 1.3853646247516966e-05,
      "loss": 0.0944,
      "num_input_tokens_seen": 49063696,
      "step": 25885
    },
    {
      "epoch": 4.223509258503956,
      "grad_norm": 0.11041847616434097,
      "learning_rate": 1.3844859428289545e-05,
      "loss": 0.1119,
      "num_input_tokens_seen": 49072624,
      "step": 25890
    },
    {
      "epoch": 4.224324985724774,
      "grad_norm": 0.6575822234153748,
      "learning_rate": 1.3836074329324984e-05,
      "loss": 0.0706,
      "num_input_tokens_seen": 49082144,
      "step": 25895
    },
    {
      "epoch": 4.225140712945591,
      "grad_norm": 0.6399835348129272,
      "learning_rate": 1.3827290951978044e-05,
      "loss": 0.221,
      "num_input_tokens_seen": 49092880,
      "step": 25900
    },
    {
      "epoch": 4.225956440166408,
      "grad_norm": 0.13405703008174896,
      "learning_rate": 1.381850929760326e-05,
      "loss": 0.079,
      "num_input_tokens_seen": 49102784,
      "step": 25905
    },
    {
      "epoch": 4.226772167387225,
      "grad_norm": 0.0399639718234539,
      "learning_rate": 1.3809729367554842e-05,
      "loss": 0.1147,
      "num_input_tokens_seen": 49111664,
      "step": 25910
    },
    {
      "epoch": 4.227587894608043,
      "grad_norm": 0.4962584674358368,
      "learning_rate": 1.3800951163186784e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 49121216,
      "step": 25915
    },
    {
      "epoch": 4.2284036218288605,
      "grad_norm": 0.04040297120809555,
      "learning_rate": 1.3792174685852801e-05,
      "loss": 0.05,
      "num_input_tokens_seen": 49130576,
      "step": 25920
    },
    {
      "epoch": 4.229219349049678,
      "grad_norm": 0.1928398460149765,
      "learning_rate": 1.378339993690632e-05,
      "loss": 0.1129,
      "num_input_tokens_seen": 49140528,
      "step": 25925
    },
    {
      "epoch": 4.230035076270495,
      "grad_norm": 0.35613930225372314,
      "learning_rate": 1.3774626917700523e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 49151008,
      "step": 25930
    },
    {
      "epoch": 4.230850803491313,
      "grad_norm": 0.04168090969324112,
      "learning_rate": 1.3765855629588334e-05,
      "loss": 0.1153,
      "num_input_tokens_seen": 49160992,
      "step": 25935
    },
    {
      "epoch": 4.23166653071213,
      "grad_norm": 0.13046365976333618,
      "learning_rate": 1.3757086073922374e-05,
      "loss": 0.1149,
      "num_input_tokens_seen": 49170192,
      "step": 25940
    },
    {
      "epoch": 4.232482257932947,
      "grad_norm": 1.4452029466629028,
      "learning_rate": 1.3748318252055038e-05,
      "loss": 0.063,
      "num_input_tokens_seen": 49178736,
      "step": 25945
    },
    {
      "epoch": 4.233297985153764,
      "grad_norm": 0.15416376292705536,
      "learning_rate": 1.3739552165338416e-05,
      "loss": 0.0328,
      "num_input_tokens_seen": 49188704,
      "step": 25950
    },
    {
      "epoch": 4.234113712374582,
      "grad_norm": 0.26718324422836304,
      "learning_rate": 1.3730787815124354e-05,
      "loss": 0.039,
      "num_input_tokens_seen": 49198064,
      "step": 25955
    },
    {
      "epoch": 4.234929439595399,
      "grad_norm": 0.6052286028862,
      "learning_rate": 1.3722025202764443e-05,
      "loss": 0.0269,
      "num_input_tokens_seen": 49207632,
      "step": 25960
    },
    {
      "epoch": 4.2357451668162165,
      "grad_norm": 0.06623474508523941,
      "learning_rate": 1.371326432960997e-05,
      "loss": 0.0139,
      "num_input_tokens_seen": 49216416,
      "step": 25965
    },
    {
      "epoch": 4.236560894037034,
      "grad_norm": 1.2958000898361206,
      "learning_rate": 1.3704505197011969e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 49224256,
      "step": 25970
    },
    {
      "epoch": 4.237376621257852,
      "grad_norm": 0.24700213968753815,
      "learning_rate": 1.3695747806321224e-05,
      "loss": 0.0863,
      "num_input_tokens_seen": 49234880,
      "step": 25975
    },
    {
      "epoch": 4.238192348478669,
      "grad_norm": 0.10428815335035324,
      "learning_rate": 1.3686992158888212e-05,
      "loss": 0.0816,
      "num_input_tokens_seen": 49244544,
      "step": 25980
    },
    {
      "epoch": 4.239008075699486,
      "grad_norm": 1.2811404466629028,
      "learning_rate": 1.367823825606319e-05,
      "loss": 0.2386,
      "num_input_tokens_seen": 49252992,
      "step": 25985
    },
    {
      "epoch": 4.239823802920303,
      "grad_norm": 0.6090632081031799,
      "learning_rate": 1.36694860991961e-05,
      "loss": 0.0626,
      "num_input_tokens_seen": 49261904,
      "step": 25990
    },
    {
      "epoch": 4.240639530141121,
      "grad_norm": 0.06687460094690323,
      "learning_rate": 1.3660735689636636e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 49270640,
      "step": 25995
    },
    {
      "epoch": 4.241455257361938,
      "grad_norm": 0.07416409254074097,
      "learning_rate": 1.365198702873424e-05,
      "loss": 0.0332,
      "num_input_tokens_seen": 49280208,
      "step": 26000
    },
    {
      "epoch": 4.241455257361938,
      "eval_loss": 0.143888920545578,
      "eval_runtime": 133.3684,
      "eval_samples_per_second": 20.432,
      "eval_steps_per_second": 10.22,
      "num_input_tokens_seen": 49280208,
      "step": 26000
    },
    {
      "epoch": 4.242270984582755,
      "grad_norm": 0.24248147010803223,
      "learning_rate": 1.364324011783804e-05,
      "loss": 0.0797,
      "num_input_tokens_seen": 49288880,
      "step": 26005
    },
    {
      "epoch": 4.243086711803572,
      "grad_norm": 0.44558000564575195,
      "learning_rate": 1.3634494958296934e-05,
      "loss": 0.2091,
      "num_input_tokens_seen": 49299408,
      "step": 26010
    },
    {
      "epoch": 4.2439024390243905,
      "grad_norm": 0.21183738112449646,
      "learning_rate": 1.3625751551459542e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 49309360,
      "step": 26015
    },
    {
      "epoch": 4.244718166245208,
      "grad_norm": 0.7379836440086365,
      "learning_rate": 1.3617009898674188e-05,
      "loss": 0.0742,
      "num_input_tokens_seen": 49319056,
      "step": 26020
    },
    {
      "epoch": 4.245533893466025,
      "grad_norm": 0.12910224497318268,
      "learning_rate": 1.3608270001288967e-05,
      "loss": 0.1323,
      "num_input_tokens_seen": 49328272,
      "step": 26025
    },
    {
      "epoch": 4.246349620686843,
      "grad_norm": 0.3515138626098633,
      "learning_rate": 1.359953186065166e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 49337264,
      "step": 26030
    },
    {
      "epoch": 4.24716534790766,
      "grad_norm": 0.027161750942468643,
      "learning_rate": 1.3590795478109814e-05,
      "loss": 0.2754,
      "num_input_tokens_seen": 49346160,
      "step": 26035
    },
    {
      "epoch": 4.247981075128477,
      "grad_norm": 0.9218365550041199,
      "learning_rate": 1.3582060855010675e-05,
      "loss": 0.0849,
      "num_input_tokens_seen": 49356112,
      "step": 26040
    },
    {
      "epoch": 4.248796802349294,
      "grad_norm": 0.47221896052360535,
      "learning_rate": 1.3573327992701245e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 49365408,
      "step": 26045
    },
    {
      "epoch": 4.249612529570112,
      "grad_norm": 1.1621969938278198,
      "learning_rate": 1.356459689252823e-05,
      "loss": 0.3191,
      "num_input_tokens_seen": 49376240,
      "step": 26050
    },
    {
      "epoch": 4.250428256790929,
      "grad_norm": 0.4782503545284271,
      "learning_rate": 1.3555867555838087e-05,
      "loss": 0.0569,
      "num_input_tokens_seen": 49386144,
      "step": 26055
    },
    {
      "epoch": 4.251243984011746,
      "grad_norm": 0.42016366124153137,
      "learning_rate": 1.3547139983976975e-05,
      "loss": 0.1047,
      "num_input_tokens_seen": 49395392,
      "step": 26060
    },
    {
      "epoch": 4.2520597112325635,
      "grad_norm": 1.192856788635254,
      "learning_rate": 1.3538414178290815e-05,
      "loss": 0.0544,
      "num_input_tokens_seen": 49405856,
      "step": 26065
    },
    {
      "epoch": 4.252875438453382,
      "grad_norm": 0.043341927230358124,
      "learning_rate": 1.3529690140125209e-05,
      "loss": 0.0911,
      "num_input_tokens_seen": 49415504,
      "step": 26070
    },
    {
      "epoch": 4.253691165674199,
      "grad_norm": 0.5032963156700134,
      "learning_rate": 1.352096787082553e-05,
      "loss": 0.1529,
      "num_input_tokens_seen": 49426032,
      "step": 26075
    },
    {
      "epoch": 4.254506892895016,
      "grad_norm": 1.4770944118499756,
      "learning_rate": 1.3512247371736871e-05,
      "loss": 0.0905,
      "num_input_tokens_seen": 49434432,
      "step": 26080
    },
    {
      "epoch": 4.255322620115833,
      "grad_norm": 0.20385532081127167,
      "learning_rate": 1.3503528644204022e-05,
      "loss": 0.0924,
      "num_input_tokens_seen": 49444688,
      "step": 26085
    },
    {
      "epoch": 4.256138347336651,
      "grad_norm": 0.562620222568512,
      "learning_rate": 1.349481168957153e-05,
      "loss": 0.115,
      "num_input_tokens_seen": 49453776,
      "step": 26090
    },
    {
      "epoch": 4.256954074557468,
      "grad_norm": 1.2046771049499512,
      "learning_rate": 1.3486096509183665e-05,
      "loss": 0.0576,
      "num_input_tokens_seen": 49461968,
      "step": 26095
    },
    {
      "epoch": 4.257769801778285,
      "grad_norm": 0.2655819058418274,
      "learning_rate": 1.3477383104384406e-05,
      "loss": 0.0773,
      "num_input_tokens_seen": 49472768,
      "step": 26100
    },
    {
      "epoch": 4.258585528999102,
      "grad_norm": 0.19098767638206482,
      "learning_rate": 1.3468671476517481e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 49482368,
      "step": 26105
    },
    {
      "epoch": 4.25940125621992,
      "grad_norm": 0.9172676801681519,
      "learning_rate": 1.3459961626926326e-05,
      "loss": 0.1048,
      "num_input_tokens_seen": 49491680,
      "step": 26110
    },
    {
      "epoch": 4.2602169834407375,
      "grad_norm": 0.11506094038486481,
      "learning_rate": 1.3451253556954101e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 49500880,
      "step": 26115
    },
    {
      "epoch": 4.261032710661555,
      "grad_norm": 0.7563544511795044,
      "learning_rate": 1.3442547267943717e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 49509472,
      "step": 26120
    },
    {
      "epoch": 4.261848437882372,
      "grad_norm": 1.5656278133392334,
      "learning_rate": 1.3433842761237774e-05,
      "loss": 0.2578,
      "num_input_tokens_seen": 49517824,
      "step": 26125
    },
    {
      "epoch": 4.26266416510319,
      "grad_norm": 0.808265745639801,
      "learning_rate": 1.3425140038178639e-05,
      "loss": 0.077,
      "num_input_tokens_seen": 49526448,
      "step": 26130
    },
    {
      "epoch": 4.263479892324007,
      "grad_norm": 1.7043527364730835,
      "learning_rate": 1.3416439100108358e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 49535664,
      "step": 26135
    },
    {
      "epoch": 4.264295619544824,
      "grad_norm": 0.44771990180015564,
      "learning_rate": 1.3407739948368734e-05,
      "loss": 0.3676,
      "num_input_tokens_seen": 49544848,
      "step": 26140
    },
    {
      "epoch": 4.265111346765641,
      "grad_norm": 1.4915354251861572,
      "learning_rate": 1.3399042584301298e-05,
      "loss": 0.2007,
      "num_input_tokens_seen": 49552592,
      "step": 26145
    },
    {
      "epoch": 4.265927073986459,
      "grad_norm": 0.10040515661239624,
      "learning_rate": 1.3390347009247272e-05,
      "loss": 0.1082,
      "num_input_tokens_seen": 49562560,
      "step": 26150
    },
    {
      "epoch": 4.266742801207276,
      "grad_norm": 0.1942933350801468,
      "learning_rate": 1.3381653224547635e-05,
      "loss": 0.0954,
      "num_input_tokens_seen": 49572512,
      "step": 26155
    },
    {
      "epoch": 4.2675585284280935,
      "grad_norm": 1.7083351612091064,
      "learning_rate": 1.3372961231543086e-05,
      "loss": 0.3336,
      "num_input_tokens_seen": 49580656,
      "step": 26160
    },
    {
      "epoch": 4.268374255648911,
      "grad_norm": 1.5469396114349365,
      "learning_rate": 1.3364271031574016e-05,
      "loss": 0.3161,
      "num_input_tokens_seen": 49591392,
      "step": 26165
    },
    {
      "epoch": 4.269189982869729,
      "grad_norm": 0.24219682812690735,
      "learning_rate": 1.335558262598059e-05,
      "loss": 0.1237,
      "num_input_tokens_seen": 49602048,
      "step": 26170
    },
    {
      "epoch": 4.270005710090546,
      "grad_norm": 0.662091851234436,
      "learning_rate": 1.3346896016102645e-05,
      "loss": 0.083,
      "num_input_tokens_seen": 49610944,
      "step": 26175
    },
    {
      "epoch": 4.270821437311363,
      "grad_norm": 0.951148509979248,
      "learning_rate": 1.3338211203279788e-05,
      "loss": 0.108,
      "num_input_tokens_seen": 49619728,
      "step": 26180
    },
    {
      "epoch": 4.27163716453218,
      "grad_norm": 0.03986675292253494,
      "learning_rate": 1.3329528188851303e-05,
      "loss": 0.2393,
      "num_input_tokens_seen": 49629680,
      "step": 26185
    },
    {
      "epoch": 4.272452891752998,
      "grad_norm": 1.8938723802566528,
      "learning_rate": 1.3320846974156242e-05,
      "loss": 0.0923,
      "num_input_tokens_seen": 49639488,
      "step": 26190
    },
    {
      "epoch": 4.273268618973815,
      "grad_norm": 0.993911623954773,
      "learning_rate": 1.3312167560533337e-05,
      "loss": 0.1681,
      "num_input_tokens_seen": 49648992,
      "step": 26195
    },
    {
      "epoch": 4.274084346194632,
      "grad_norm": 0.4113839864730835,
      "learning_rate": 1.3303489949321082e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 49658080,
      "step": 26200
    },
    {
      "epoch": 4.274084346194632,
      "eval_loss": 0.144459530711174,
      "eval_runtime": 133.4234,
      "eval_samples_per_second": 20.424,
      "eval_steps_per_second": 10.216,
      "num_input_tokens_seen": 49658080,
      "step": 26200
    },
    {
      "epoch": 4.27490007341545,
      "grad_norm": 0.6689632534980774,
      "learning_rate": 1.3294814141857653e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 49667248,
      "step": 26205
    },
    {
      "epoch": 4.275715800636267,
      "grad_norm": 0.17941929399967194,
      "learning_rate": 1.3286140139480992e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 49675952,
      "step": 26210
    },
    {
      "epoch": 4.276531527857085,
      "grad_norm": 1.287448763847351,
      "learning_rate": 1.3277467943528719e-05,
      "loss": 0.1529,
      "num_input_tokens_seen": 49685664,
      "step": 26215
    },
    {
      "epoch": 4.277347255077902,
      "grad_norm": 1.6487261056900024,
      "learning_rate": 1.3268797555338203e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 49695008,
      "step": 26220
    },
    {
      "epoch": 4.27816298229872,
      "grad_norm": 0.056204769760370255,
      "learning_rate": 1.3260128976246533e-05,
      "loss": 0.0754,
      "num_input_tokens_seen": 49704448,
      "step": 26225
    },
    {
      "epoch": 4.278978709519537,
      "grad_norm": 0.27361899614334106,
      "learning_rate": 1.32514622075905e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 49713248,
      "step": 26230
    },
    {
      "epoch": 4.279794436740354,
      "grad_norm": 1.3643349409103394,
      "learning_rate": 1.3242797250706638e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 49722752,
      "step": 26235
    },
    {
      "epoch": 4.280610163961171,
      "grad_norm": 0.161720409989357,
      "learning_rate": 1.3234134106931195e-05,
      "loss": 0.0279,
      "num_input_tokens_seen": 49732736,
      "step": 26240
    },
    {
      "epoch": 4.281425891181989,
      "grad_norm": 1.0392273664474487,
      "learning_rate": 1.322547277760013e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 49742768,
      "step": 26245
    },
    {
      "epoch": 4.282241618402806,
      "grad_norm": 0.09429457783699036,
      "learning_rate": 1.3216813264049132e-05,
      "loss": 0.0393,
      "num_input_tokens_seen": 49752192,
      "step": 26250
    },
    {
      "epoch": 4.283057345623623,
      "grad_norm": 0.140595942735672,
      "learning_rate": 1.32081555676136e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 49761728,
      "step": 26255
    },
    {
      "epoch": 4.2838730728444405,
      "grad_norm": 1.5614999532699585,
      "learning_rate": 1.3199499689628674e-05,
      "loss": 0.1761,
      "num_input_tokens_seen": 49771568,
      "step": 26260
    },
    {
      "epoch": 4.2846888000652585,
      "grad_norm": 0.32987669110298157,
      "learning_rate": 1.3190845631429192e-05,
      "loss": 0.1234,
      "num_input_tokens_seen": 49780976,
      "step": 26265
    },
    {
      "epoch": 4.285504527286076,
      "grad_norm": 0.14903628826141357,
      "learning_rate": 1.3182193394349704e-05,
      "loss": 0.203,
      "num_input_tokens_seen": 49791120,
      "step": 26270
    },
    {
      "epoch": 4.286320254506893,
      "grad_norm": 0.17745114862918854,
      "learning_rate": 1.3173542979724507e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 49799952,
      "step": 26275
    },
    {
      "epoch": 4.28713598172771,
      "grad_norm": 0.4059651494026184,
      "learning_rate": 1.3164894388887617e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 49809008,
      "step": 26280
    },
    {
      "epoch": 4.287951708948528,
      "grad_norm": 0.39901629090309143,
      "learning_rate": 1.3156247623172727e-05,
      "loss": 0.1729,
      "num_input_tokens_seen": 49818992,
      "step": 26285
    },
    {
      "epoch": 4.288767436169345,
      "grad_norm": 0.4663274586200714,
      "learning_rate": 1.3147602683913302e-05,
      "loss": 0.1745,
      "num_input_tokens_seen": 49828976,
      "step": 26290
    },
    {
      "epoch": 4.289583163390162,
      "grad_norm": 1.4826586246490479,
      "learning_rate": 1.3138959572442481e-05,
      "loss": 0.1842,
      "num_input_tokens_seen": 49837712,
      "step": 26295
    },
    {
      "epoch": 4.290398890610979,
      "grad_norm": 0.8556394577026367,
      "learning_rate": 1.3130318290093146e-05,
      "loss": 0.0855,
      "num_input_tokens_seen": 49846128,
      "step": 26300
    },
    {
      "epoch": 4.291214617831797,
      "grad_norm": 0.9907035231590271,
      "learning_rate": 1.3121678838197909e-05,
      "loss": 0.1224,
      "num_input_tokens_seen": 49856192,
      "step": 26305
    },
    {
      "epoch": 4.2920303450526145,
      "grad_norm": 0.06443056464195251,
      "learning_rate": 1.3113041218089056e-05,
      "loss": 0.1333,
      "num_input_tokens_seen": 49865936,
      "step": 26310
    },
    {
      "epoch": 4.292846072273432,
      "grad_norm": 0.2863946557044983,
      "learning_rate": 1.3104405431098626e-05,
      "loss": 0.0735,
      "num_input_tokens_seen": 49875488,
      "step": 26315
    },
    {
      "epoch": 4.293661799494249,
      "grad_norm": 0.17515483498573303,
      "learning_rate": 1.3095771478558377e-05,
      "loss": 0.1791,
      "num_input_tokens_seen": 49884800,
      "step": 26320
    },
    {
      "epoch": 4.294477526715067,
      "grad_norm": 0.4113374352455139,
      "learning_rate": 1.3087139361799766e-05,
      "loss": 0.2328,
      "num_input_tokens_seen": 49894176,
      "step": 26325
    },
    {
      "epoch": 4.295293253935884,
      "grad_norm": 0.11036481708288193,
      "learning_rate": 1.3078509082153964e-05,
      "loss": 0.0293,
      "num_input_tokens_seen": 49903488,
      "step": 26330
    },
    {
      "epoch": 4.296108981156701,
      "grad_norm": 0.03386333957314491,
      "learning_rate": 1.3069880640951885e-05,
      "loss": 0.1043,
      "num_input_tokens_seen": 49912096,
      "step": 26335
    },
    {
      "epoch": 4.296924708377518,
      "grad_norm": 0.20545770227909088,
      "learning_rate": 1.3061254039524123e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 49921408,
      "step": 26340
    },
    {
      "epoch": 4.297740435598336,
      "grad_norm": 0.08883549273014069,
      "learning_rate": 1.3052629279201028e-05,
      "loss": 0.0988,
      "num_input_tokens_seen": 49931168,
      "step": 26345
    },
    {
      "epoch": 4.298556162819153,
      "grad_norm": 0.511010468006134,
      "learning_rate": 1.3044006361312633e-05,
      "loss": 0.0904,
      "num_input_tokens_seen": 49941024,
      "step": 26350
    },
    {
      "epoch": 4.2993718900399704,
      "grad_norm": 0.10744424164295197,
      "learning_rate": 1.30353852871887e-05,
      "loss": 0.0776,
      "num_input_tokens_seen": 49949232,
      "step": 26355
    },
    {
      "epoch": 4.300187617260788,
      "grad_norm": 0.4676356315612793,
      "learning_rate": 1.302676605815873e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 49959072,
      "step": 26360
    },
    {
      "epoch": 4.301003344481606,
      "grad_norm": 0.22593607008457184,
      "learning_rate": 1.3018148675551884e-05,
      "loss": 0.2799,
      "num_input_tokens_seen": 49968496,
      "step": 26365
    },
    {
      "epoch": 4.301819071702423,
      "grad_norm": 1.0693137645721436,
      "learning_rate": 1.3009533140697094e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 49977792,
      "step": 26370
    },
    {
      "epoch": 4.30263479892324,
      "grad_norm": 0.9336548447608948,
      "learning_rate": 1.3000919454922966e-05,
      "loss": 0.1175,
      "num_input_tokens_seen": 49987152,
      "step": 26375
    },
    {
      "epoch": 4.303450526144058,
      "grad_norm": 0.7498557567596436,
      "learning_rate": 1.299230761955785e-05,
      "loss": 0.2439,
      "num_input_tokens_seen": 49997920,
      "step": 26380
    },
    {
      "epoch": 4.304266253364875,
      "grad_norm": 1.49233877658844,
      "learning_rate": 1.2983697635929807e-05,
      "loss": 0.1209,
      "num_input_tokens_seen": 50007280,
      "step": 26385
    },
    {
      "epoch": 4.305081980585692,
      "grad_norm": 1.9304537773132324,
      "learning_rate": 1.2975089505366584e-05,
      "loss": 0.2547,
      "num_input_tokens_seen": 50015600,
      "step": 26390
    },
    {
      "epoch": 4.305897707806509,
      "grad_norm": 1.8568865060806274,
      "learning_rate": 1.2966483229195683e-05,
      "loss": 0.1199,
      "num_input_tokens_seen": 50025120,
      "step": 26395
    },
    {
      "epoch": 4.306713435027326,
      "grad_norm": 0.47539517283439636,
      "learning_rate": 1.2957878808744283e-05,
      "loss": 0.0985,
      "num_input_tokens_seen": 50034848,
      "step": 26400
    },
    {
      "epoch": 4.306713435027326,
      "eval_loss": 0.14291416108608246,
      "eval_runtime": 133.1849,
      "eval_samples_per_second": 20.46,
      "eval_steps_per_second": 10.234,
      "num_input_tokens_seen": 50034848,
      "step": 26400
    },
    {
      "epoch": 4.307529162248144,
      "grad_norm": 0.6218879222869873,
      "learning_rate": 1.294927624533931e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 50044288,
      "step": 26405
    },
    {
      "epoch": 4.3083448894689615,
      "grad_norm": 0.40001380443573,
      "learning_rate": 1.2940675540307378e-05,
      "loss": 0.1051,
      "num_input_tokens_seen": 50054176,
      "step": 26410
    },
    {
      "epoch": 4.309160616689779,
      "grad_norm": 0.22472678124904633,
      "learning_rate": 1.2932076694974814e-05,
      "loss": 0.1761,
      "num_input_tokens_seen": 50063136,
      "step": 26415
    },
    {
      "epoch": 4.309976343910597,
      "grad_norm": 0.8644217848777771,
      "learning_rate": 1.2923479710667682e-05,
      "loss": 0.1019,
      "num_input_tokens_seen": 50072096,
      "step": 26420
    },
    {
      "epoch": 4.310792071131414,
      "grad_norm": 0.6823829412460327,
      "learning_rate": 1.2914884588711751e-05,
      "loss": 0.1131,
      "num_input_tokens_seen": 50081120,
      "step": 26425
    },
    {
      "epoch": 4.311607798352231,
      "grad_norm": 0.2957998216152191,
      "learning_rate": 1.2906291330432475e-05,
      "loss": 0.0335,
      "num_input_tokens_seen": 50090304,
      "step": 26430
    },
    {
      "epoch": 4.312423525573048,
      "grad_norm": 0.06697936356067657,
      "learning_rate": 1.2897699937155055e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 50100400,
      "step": 26435
    },
    {
      "epoch": 4.313239252793866,
      "grad_norm": 0.42882341146469116,
      "learning_rate": 1.2889110410204403e-05,
      "loss": 0.0843,
      "num_input_tokens_seen": 50109632,
      "step": 26440
    },
    {
      "epoch": 4.314054980014683,
      "grad_norm": 0.8721769452095032,
      "learning_rate": 1.2880522750905111e-05,
      "loss": 0.0882,
      "num_input_tokens_seen": 50119552,
      "step": 26445
    },
    {
      "epoch": 4.3148707072355,
      "grad_norm": 0.25740861892700195,
      "learning_rate": 1.2871936960581523e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 50129008,
      "step": 26450
    },
    {
      "epoch": 4.3156864344563175,
      "grad_norm": 0.8594481945037842,
      "learning_rate": 1.2863353040557658e-05,
      "loss": 0.0739,
      "num_input_tokens_seen": 50137904,
      "step": 26455
    },
    {
      "epoch": 4.3165021616771355,
      "grad_norm": 1.3238894939422607,
      "learning_rate": 1.2854770992157273e-05,
      "loss": 0.2468,
      "num_input_tokens_seen": 50147632,
      "step": 26460
    },
    {
      "epoch": 4.317317888897953,
      "grad_norm": 0.1835549771785736,
      "learning_rate": 1.2846190816703835e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 50157104,
      "step": 26465
    },
    {
      "epoch": 4.31813361611877,
      "grad_norm": 0.22167974710464478,
      "learning_rate": 1.2837612515520498e-05,
      "loss": 0.065,
      "num_input_tokens_seen": 50166640,
      "step": 26470
    },
    {
      "epoch": 4.318949343339587,
      "grad_norm": 0.051136747002601624,
      "learning_rate": 1.2829036089930163e-05,
      "loss": 0.068,
      "num_input_tokens_seen": 50176848,
      "step": 26475
    },
    {
      "epoch": 4.319765070560405,
      "grad_norm": 0.10044422745704651,
      "learning_rate": 1.2820461541255412e-05,
      "loss": 0.0408,
      "num_input_tokens_seen": 50186064,
      "step": 26480
    },
    {
      "epoch": 4.320580797781222,
      "grad_norm": 0.38463953137397766,
      "learning_rate": 1.2811888870818543e-05,
      "loss": 0.1313,
      "num_input_tokens_seen": 50195088,
      "step": 26485
    },
    {
      "epoch": 4.321396525002039,
      "grad_norm": 0.7232702374458313,
      "learning_rate": 1.2803318079941581e-05,
      "loss": 0.1153,
      "num_input_tokens_seen": 50204640,
      "step": 26490
    },
    {
      "epoch": 4.322212252222856,
      "grad_norm": 0.7689719200134277,
      "learning_rate": 1.2794749169946235e-05,
      "loss": 0.2398,
      "num_input_tokens_seen": 50214512,
      "step": 26495
    },
    {
      "epoch": 4.323027979443674,
      "grad_norm": 0.7663591504096985,
      "learning_rate": 1.2786182142153952e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 50225136,
      "step": 26500
    },
    {
      "epoch": 4.3238437066644915,
      "grad_norm": 1.3400646448135376,
      "learning_rate": 1.2777616997885878e-05,
      "loss": 0.0847,
      "num_input_tokens_seen": 50234880,
      "step": 26505
    },
    {
      "epoch": 4.324659433885309,
      "grad_norm": 1.343153715133667,
      "learning_rate": 1.2769053738462847e-05,
      "loss": 0.1113,
      "num_input_tokens_seen": 50244208,
      "step": 26510
    },
    {
      "epoch": 4.325475161106126,
      "grad_norm": 0.12348069250583649,
      "learning_rate": 1.2760492365205434e-05,
      "loss": 0.0548,
      "num_input_tokens_seen": 50254880,
      "step": 26515
    },
    {
      "epoch": 4.326290888326944,
      "grad_norm": 0.10591522604227066,
      "learning_rate": 1.2751932879433919e-05,
      "loss": 0.2469,
      "num_input_tokens_seen": 50263568,
      "step": 26520
    },
    {
      "epoch": 4.327106615547761,
      "grad_norm": 1.4052270650863647,
      "learning_rate": 1.2743375282468267e-05,
      "loss": 0.1789,
      "num_input_tokens_seen": 50272736,
      "step": 26525
    },
    {
      "epoch": 4.327922342768578,
      "grad_norm": 1.4058942794799805,
      "learning_rate": 1.2734819575628182e-05,
      "loss": 0.0753,
      "num_input_tokens_seen": 50281232,
      "step": 26530
    },
    {
      "epoch": 4.328738069989395,
      "grad_norm": 0.6250662803649902,
      "learning_rate": 1.2726265760233039e-05,
      "loss": 0.203,
      "num_input_tokens_seen": 50290288,
      "step": 26535
    },
    {
      "epoch": 4.329553797210213,
      "grad_norm": 0.9020418524742126,
      "learning_rate": 1.271771383760197e-05,
      "loss": 0.1135,
      "num_input_tokens_seen": 50300048,
      "step": 26540
    },
    {
      "epoch": 4.33036952443103,
      "grad_norm": 0.06868349760770798,
      "learning_rate": 1.2709163809053764e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 50310624,
      "step": 26545
    },
    {
      "epoch": 4.331185251651847,
      "grad_norm": 0.7600064277648926,
      "learning_rate": 1.2700615675906963e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 50319888,
      "step": 26550
    },
    {
      "epoch": 4.332000978872665,
      "grad_norm": 1.553351640701294,
      "learning_rate": 1.269206943947978e-05,
      "loss": 0.0847,
      "num_input_tokens_seen": 50329376,
      "step": 26555
    },
    {
      "epoch": 4.332816706093483,
      "grad_norm": 0.7943693399429321,
      "learning_rate": 1.2683525101090177e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 50338064,
      "step": 26560
    },
    {
      "epoch": 4.3336324333143,
      "grad_norm": 1.6170930862426758,
      "learning_rate": 1.2674982662055765e-05,
      "loss": 0.1942,
      "num_input_tokens_seen": 50348128,
      "step": 26565
    },
    {
      "epoch": 4.334448160535117,
      "grad_norm": 0.7210689187049866,
      "learning_rate": 1.2666442123693922e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 50357312,
      "step": 26570
    },
    {
      "epoch": 4.335263887755934,
      "grad_norm": 0.18305248022079468,
      "learning_rate": 1.265790348732169e-05,
      "loss": 0.1175,
      "num_input_tokens_seen": 50366592,
      "step": 26575
    },
    {
      "epoch": 4.336079614976752,
      "grad_norm": 0.0974150002002716,
      "learning_rate": 1.264936675425584e-05,
      "loss": 0.1124,
      "num_input_tokens_seen": 50376112,
      "step": 26580
    },
    {
      "epoch": 4.336895342197569,
      "grad_norm": 1.8918788433074951,
      "learning_rate": 1.2640831925812852e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 50386192,
      "step": 26585
    },
    {
      "epoch": 4.337711069418386,
      "grad_norm": 0.07206472754478455,
      "learning_rate": 1.263229900330889e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 50394864,
      "step": 26590
    },
    {
      "epoch": 4.338526796639204,
      "grad_norm": 0.6959061622619629,
      "learning_rate": 1.2623767988059843e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 50404960,
      "step": 26595
    },
    {
      "epoch": 4.339342523860021,
      "grad_norm": 1.7508801221847534,
      "learning_rate": 1.2615238881381309e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 50413376,
      "step": 26600
    },
    {
      "epoch": 4.339342523860021,
      "eval_loss": 0.14178642630577087,
      "eval_runtime": 133.2931,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 50413376,
      "step": 26600
    },
    {
      "epoch": 4.3401582510808385,
      "grad_norm": 0.09956388175487518,
      "learning_rate": 1.2606711684588568e-05,
      "loss": 0.068,
      "num_input_tokens_seen": 50423824,
      "step": 26605
    },
    {
      "epoch": 4.340973978301656,
      "grad_norm": 0.6368354558944702,
      "learning_rate": 1.2598186398996636e-05,
      "loss": 0.1339,
      "num_input_tokens_seen": 50433056,
      "step": 26610
    },
    {
      "epoch": 4.341789705522474,
      "grad_norm": 0.6288803815841675,
      "learning_rate": 1.2589663025920207e-05,
      "loss": 0.0937,
      "num_input_tokens_seen": 50442176,
      "step": 26615
    },
    {
      "epoch": 4.342605432743291,
      "grad_norm": 0.5312763452529907,
      "learning_rate": 1.2581141566673705e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 50452032,
      "step": 26620
    },
    {
      "epoch": 4.343421159964108,
      "grad_norm": 1.1066079139709473,
      "learning_rate": 1.257262202257124e-05,
      "loss": 0.1009,
      "num_input_tokens_seen": 50462768,
      "step": 26625
    },
    {
      "epoch": 4.344236887184925,
      "grad_norm": 0.7667326331138611,
      "learning_rate": 1.2564104394926618e-05,
      "loss": 0.2149,
      "num_input_tokens_seen": 50473216,
      "step": 26630
    },
    {
      "epoch": 4.345052614405743,
      "grad_norm": 0.10744719207286835,
      "learning_rate": 1.2555588685053383e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 50483456,
      "step": 26635
    },
    {
      "epoch": 4.34586834162656,
      "grad_norm": 0.2915803790092468,
      "learning_rate": 1.2547074894264762e-05,
      "loss": 0.187,
      "num_input_tokens_seen": 50492992,
      "step": 26640
    },
    {
      "epoch": 4.346684068847377,
      "grad_norm": 0.23110641539096832,
      "learning_rate": 1.2538563023873679e-05,
      "loss": 0.0877,
      "num_input_tokens_seen": 50502208,
      "step": 26645
    },
    {
      "epoch": 4.3474997960681945,
      "grad_norm": 1.3924510478973389,
      "learning_rate": 1.2530053075192789e-05,
      "loss": 0.1207,
      "num_input_tokens_seen": 50511616,
      "step": 26650
    },
    {
      "epoch": 4.3483155232890125,
      "grad_norm": 0.2802341878414154,
      "learning_rate": 1.252154504953441e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 50521952,
      "step": 26655
    },
    {
      "epoch": 4.34913125050983,
      "grad_norm": 0.2023971676826477,
      "learning_rate": 1.25130389482106e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 50531296,
      "step": 26660
    },
    {
      "epoch": 4.349946977730647,
      "grad_norm": 0.2738489508628845,
      "learning_rate": 1.2504534772533116e-05,
      "loss": 0.2924,
      "num_input_tokens_seen": 50540688,
      "step": 26665
    },
    {
      "epoch": 4.350762704951464,
      "grad_norm": 0.1733650416135788,
      "learning_rate": 1.2496032523813387e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 50550320,
      "step": 26670
    },
    {
      "epoch": 4.351578432172282,
      "grad_norm": 0.3825167417526245,
      "learning_rate": 1.2487532203362576e-05,
      "loss": 0.0563,
      "num_input_tokens_seen": 50559424,
      "step": 26675
    },
    {
      "epoch": 4.352394159393099,
      "grad_norm": 2.187624454498291,
      "learning_rate": 1.247903381249155e-05,
      "loss": 0.273,
      "num_input_tokens_seen": 50569536,
      "step": 26680
    },
    {
      "epoch": 4.353209886613916,
      "grad_norm": 1.140194058418274,
      "learning_rate": 1.2470537352510853e-05,
      "loss": 0.2831,
      "num_input_tokens_seen": 50579648,
      "step": 26685
    },
    {
      "epoch": 4.354025613834733,
      "grad_norm": 0.5594476461410522,
      "learning_rate": 1.2462042824730758e-05,
      "loss": 0.0726,
      "num_input_tokens_seen": 50589328,
      "step": 26690
    },
    {
      "epoch": 4.354841341055551,
      "grad_norm": 1.4460549354553223,
      "learning_rate": 1.245355023046122e-05,
      "loss": 0.2852,
      "num_input_tokens_seen": 50598208,
      "step": 26695
    },
    {
      "epoch": 4.3556570682763684,
      "grad_norm": 0.8971638679504395,
      "learning_rate": 1.2445059571011896e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 50607920,
      "step": 26700
    },
    {
      "epoch": 4.356472795497186,
      "grad_norm": 0.33125805854797363,
      "learning_rate": 1.2436570847692173e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 50617712,
      "step": 26705
    },
    {
      "epoch": 4.357288522718003,
      "grad_norm": 0.21719679236412048,
      "learning_rate": 1.2428084061811096e-05,
      "loss": 0.2523,
      "num_input_tokens_seen": 50626640,
      "step": 26710
    },
    {
      "epoch": 4.358104249938821,
      "grad_norm": 0.31875890493392944,
      "learning_rate": 1.2419599214677447e-05,
      "loss": 0.2542,
      "num_input_tokens_seen": 50635984,
      "step": 26715
    },
    {
      "epoch": 4.358919977159638,
      "grad_norm": 1.3757350444793701,
      "learning_rate": 1.2411116307599702e-05,
      "loss": 0.1515,
      "num_input_tokens_seen": 50645504,
      "step": 26720
    },
    {
      "epoch": 4.359735704380455,
      "grad_norm": 1.0839565992355347,
      "learning_rate": 1.2402635341886016e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 50653824,
      "step": 26725
    },
    {
      "epoch": 4.360551431601272,
      "grad_norm": 2.5089824199676514,
      "learning_rate": 1.2394156318844278e-05,
      "loss": 0.1931,
      "num_input_tokens_seen": 50663472,
      "step": 26730
    },
    {
      "epoch": 4.36136715882209,
      "grad_norm": 0.054119788110256195,
      "learning_rate": 1.2385679239782039e-05,
      "loss": 0.0452,
      "num_input_tokens_seen": 50672272,
      "step": 26735
    },
    {
      "epoch": 4.362182886042907,
      "grad_norm": 1.0271215438842773,
      "learning_rate": 1.2377204106006585e-05,
      "loss": 0.0897,
      "num_input_tokens_seen": 50682400,
      "step": 26740
    },
    {
      "epoch": 4.362998613263724,
      "grad_norm": 0.26071280241012573,
      "learning_rate": 1.2368730918824891e-05,
      "loss": 0.2039,
      "num_input_tokens_seen": 50692240,
      "step": 26745
    },
    {
      "epoch": 4.3638143404845415,
      "grad_norm": 0.14319907128810883,
      "learning_rate": 1.236025967954362e-05,
      "loss": 0.1122,
      "num_input_tokens_seen": 50700608,
      "step": 26750
    },
    {
      "epoch": 4.3646300677053596,
      "grad_norm": 0.09581250697374344,
      "learning_rate": 1.2351790389469153e-05,
      "loss": 0.033,
      "num_input_tokens_seen": 50710528,
      "step": 26755
    },
    {
      "epoch": 4.365445794926177,
      "grad_norm": 0.45942631363868713,
      "learning_rate": 1.234332304990755e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 50719136,
      "step": 26760
    },
    {
      "epoch": 4.366261522146994,
      "grad_norm": 0.05872032791376114,
      "learning_rate": 1.2334857662164593e-05,
      "loss": 0.0812,
      "num_input_tokens_seen": 50728416,
      "step": 26765
    },
    {
      "epoch": 4.367077249367812,
      "grad_norm": 0.2675348222255707,
      "learning_rate": 1.2326394227545743e-05,
      "loss": 0.0984,
      "num_input_tokens_seen": 50738416,
      "step": 26770
    },
    {
      "epoch": 4.367892976588629,
      "grad_norm": 0.2254408597946167,
      "learning_rate": 1.2317932747356162e-05,
      "loss": 0.1879,
      "num_input_tokens_seen": 50747392,
      "step": 26775
    },
    {
      "epoch": 4.368708703809446,
      "grad_norm": 0.5779612064361572,
      "learning_rate": 1.2309473222900726e-05,
      "loss": 0.0668,
      "num_input_tokens_seen": 50758096,
      "step": 26780
    },
    {
      "epoch": 4.369524431030263,
      "grad_norm": 1.2383235692977905,
      "learning_rate": 1.2301015655484006e-05,
      "loss": 0.1958,
      "num_input_tokens_seen": 50767632,
      "step": 26785
    },
    {
      "epoch": 4.370340158251081,
      "grad_norm": 0.3996729552745819,
      "learning_rate": 1.2292560046410245e-05,
      "loss": 0.1761,
      "num_input_tokens_seen": 50776096,
      "step": 26790
    },
    {
      "epoch": 4.371155885471898,
      "grad_norm": 1.2995659112930298,
      "learning_rate": 1.228410639698343e-05,
      "loss": 0.3027,
      "num_input_tokens_seen": 50784816,
      "step": 26795
    },
    {
      "epoch": 4.3719716126927155,
      "grad_norm": 0.7812651991844177,
      "learning_rate": 1.2275654708507195e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 50793248,
      "step": 26800
    },
    {
      "epoch": 4.3719716126927155,
      "eval_loss": 0.1420937329530716,
      "eval_runtime": 133.2472,
      "eval_samples_per_second": 20.451,
      "eval_steps_per_second": 10.229,
      "num_input_tokens_seen": 50793248,
      "step": 26800
    },
    {
      "epoch": 4.372787339913533,
      "grad_norm": 0.5827279686927795,
      "learning_rate": 1.2267204982284908e-05,
      "loss": 0.0912,
      "num_input_tokens_seen": 50802480,
      "step": 26805
    },
    {
      "epoch": 4.373603067134351,
      "grad_norm": 1.012304425239563,
      "learning_rate": 1.2258757219619635e-05,
      "loss": 0.1937,
      "num_input_tokens_seen": 50812704,
      "step": 26810
    },
    {
      "epoch": 4.374418794355168,
      "grad_norm": 1.1653684377670288,
      "learning_rate": 1.2250311421814104e-05,
      "loss": 0.2513,
      "num_input_tokens_seen": 50821488,
      "step": 26815
    },
    {
      "epoch": 4.375234521575985,
      "grad_norm": 0.038178399205207825,
      "learning_rate": 1.2241867590170772e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 50831712,
      "step": 26820
    },
    {
      "epoch": 4.376050248796802,
      "grad_norm": 0.3178709149360657,
      "learning_rate": 1.2233425725991799e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 50840480,
      "step": 26825
    },
    {
      "epoch": 4.37686597601762,
      "grad_norm": 0.054768458008766174,
      "learning_rate": 1.2224985830579003e-05,
      "loss": 0.0772,
      "num_input_tokens_seen": 50849872,
      "step": 26830
    },
    {
      "epoch": 4.377681703238437,
      "grad_norm": 0.3866581618785858,
      "learning_rate": 1.2216547905233944e-05,
      "loss": 0.0792,
      "num_input_tokens_seen": 50860048,
      "step": 26835
    },
    {
      "epoch": 4.378497430459254,
      "grad_norm": 0.895570695400238,
      "learning_rate": 1.2208111951257842e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 50869360,
      "step": 26840
    },
    {
      "epoch": 4.3793131576800715,
      "grad_norm": 0.7489267587661743,
      "learning_rate": 1.2199677969951622e-05,
      "loss": 0.2379,
      "num_input_tokens_seen": 50879536,
      "step": 26845
    },
    {
      "epoch": 4.3801288849008895,
      "grad_norm": 0.5540897250175476,
      "learning_rate": 1.2191245962615927e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 50889424,
      "step": 26850
    },
    {
      "epoch": 4.380944612121707,
      "grad_norm": 1.065796971321106,
      "learning_rate": 1.218281593055106e-05,
      "loss": 0.122,
      "num_input_tokens_seen": 50898432,
      "step": 26855
    },
    {
      "epoch": 4.381760339342524,
      "grad_norm": 0.5088992714881897,
      "learning_rate": 1.217438787505705e-05,
      "loss": 0.1974,
      "num_input_tokens_seen": 50907664,
      "step": 26860
    },
    {
      "epoch": 4.382576066563341,
      "grad_norm": 0.9290018081665039,
      "learning_rate": 1.2165961797433615e-05,
      "loss": 0.0792,
      "num_input_tokens_seen": 50917664,
      "step": 26865
    },
    {
      "epoch": 4.383391793784159,
      "grad_norm": 0.34003207087516785,
      "learning_rate": 1.215753769898014e-05,
      "loss": 0.0369,
      "num_input_tokens_seen": 50927088,
      "step": 26870
    },
    {
      "epoch": 4.384207521004976,
      "grad_norm": 1.0289396047592163,
      "learning_rate": 1.2149115580995755e-05,
      "loss": 0.1285,
      "num_input_tokens_seen": 50935792,
      "step": 26875
    },
    {
      "epoch": 4.385023248225793,
      "grad_norm": 0.1742076426744461,
      "learning_rate": 1.2140695444779227e-05,
      "loss": 0.1134,
      "num_input_tokens_seen": 50944832,
      "step": 26880
    },
    {
      "epoch": 4.38583897544661,
      "grad_norm": 0.18069857358932495,
      "learning_rate": 1.2132277291629066e-05,
      "loss": 0.05,
      "num_input_tokens_seen": 50955072,
      "step": 26885
    },
    {
      "epoch": 4.386654702667428,
      "grad_norm": 0.7233547568321228,
      "learning_rate": 1.2123861122843458e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 50964608,
      "step": 26890
    },
    {
      "epoch": 4.387470429888245,
      "grad_norm": 0.5723063349723816,
      "learning_rate": 1.2115446939720271e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 50974960,
      "step": 26895
    },
    {
      "epoch": 4.388286157109063,
      "grad_norm": 0.7944124341011047,
      "learning_rate": 1.210703474355708e-05,
      "loss": 0.2275,
      "num_input_tokens_seen": 50984128,
      "step": 26900
    },
    {
      "epoch": 4.38910188432988,
      "grad_norm": 0.6433953046798706,
      "learning_rate": 1.2098624535651164e-05,
      "loss": 0.2433,
      "num_input_tokens_seen": 50994320,
      "step": 26905
    },
    {
      "epoch": 4.389917611550698,
      "grad_norm": 0.7092493176460266,
      "learning_rate": 1.2090216317299477e-05,
      "loss": 0.0965,
      "num_input_tokens_seen": 51003552,
      "step": 26910
    },
    {
      "epoch": 4.390733338771515,
      "grad_norm": 0.1312032788991928,
      "learning_rate": 1.2081810089798668e-05,
      "loss": 0.0277,
      "num_input_tokens_seen": 51013424,
      "step": 26915
    },
    {
      "epoch": 4.391549065992332,
      "grad_norm": 0.20252981781959534,
      "learning_rate": 1.2073405854445072e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 51022208,
      "step": 26920
    },
    {
      "epoch": 4.392364793213149,
      "grad_norm": 1.7271838188171387,
      "learning_rate": 1.206500361253474e-05,
      "loss": 0.0846,
      "num_input_tokens_seen": 51031216,
      "step": 26925
    },
    {
      "epoch": 4.393180520433967,
      "grad_norm": 1.4182292222976685,
      "learning_rate": 1.2056603365363409e-05,
      "loss": 0.2406,
      "num_input_tokens_seen": 51039904,
      "step": 26930
    },
    {
      "epoch": 4.393996247654784,
      "grad_norm": 0.6881333589553833,
      "learning_rate": 1.2048205114226487e-05,
      "loss": 0.2213,
      "num_input_tokens_seen": 51049440,
      "step": 26935
    },
    {
      "epoch": 4.394811974875601,
      "grad_norm": 1.4487842321395874,
      "learning_rate": 1.2039808860419102e-05,
      "loss": 0.1304,
      "num_input_tokens_seen": 51059152,
      "step": 26940
    },
    {
      "epoch": 4.395627702096419,
      "grad_norm": 0.7600029110908508,
      "learning_rate": 1.2031414605236066e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 51069184,
      "step": 26945
    },
    {
      "epoch": 4.3964434293172365,
      "grad_norm": 0.061623234301805496,
      "learning_rate": 1.2023022349971862e-05,
      "loss": 0.0549,
      "num_input_tokens_seen": 51077360,
      "step": 26950
    },
    {
      "epoch": 4.397259156538054,
      "grad_norm": 1.3401293754577637,
      "learning_rate": 1.20146320959207e-05,
      "loss": 0.1616,
      "num_input_tokens_seen": 51086480,
      "step": 26955
    },
    {
      "epoch": 4.398074883758871,
      "grad_norm": 0.1665574014186859,
      "learning_rate": 1.2006243844376445e-05,
      "loss": 0.0818,
      "num_input_tokens_seen": 51096656,
      "step": 26960
    },
    {
      "epoch": 4.398890610979688,
      "grad_norm": 1.7136021852493286,
      "learning_rate": 1.1997857596632678e-05,
      "loss": 0.0872,
      "num_input_tokens_seen": 51106592,
      "step": 26965
    },
    {
      "epoch": 4.399706338200506,
      "grad_norm": 1.3818132877349854,
      "learning_rate": 1.1989473353982672e-05,
      "loss": 0.1745,
      "num_input_tokens_seen": 51114560,
      "step": 26970
    },
    {
      "epoch": 4.400522065421323,
      "grad_norm": 0.31699275970458984,
      "learning_rate": 1.198109111771937e-05,
      "loss": 0.1812,
      "num_input_tokens_seen": 51123776,
      "step": 26975
    },
    {
      "epoch": 4.40133779264214,
      "grad_norm": 0.09631223231554031,
      "learning_rate": 1.197271088913543e-05,
      "loss": 0.0444,
      "num_input_tokens_seen": 51133296,
      "step": 26980
    },
    {
      "epoch": 4.402153519862958,
      "grad_norm": 1.3127562999725342,
      "learning_rate": 1.1964332669523182e-05,
      "loss": 0.0791,
      "num_input_tokens_seen": 51142848,
      "step": 26985
    },
    {
      "epoch": 4.402969247083775,
      "grad_norm": 0.08236891031265259,
      "learning_rate": 1.1955956460174645e-05,
      "loss": 0.0322,
      "num_input_tokens_seen": 51152160,
      "step": 26990
    },
    {
      "epoch": 4.4037849743045925,
      "grad_norm": 0.16053080558776855,
      "learning_rate": 1.1947582262381552e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 51161584,
      "step": 26995
    },
    {
      "epoch": 4.40460070152541,
      "grad_norm": 1.05189847946167,
      "learning_rate": 1.1939210077435293e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 51170976,
      "step": 27000
    },
    {
      "epoch": 4.40460070152541,
      "eval_loss": 0.14137300848960876,
      "eval_runtime": 133.2135,
      "eval_samples_per_second": 20.456,
      "eval_steps_per_second": 10.232,
      "num_input_tokens_seen": 51170976,
      "step": 27000
    },
    {
      "epoch": 4.405416428746228,
      "grad_norm": 0.12628260254859924,
      "learning_rate": 1.193083990662697e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 51180560,
      "step": 27005
    },
    {
      "epoch": 4.406232155967045,
      "grad_norm": 0.3567066788673401,
      "learning_rate": 1.192247175124738e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 51190784,
      "step": 27010
    },
    {
      "epoch": 4.407047883187862,
      "grad_norm": 1.1193310022354126,
      "learning_rate": 1.191410561258698e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 51201424,
      "step": 27015
    },
    {
      "epoch": 4.407863610408679,
      "grad_norm": 1.3746317625045776,
      "learning_rate": 1.1905741491935944e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 51211536,
      "step": 27020
    },
    {
      "epoch": 4.408679337629497,
      "grad_norm": 0.13273559510707855,
      "learning_rate": 1.1897379390584129e-05,
      "loss": 0.1292,
      "num_input_tokens_seen": 51220704,
      "step": 27025
    },
    {
      "epoch": 4.409495064850314,
      "grad_norm": 1.239406704902649,
      "learning_rate": 1.1889019309821062e-05,
      "loss": 0.2673,
      "num_input_tokens_seen": 51229376,
      "step": 27030
    },
    {
      "epoch": 4.410310792071131,
      "grad_norm": 0.4319859743118286,
      "learning_rate": 1.188066125093599e-05,
      "loss": 0.1086,
      "num_input_tokens_seen": 51239424,
      "step": 27035
    },
    {
      "epoch": 4.411126519291948,
      "grad_norm": 0.48509353399276733,
      "learning_rate": 1.1872305215217811e-05,
      "loss": 0.0638,
      "num_input_tokens_seen": 51248384,
      "step": 27040
    },
    {
      "epoch": 4.4119422465127665,
      "grad_norm": 1.2922486066818237,
      "learning_rate": 1.186395120395514e-05,
      "loss": 0.1863,
      "num_input_tokens_seen": 51257856,
      "step": 27045
    },
    {
      "epoch": 4.412757973733584,
      "grad_norm": 1.001059889793396,
      "learning_rate": 1.1855599218436283e-05,
      "loss": 0.0574,
      "num_input_tokens_seen": 51266256,
      "step": 27050
    },
    {
      "epoch": 4.413573700954401,
      "grad_norm": 1.2853665351867676,
      "learning_rate": 1.1847249259949209e-05,
      "loss": 0.1937,
      "num_input_tokens_seen": 51276128,
      "step": 27055
    },
    {
      "epoch": 4.414389428175218,
      "grad_norm": 1.0141421556472778,
      "learning_rate": 1.1838901329781574e-05,
      "loss": 0.0674,
      "num_input_tokens_seen": 51285504,
      "step": 27060
    },
    {
      "epoch": 4.415205155396036,
      "grad_norm": 0.41450732946395874,
      "learning_rate": 1.1830555429220758e-05,
      "loss": 0.1457,
      "num_input_tokens_seen": 51294320,
      "step": 27065
    },
    {
      "epoch": 4.416020882616853,
      "grad_norm": 0.17285320162773132,
      "learning_rate": 1.1822211559553784e-05,
      "loss": 0.0857,
      "num_input_tokens_seen": 51303952,
      "step": 27070
    },
    {
      "epoch": 4.41683660983767,
      "grad_norm": 1.293521761894226,
      "learning_rate": 1.18138697220674e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 51314336,
      "step": 27075
    },
    {
      "epoch": 4.417652337058487,
      "grad_norm": 0.5184613466262817,
      "learning_rate": 1.1805529918048e-05,
      "loss": 0.1636,
      "num_input_tokens_seen": 51324656,
      "step": 27080
    },
    {
      "epoch": 4.418468064279305,
      "grad_norm": 0.23193760216236115,
      "learning_rate": 1.1797192148781702e-05,
      "loss": 0.0234,
      "num_input_tokens_seen": 51333664,
      "step": 27085
    },
    {
      "epoch": 4.419283791500122,
      "grad_norm": 0.4052279591560364,
      "learning_rate": 1.1788856415554297e-05,
      "loss": 0.1976,
      "num_input_tokens_seen": 51343824,
      "step": 27090
    },
    {
      "epoch": 4.4200995187209395,
      "grad_norm": 0.13343511521816254,
      "learning_rate": 1.1780522719651249e-05,
      "loss": 0.0716,
      "num_input_tokens_seen": 51352864,
      "step": 27095
    },
    {
      "epoch": 4.420915245941757,
      "grad_norm": 0.5373573303222656,
      "learning_rate": 1.1772191062357721e-05,
      "loss": 0.087,
      "num_input_tokens_seen": 51362864,
      "step": 27100
    },
    {
      "epoch": 4.421730973162575,
      "grad_norm": 0.30107590556144714,
      "learning_rate": 1.1763861444958573e-05,
      "loss": 0.0308,
      "num_input_tokens_seen": 51372928,
      "step": 27105
    },
    {
      "epoch": 4.422546700383392,
      "grad_norm": 0.7128900289535522,
      "learning_rate": 1.1755533868738317e-05,
      "loss": 0.0501,
      "num_input_tokens_seen": 51383680,
      "step": 27110
    },
    {
      "epoch": 4.423362427604209,
      "grad_norm": 1.4681414365768433,
      "learning_rate": 1.1747208334981185e-05,
      "loss": 0.2202,
      "num_input_tokens_seen": 51393088,
      "step": 27115
    },
    {
      "epoch": 4.424178154825027,
      "grad_norm": 0.20391882956027985,
      "learning_rate": 1.1738884844971067e-05,
      "loss": 0.1811,
      "num_input_tokens_seen": 51402672,
      "step": 27120
    },
    {
      "epoch": 4.424993882045844,
      "grad_norm": 0.40445998311042786,
      "learning_rate": 1.1730563399991563e-05,
      "loss": 0.1203,
      "num_input_tokens_seen": 51412448,
      "step": 27125
    },
    {
      "epoch": 4.425809609266661,
      "grad_norm": 0.08933311700820923,
      "learning_rate": 1.1722244001325938e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 51422160,
      "step": 27130
    },
    {
      "epoch": 4.426625336487478,
      "grad_norm": 0.10419183224439621,
      "learning_rate": 1.1713926650257137e-05,
      "loss": 0.0769,
      "num_input_tokens_seen": 51431568,
      "step": 27135
    },
    {
      "epoch": 4.4274410637082955,
      "grad_norm": 0.1636592596769333,
      "learning_rate": 1.170561134806781e-05,
      "loss": 0.0269,
      "num_input_tokens_seen": 51441504,
      "step": 27140
    },
    {
      "epoch": 4.4282567909291135,
      "grad_norm": 0.18205556273460388,
      "learning_rate": 1.1697298096040287e-05,
      "loss": 0.0771,
      "num_input_tokens_seen": 51451376,
      "step": 27145
    },
    {
      "epoch": 4.429072518149931,
      "grad_norm": 0.13897116482257843,
      "learning_rate": 1.1688986895456567e-05,
      "loss": 0.0757,
      "num_input_tokens_seen": 51459728,
      "step": 27150
    },
    {
      "epoch": 4.429888245370748,
      "grad_norm": 1.3399856090545654,
      "learning_rate": 1.1680677747598349e-05,
      "loss": 0.1336,
      "num_input_tokens_seen": 51468816,
      "step": 27155
    },
    {
      "epoch": 4.430703972591566,
      "grad_norm": 0.0700974091887474,
      "learning_rate": 1.1672370653746995e-05,
      "loss": 0.1306,
      "num_input_tokens_seen": 51478512,
      "step": 27160
    },
    {
      "epoch": 4.431519699812383,
      "grad_norm": 0.2823188006877899,
      "learning_rate": 1.166406561518357e-05,
      "loss": 0.0561,
      "num_input_tokens_seen": 51487456,
      "step": 27165
    },
    {
      "epoch": 4.4323354270332,
      "grad_norm": 0.9843860864639282,
      "learning_rate": 1.1655762633188826e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 51497136,
      "step": 27170
    },
    {
      "epoch": 4.433151154254017,
      "grad_norm": 0.2917509078979492,
      "learning_rate": 1.1647461709043172e-05,
      "loss": 0.0998,
      "num_input_tokens_seen": 51507456,
      "step": 27175
    },
    {
      "epoch": 4.433966881474835,
      "grad_norm": 0.56422358751297,
      "learning_rate": 1.1639162844026722e-05,
      "loss": 0.0875,
      "num_input_tokens_seen": 51518480,
      "step": 27180
    },
    {
      "epoch": 4.434782608695652,
      "grad_norm": 1.6276400089263916,
      "learning_rate": 1.163086603941927e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 51529136,
      "step": 27185
    },
    {
      "epoch": 4.4355983359164695,
      "grad_norm": 0.806408703327179,
      "learning_rate": 1.1622571296500273e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 51539376,
      "step": 27190
    },
    {
      "epoch": 4.436414063137287,
      "grad_norm": 0.06319291144609451,
      "learning_rate": 1.1614278616548904e-05,
      "loss": 0.1481,
      "num_input_tokens_seen": 51549776,
      "step": 27195
    },
    {
      "epoch": 4.437229790358105,
      "grad_norm": 0.8136113286018372,
      "learning_rate": 1.1605988000843986e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 51559504,
      "step": 27200
    },
    {
      "epoch": 4.437229790358105,
      "eval_loss": 0.14091970026493073,
      "eval_runtime": 133.3017,
      "eval_samples_per_second": 20.442,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 51559504,
      "step": 27200
    },
    {
      "epoch": 4.438045517578922,
      "grad_norm": 0.14994236826896667,
      "learning_rate": 1.1597699450664028e-05,
      "loss": 0.0584,
      "num_input_tokens_seen": 51568240,
      "step": 27205
    },
    {
      "epoch": 4.438861244799739,
      "grad_norm": 0.21359674632549286,
      "learning_rate": 1.1589412967287252e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 51576560,
      "step": 27210
    },
    {
      "epoch": 4.439676972020556,
      "grad_norm": 1.0660008192062378,
      "learning_rate": 1.1581128551991514e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 51585904,
      "step": 27215
    },
    {
      "epoch": 4.440492699241374,
      "grad_norm": 1.3712197542190552,
      "learning_rate": 1.1572846206054383e-05,
      "loss": 0.2084,
      "num_input_tokens_seen": 51594832,
      "step": 27220
    },
    {
      "epoch": 4.441308426462191,
      "grad_norm": 3.5255558490753174,
      "learning_rate": 1.1564565930753113e-05,
      "loss": 0.2757,
      "num_input_tokens_seen": 51605152,
      "step": 27225
    },
    {
      "epoch": 4.442124153683008,
      "grad_norm": 0.34231579303741455,
      "learning_rate": 1.1556287727364606e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 51615120,
      "step": 27230
    },
    {
      "epoch": 4.442939880903825,
      "grad_norm": 0.8286688327789307,
      "learning_rate": 1.1548011597165489e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 51624816,
      "step": 27235
    },
    {
      "epoch": 4.443755608124643,
      "grad_norm": 0.2791954278945923,
      "learning_rate": 1.1539737541432019e-05,
      "loss": 0.1819,
      "num_input_tokens_seen": 51633440,
      "step": 27240
    },
    {
      "epoch": 4.444571335345461,
      "grad_norm": 0.9210508465766907,
      "learning_rate": 1.1531465561440174e-05,
      "loss": 0.1924,
      "num_input_tokens_seen": 51642336,
      "step": 27245
    },
    {
      "epoch": 4.445387062566278,
      "grad_norm": 0.37476393580436707,
      "learning_rate": 1.1523195658465605e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 51651488,
      "step": 27250
    },
    {
      "epoch": 4.446202789787095,
      "grad_norm": 1.6110395193099976,
      "learning_rate": 1.1514927833783618e-05,
      "loss": 0.2967,
      "num_input_tokens_seen": 51660512,
      "step": 27255
    },
    {
      "epoch": 4.447018517007913,
      "grad_norm": 0.1629030555486679,
      "learning_rate": 1.150666208866922e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 51669952,
      "step": 27260
    },
    {
      "epoch": 4.44783424422873,
      "grad_norm": 0.9758684039115906,
      "learning_rate": 1.1498398424397106e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 51679136,
      "step": 27265
    },
    {
      "epoch": 4.448649971449547,
      "grad_norm": 0.9560902118682861,
      "learning_rate": 1.1490136842241628e-05,
      "loss": 0.0705,
      "num_input_tokens_seen": 51689200,
      "step": 27270
    },
    {
      "epoch": 4.449465698670364,
      "grad_norm": 0.5324872136116028,
      "learning_rate": 1.1481877343476813e-05,
      "loss": 0.0647,
      "num_input_tokens_seen": 51697648,
      "step": 27275
    },
    {
      "epoch": 4.450281425891182,
      "grad_norm": 2.0817766189575195,
      "learning_rate": 1.14736199293764e-05,
      "loss": 0.1879,
      "num_input_tokens_seen": 51707328,
      "step": 27280
    },
    {
      "epoch": 4.451097153111999,
      "grad_norm": 0.06217152252793312,
      "learning_rate": 1.1465364601213771e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 51717312,
      "step": 27285
    },
    {
      "epoch": 4.4519128803328165,
      "grad_norm": 0.9472647309303284,
      "learning_rate": 1.1457111360262012e-05,
      "loss": 0.0846,
      "num_input_tokens_seen": 51726224,
      "step": 27290
    },
    {
      "epoch": 4.4527286075536345,
      "grad_norm": 0.0765179693698883,
      "learning_rate": 1.1448860207793869e-05,
      "loss": 0.0336,
      "num_input_tokens_seen": 51735872,
      "step": 27295
    },
    {
      "epoch": 4.453544334774452,
      "grad_norm": 0.25605931878089905,
      "learning_rate": 1.144061114508177e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 51745552,
      "step": 27300
    },
    {
      "epoch": 4.454360061995269,
      "grad_norm": 0.3021511137485504,
      "learning_rate": 1.1432364173397842e-05,
      "loss": 0.0771,
      "num_input_tokens_seen": 51754192,
      "step": 27305
    },
    {
      "epoch": 4.455175789216086,
      "grad_norm": 0.872700572013855,
      "learning_rate": 1.1424119294013852e-05,
      "loss": 0.0777,
      "num_input_tokens_seen": 51763168,
      "step": 27310
    },
    {
      "epoch": 4.455991516436903,
      "grad_norm": 0.40462395548820496,
      "learning_rate": 1.1415876508201279e-05,
      "loss": 0.1061,
      "num_input_tokens_seen": 51773136,
      "step": 27315
    },
    {
      "epoch": 4.456807243657721,
      "grad_norm": 0.030635377392172813,
      "learning_rate": 1.140763581723125e-05,
      "loss": 0.0568,
      "num_input_tokens_seen": 51781728,
      "step": 27320
    },
    {
      "epoch": 4.457622970878538,
      "grad_norm": 1.7066810131072998,
      "learning_rate": 1.1399397222374588e-05,
      "loss": 0.0627,
      "num_input_tokens_seen": 51791184,
      "step": 27325
    },
    {
      "epoch": 4.458438698099355,
      "grad_norm": 1.2433029413223267,
      "learning_rate": 1.1391160724901804e-05,
      "loss": 0.0701,
      "num_input_tokens_seen": 51801808,
      "step": 27330
    },
    {
      "epoch": 4.459254425320173,
      "grad_norm": 0.3972489833831787,
      "learning_rate": 1.138292632608304e-05,
      "loss": 0.0426,
      "num_input_tokens_seen": 51810336,
      "step": 27335
    },
    {
      "epoch": 4.4600701525409905,
      "grad_norm": 0.8784456253051758,
      "learning_rate": 1.1374694027188174e-05,
      "loss": 0.1344,
      "num_input_tokens_seen": 51819760,
      "step": 27340
    },
    {
      "epoch": 4.460885879761808,
      "grad_norm": 0.7401664853096008,
      "learning_rate": 1.1366463829486711e-05,
      "loss": 0.1134,
      "num_input_tokens_seen": 51827936,
      "step": 27345
    },
    {
      "epoch": 4.461701606982625,
      "grad_norm": 0.05567861720919609,
      "learning_rate": 1.1358235734247849e-05,
      "loss": 0.0868,
      "num_input_tokens_seen": 51836512,
      "step": 27350
    },
    {
      "epoch": 4.462517334203443,
      "grad_norm": 0.9404476881027222,
      "learning_rate": 1.1350009742740478e-05,
      "loss": 0.1201,
      "num_input_tokens_seen": 51845312,
      "step": 27355
    },
    {
      "epoch": 4.46333306142426,
      "grad_norm": 0.9934600591659546,
      "learning_rate": 1.134178585623313e-05,
      "loss": 0.2059,
      "num_input_tokens_seen": 51855168,
      "step": 27360
    },
    {
      "epoch": 4.464148788645077,
      "grad_norm": 1.1535437107086182,
      "learning_rate": 1.1333564075994047e-05,
      "loss": 0.169,
      "num_input_tokens_seen": 51863712,
      "step": 27365
    },
    {
      "epoch": 4.464964515865894,
      "grad_norm": 0.43002572655677795,
      "learning_rate": 1.1325344403291133e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 51873088,
      "step": 27370
    },
    {
      "epoch": 4.465780243086712,
      "grad_norm": 0.1125381663441658,
      "learning_rate": 1.1317126839391951e-05,
      "loss": 0.053,
      "num_input_tokens_seen": 51882128,
      "step": 27375
    },
    {
      "epoch": 4.466595970307529,
      "grad_norm": 1.3933615684509277,
      "learning_rate": 1.1308911385563766e-05,
      "loss": 0.2197,
      "num_input_tokens_seen": 51890048,
      "step": 27380
    },
    {
      "epoch": 4.467411697528346,
      "grad_norm": 0.03934551775455475,
      "learning_rate": 1.1300698043073494e-05,
      "loss": 0.1761,
      "num_input_tokens_seen": 51898592,
      "step": 27385
    },
    {
      "epoch": 4.468227424749164,
      "grad_norm": 0.6030627489089966,
      "learning_rate": 1.1292486813187736e-05,
      "loss": 0.1013,
      "num_input_tokens_seen": 51908624,
      "step": 27390
    },
    {
      "epoch": 4.469043151969982,
      "grad_norm": 0.2185482531785965,
      "learning_rate": 1.1284277697172782e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 51918752,
      "step": 27395
    },
    {
      "epoch": 4.469858879190799,
      "grad_norm": 0.2487226277589798,
      "learning_rate": 1.127607069629456e-05,
      "loss": 0.0327,
      "num_input_tokens_seen": 51928704,
      "step": 27400
    },
    {
      "epoch": 4.469858879190799,
      "eval_loss": 0.14145998656749725,
      "eval_runtime": 133.2109,
      "eval_samples_per_second": 20.456,
      "eval_steps_per_second": 10.232,
      "num_input_tokens_seen": 51928704,
      "step": 27400
    },
    {
      "epoch": 4.470674606411616,
      "grad_norm": 0.08640682697296143,
      "learning_rate": 1.1267865811818701e-05,
      "loss": 0.0867,
      "num_input_tokens_seen": 51938480,
      "step": 27405
    },
    {
      "epoch": 4.471490333632433,
      "grad_norm": 0.05588792636990547,
      "learning_rate": 1.1259663045010513e-05,
      "loss": 0.0676,
      "num_input_tokens_seen": 51947376,
      "step": 27410
    },
    {
      "epoch": 4.472306060853251,
      "grad_norm": 1.5083585977554321,
      "learning_rate": 1.1251462397134957e-05,
      "loss": 0.2306,
      "num_input_tokens_seen": 51957136,
      "step": 27415
    },
    {
      "epoch": 4.473121788074068,
      "grad_norm": 1.3415488004684448,
      "learning_rate": 1.1243263869456664e-05,
      "loss": 0.0433,
      "num_input_tokens_seen": 51966752,
      "step": 27420
    },
    {
      "epoch": 4.473937515294885,
      "grad_norm": 0.11117078363895416,
      "learning_rate": 1.1235067463239967e-05,
      "loss": 0.0567,
      "num_input_tokens_seen": 51975520,
      "step": 27425
    },
    {
      "epoch": 4.474753242515702,
      "grad_norm": 0.4976637065410614,
      "learning_rate": 1.122687317974884e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 51983584,
      "step": 27430
    },
    {
      "epoch": 4.47556896973652,
      "grad_norm": 0.4143129587173462,
      "learning_rate": 1.1218681020246963e-05,
      "loss": 0.2091,
      "num_input_tokens_seen": 51993088,
      "step": 27435
    },
    {
      "epoch": 4.4763846969573375,
      "grad_norm": 0.045960623770952225,
      "learning_rate": 1.1210490985997652e-05,
      "loss": 0.1665,
      "num_input_tokens_seen": 52002912,
      "step": 27440
    },
    {
      "epoch": 4.477200424178155,
      "grad_norm": 0.30021411180496216,
      "learning_rate": 1.1202303078263917e-05,
      "loss": 0.1273,
      "num_input_tokens_seen": 52011392,
      "step": 27445
    },
    {
      "epoch": 4.478016151398972,
      "grad_norm": 0.2322993278503418,
      "learning_rate": 1.1194117298308451e-05,
      "loss": 0.0989,
      "num_input_tokens_seen": 52020816,
      "step": 27450
    },
    {
      "epoch": 4.47883187861979,
      "grad_norm": 1.568151831626892,
      "learning_rate": 1.1185933647393585e-05,
      "loss": 0.234,
      "num_input_tokens_seen": 52030576,
      "step": 27455
    },
    {
      "epoch": 4.479647605840607,
      "grad_norm": 0.16049548983573914,
      "learning_rate": 1.1177752126781354e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 52040432,
      "step": 27460
    },
    {
      "epoch": 4.480463333061424,
      "grad_norm": 0.4738207459449768,
      "learning_rate": 1.1169572737733441e-05,
      "loss": 0.0997,
      "num_input_tokens_seen": 52050400,
      "step": 27465
    },
    {
      "epoch": 4.481279060282241,
      "grad_norm": 0.15287211537361145,
      "learning_rate": 1.1161395481511216e-05,
      "loss": 0.1,
      "num_input_tokens_seen": 52059792,
      "step": 27470
    },
    {
      "epoch": 4.482094787503059,
      "grad_norm": 1.25472891330719,
      "learning_rate": 1.1153220359375722e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 52068864,
      "step": 27475
    },
    {
      "epoch": 4.482910514723876,
      "grad_norm": 1.3113585710525513,
      "learning_rate": 1.114504737258765e-05,
      "loss": 0.1503,
      "num_input_tokens_seen": 52077760,
      "step": 27480
    },
    {
      "epoch": 4.4837262419446935,
      "grad_norm": 1.2679725885391235,
      "learning_rate": 1.1136876522407393e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 52085968,
      "step": 27485
    },
    {
      "epoch": 4.484541969165511,
      "grad_norm": 0.5404841303825378,
      "learning_rate": 1.1128707810094985e-05,
      "loss": 0.1874,
      "num_input_tokens_seen": 52094592,
      "step": 27490
    },
    {
      "epoch": 4.485357696386329,
      "grad_norm": 0.29679661989212036,
      "learning_rate": 1.1120541236910157e-05,
      "loss": 0.0352,
      "num_input_tokens_seen": 52103216,
      "step": 27495
    },
    {
      "epoch": 4.486173423607146,
      "grad_norm": 0.04716229438781738,
      "learning_rate": 1.111237680411229e-05,
      "loss": 0.1032,
      "num_input_tokens_seen": 52112496,
      "step": 27500
    },
    {
      "epoch": 4.486989150827963,
      "grad_norm": 0.21373732388019562,
      "learning_rate": 1.1104214512960433e-05,
      "loss": 0.0998,
      "num_input_tokens_seen": 52121312,
      "step": 27505
    },
    {
      "epoch": 4.487804878048781,
      "grad_norm": 0.31351685523986816,
      "learning_rate": 1.1096054364713327e-05,
      "loss": 0.0875,
      "num_input_tokens_seen": 52129920,
      "step": 27510
    },
    {
      "epoch": 4.488620605269598,
      "grad_norm": 0.12638506293296814,
      "learning_rate": 1.1087896360629371e-05,
      "loss": 0.1179,
      "num_input_tokens_seen": 52137904,
      "step": 27515
    },
    {
      "epoch": 4.489436332490415,
      "grad_norm": 0.5691043734550476,
      "learning_rate": 1.107974050196662e-05,
      "loss": 0.0203,
      "num_input_tokens_seen": 52147856,
      "step": 27520
    },
    {
      "epoch": 4.490252059711232,
      "grad_norm": 0.0926654264330864,
      "learning_rate": 1.1071586789982816e-05,
      "loss": 0.2257,
      "num_input_tokens_seen": 52158416,
      "step": 27525
    },
    {
      "epoch": 4.49106778693205,
      "grad_norm": 0.3056863248348236,
      "learning_rate": 1.1063435225935373e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 52166832,
      "step": 27530
    },
    {
      "epoch": 4.4918835141528675,
      "grad_norm": 0.5530314445495605,
      "learning_rate": 1.1055285811081348e-05,
      "loss": 0.0868,
      "num_input_tokens_seen": 52176832,
      "step": 27535
    },
    {
      "epoch": 4.492699241373685,
      "grad_norm": 0.10085386782884598,
      "learning_rate": 1.1047138546677499e-05,
      "loss": 0.0824,
      "num_input_tokens_seen": 52185728,
      "step": 27540
    },
    {
      "epoch": 4.493514968594502,
      "grad_norm": 0.11893795430660248,
      "learning_rate": 1.1038993433980219e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 52194736,
      "step": 27545
    },
    {
      "epoch": 4.49433069581532,
      "grad_norm": 0.06913020461797714,
      "learning_rate": 1.1030850474245597e-05,
      "loss": 0.2079,
      "num_input_tokens_seen": 52203968,
      "step": 27550
    },
    {
      "epoch": 4.495146423036137,
      "grad_norm": 0.4913328289985657,
      "learning_rate": 1.102270966872939e-05,
      "loss": 0.3009,
      "num_input_tokens_seen": 52212912,
      "step": 27555
    },
    {
      "epoch": 4.495962150256954,
      "grad_norm": 0.841378927230835,
      "learning_rate": 1.1014571018687e-05,
      "loss": 0.1853,
      "num_input_tokens_seen": 52222128,
      "step": 27560
    },
    {
      "epoch": 4.496777877477771,
      "grad_norm": 0.7370992302894592,
      "learning_rate": 1.1006434525373502e-05,
      "loss": 0.1116,
      "num_input_tokens_seen": 52231376,
      "step": 27565
    },
    {
      "epoch": 4.497593604698589,
      "grad_norm": 0.47933292388916016,
      "learning_rate": 1.0998300190043664e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 52241680,
      "step": 27570
    },
    {
      "epoch": 4.498409331919406,
      "grad_norm": 0.39628857374191284,
      "learning_rate": 1.0990168013951882e-05,
      "loss": 0.1922,
      "num_input_tokens_seen": 52251920,
      "step": 27575
    },
    {
      "epoch": 4.499225059140223,
      "grad_norm": 1.3303701877593994,
      "learning_rate": 1.0982037998352263e-05,
      "loss": 0.1367,
      "num_input_tokens_seen": 52261184,
      "step": 27580
    },
    {
      "epoch": 4.5000407863610405,
      "grad_norm": 0.8657395839691162,
      "learning_rate": 1.0973910144498534e-05,
      "loss": 0.3099,
      "num_input_tokens_seen": 52271072,
      "step": 27585
    },
    {
      "epoch": 4.500856513581859,
      "grad_norm": 0.2819633185863495,
      "learning_rate": 1.0965784453644123e-05,
      "loss": 0.078,
      "num_input_tokens_seen": 52280368,
      "step": 27590
    },
    {
      "epoch": 4.501672240802676,
      "grad_norm": 0.8135043978691101,
      "learning_rate": 1.0957660927042127e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 52289424,
      "step": 27595
    },
    {
      "epoch": 4.502487968023493,
      "grad_norm": 0.7774990200996399,
      "learning_rate": 1.094953956594527e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 52297776,
      "step": 27600
    },
    {
      "epoch": 4.502487968023493,
      "eval_loss": 0.14110371470451355,
      "eval_runtime": 133.2797,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 52297776,
      "step": 27600
    },
    {
      "epoch": 4.50330369524431,
      "grad_norm": 0.2132091373205185,
      "learning_rate": 1.0941420371605981e-05,
      "loss": 0.0792,
      "num_input_tokens_seen": 52307984,
      "step": 27605
    },
    {
      "epoch": 4.504119422465128,
      "grad_norm": 0.2076333463191986,
      "learning_rate": 1.0933303345276354e-05,
      "loss": 0.1155,
      "num_input_tokens_seen": 52317856,
      "step": 27610
    },
    {
      "epoch": 4.504935149685945,
      "grad_norm": 0.42644238471984863,
      "learning_rate": 1.0925188488208112e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 52327536,
      "step": 27615
    },
    {
      "epoch": 4.505750876906762,
      "grad_norm": 0.41861119866371155,
      "learning_rate": 1.0917075801652694e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 52336928,
      "step": 27620
    },
    {
      "epoch": 4.506566604127579,
      "grad_norm": 0.43271705508232117,
      "learning_rate": 1.0908965286861151e-05,
      "loss": 0.0458,
      "num_input_tokens_seen": 52345712,
      "step": 27625
    },
    {
      "epoch": 4.507382331348397,
      "grad_norm": 1.4301786422729492,
      "learning_rate": 1.090085694508425e-05,
      "loss": 0.1246,
      "num_input_tokens_seen": 52354480,
      "step": 27630
    },
    {
      "epoch": 4.5081980585692145,
      "grad_norm": 0.49692633748054504,
      "learning_rate": 1.089275077757238e-05,
      "loss": 0.1645,
      "num_input_tokens_seen": 52363088,
      "step": 27635
    },
    {
      "epoch": 4.509013785790032,
      "grad_norm": 0.0476246140897274,
      "learning_rate": 1.0884646785575633e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 52371248,
      "step": 27640
    },
    {
      "epoch": 4.50982951301085,
      "grad_norm": 0.05349041149020195,
      "learning_rate": 1.0876544970343728e-05,
      "loss": 0.1998,
      "num_input_tokens_seen": 52379760,
      "step": 27645
    },
    {
      "epoch": 4.510645240231667,
      "grad_norm": 0.2280408889055252,
      "learning_rate": 1.0868445333126082e-05,
      "loss": 0.0772,
      "num_input_tokens_seen": 52389136,
      "step": 27650
    },
    {
      "epoch": 4.511460967452484,
      "grad_norm": 0.4499795734882355,
      "learning_rate": 1.0860347875171745e-05,
      "loss": 0.0575,
      "num_input_tokens_seen": 52397680,
      "step": 27655
    },
    {
      "epoch": 4.512276694673301,
      "grad_norm": 0.47501900792121887,
      "learning_rate": 1.0852252597729465e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 52406240,
      "step": 27660
    },
    {
      "epoch": 4.513092421894118,
      "grad_norm": 0.08142238855361938,
      "learning_rate": 1.0844159502047615e-05,
      "loss": 0.0505,
      "num_input_tokens_seen": 52416208,
      "step": 27665
    },
    {
      "epoch": 4.513908149114936,
      "grad_norm": 1.7086091041564941,
      "learning_rate": 1.0836068589374265e-05,
      "loss": 0.1289,
      "num_input_tokens_seen": 52425024,
      "step": 27670
    },
    {
      "epoch": 4.514723876335753,
      "grad_norm": 1.026830792427063,
      "learning_rate": 1.0827979860957144e-05,
      "loss": 0.1083,
      "num_input_tokens_seen": 52436112,
      "step": 27675
    },
    {
      "epoch": 4.5155396035565705,
      "grad_norm": 0.02166321687400341,
      "learning_rate": 1.0819893318043615e-05,
      "loss": 0.0343,
      "num_input_tokens_seen": 52445184,
      "step": 27680
    },
    {
      "epoch": 4.5163553307773885,
      "grad_norm": 0.2818484604358673,
      "learning_rate": 1.0811808961880734e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 52455264,
      "step": 27685
    },
    {
      "epoch": 4.517171057998206,
      "grad_norm": 0.8111869692802429,
      "learning_rate": 1.080372679371522e-05,
      "loss": 0.0993,
      "num_input_tokens_seen": 52463344,
      "step": 27690
    },
    {
      "epoch": 4.517986785219023,
      "grad_norm": 0.636929988861084,
      "learning_rate": 1.0795646814793428e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 52471984,
      "step": 27695
    },
    {
      "epoch": 4.51880251243984,
      "grad_norm": 0.36309173703193665,
      "learning_rate": 1.078756902636141e-05,
      "loss": 0.0753,
      "num_input_tokens_seen": 52481216,
      "step": 27700
    },
    {
      "epoch": 4.519618239660657,
      "grad_norm": 1.5168009996414185,
      "learning_rate": 1.077949342966485e-05,
      "loss": 0.1404,
      "num_input_tokens_seen": 52490384,
      "step": 27705
    },
    {
      "epoch": 4.520433966881475,
      "grad_norm": 0.2596338391304016,
      "learning_rate": 1.0771420025949103e-05,
      "loss": 0.1455,
      "num_input_tokens_seen": 52498720,
      "step": 27710
    },
    {
      "epoch": 4.521249694102292,
      "grad_norm": 0.20175841450691223,
      "learning_rate": 1.0763348816459204e-05,
      "loss": 0.044,
      "num_input_tokens_seen": 52507456,
      "step": 27715
    },
    {
      "epoch": 4.522065421323109,
      "grad_norm": 1.4529461860656738,
      "learning_rate": 1.0755279802439816e-05,
      "loss": 0.2151,
      "num_input_tokens_seen": 52517936,
      "step": 27720
    },
    {
      "epoch": 4.522881148543927,
      "grad_norm": 0.3853796422481537,
      "learning_rate": 1.0747212985135293e-05,
      "loss": 0.0373,
      "num_input_tokens_seen": 52526928,
      "step": 27725
    },
    {
      "epoch": 4.523696875764744,
      "grad_norm": 1.7144659757614136,
      "learning_rate": 1.073914836578965e-05,
      "loss": 0.1512,
      "num_input_tokens_seen": 52536224,
      "step": 27730
    },
    {
      "epoch": 4.524512602985562,
      "grad_norm": 0.08957459777593613,
      "learning_rate": 1.0731085945646529e-05,
      "loss": 0.0705,
      "num_input_tokens_seen": 52545600,
      "step": 27735
    },
    {
      "epoch": 4.525328330206379,
      "grad_norm": 0.2522841989994049,
      "learning_rate": 1.0723025725949285e-05,
      "loss": 0.0754,
      "num_input_tokens_seen": 52555488,
      "step": 27740
    },
    {
      "epoch": 4.526144057427197,
      "grad_norm": 1.0104821920394897,
      "learning_rate": 1.0714967707940875e-05,
      "loss": 0.0786,
      "num_input_tokens_seen": 52566096,
      "step": 27745
    },
    {
      "epoch": 4.526959784648014,
      "grad_norm": 1.1269630193710327,
      "learning_rate": 1.0706911892863963e-05,
      "loss": 0.0723,
      "num_input_tokens_seen": 52575072,
      "step": 27750
    },
    {
      "epoch": 4.527775511868831,
      "grad_norm": 0.05135876685380936,
      "learning_rate": 1.0698858281960866e-05,
      "loss": 0.1622,
      "num_input_tokens_seen": 52584016,
      "step": 27755
    },
    {
      "epoch": 4.528591239089648,
      "grad_norm": 1.195358395576477,
      "learning_rate": 1.069080687647353e-05,
      "loss": 0.0933,
      "num_input_tokens_seen": 52593776,
      "step": 27760
    },
    {
      "epoch": 4.529406966310466,
      "grad_norm": 0.09952865540981293,
      "learning_rate": 1.0682757677643596e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 52602272,
      "step": 27765
    },
    {
      "epoch": 4.530222693531283,
      "grad_norm": 0.2727014720439911,
      "learning_rate": 1.0674710686712359e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 52611648,
      "step": 27770
    },
    {
      "epoch": 4.5310384207521,
      "grad_norm": 0.6036110520362854,
      "learning_rate": 1.0666665904920756e-05,
      "loss": 0.0719,
      "num_input_tokens_seen": 52622144,
      "step": 27775
    },
    {
      "epoch": 4.5318541479729175,
      "grad_norm": 0.5701115131378174,
      "learning_rate": 1.0658623333509385e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 52632080,
      "step": 27780
    },
    {
      "epoch": 4.5326698751937355,
      "grad_norm": 0.3832101821899414,
      "learning_rate": 1.0650582973718532e-05,
      "loss": 0.0186,
      "num_input_tokens_seen": 52641168,
      "step": 27785
    },
    {
      "epoch": 4.533485602414553,
      "grad_norm": 2.8497774600982666,
      "learning_rate": 1.0642544826788098e-05,
      "loss": 0.2078,
      "num_input_tokens_seen": 52650528,
      "step": 27790
    },
    {
      "epoch": 4.53430132963537,
      "grad_norm": 0.0777372494339943,
      "learning_rate": 1.063450889395769e-05,
      "loss": 0.2147,
      "num_input_tokens_seen": 52660272,
      "step": 27795
    },
    {
      "epoch": 4.535117056856187,
      "grad_norm": 0.21167223155498505,
      "learning_rate": 1.062647517646653e-05,
      "loss": 0.0763,
      "num_input_tokens_seen": 52669472,
      "step": 27800
    },
    {
      "epoch": 4.535117056856187,
      "eval_loss": 0.1419409066438675,
      "eval_runtime": 133.2208,
      "eval_samples_per_second": 20.455,
      "eval_steps_per_second": 10.231,
      "num_input_tokens_seen": 52669472,
      "step": 27800
    },
    {
      "epoch": 4.535932784077005,
      "grad_norm": 0.6855279803276062,
      "learning_rate": 1.0618443675553527e-05,
      "loss": 0.1836,
      "num_input_tokens_seen": 52678480,
      "step": 27805
    },
    {
      "epoch": 4.536748511297822,
      "grad_norm": 0.36871227622032166,
      "learning_rate": 1.0610414392457247e-05,
      "loss": 0.0588,
      "num_input_tokens_seen": 52686736,
      "step": 27810
    },
    {
      "epoch": 4.537564238518639,
      "grad_norm": 1.462719202041626,
      "learning_rate": 1.0602387328415888e-05,
      "loss": 0.1455,
      "num_input_tokens_seen": 52698000,
      "step": 27815
    },
    {
      "epoch": 4.538379965739456,
      "grad_norm": 0.22410297393798828,
      "learning_rate": 1.0594362484667347e-05,
      "loss": 0.095,
      "num_input_tokens_seen": 52708400,
      "step": 27820
    },
    {
      "epoch": 4.539195692960274,
      "grad_norm": 0.30836114287376404,
      "learning_rate": 1.0586339862449132e-05,
      "loss": 0.1983,
      "num_input_tokens_seen": 52717616,
      "step": 27825
    },
    {
      "epoch": 4.5400114201810915,
      "grad_norm": 1.2150899171829224,
      "learning_rate": 1.0578319462998445e-05,
      "loss": 0.0796,
      "num_input_tokens_seen": 52726368,
      "step": 27830
    },
    {
      "epoch": 4.540827147401909,
      "grad_norm": 0.1801927089691162,
      "learning_rate": 1.057030128755214e-05,
      "loss": 0.0625,
      "num_input_tokens_seen": 52736368,
      "step": 27835
    },
    {
      "epoch": 4.541642874622726,
      "grad_norm": 0.894519567489624,
      "learning_rate": 1.0562285337346703e-05,
      "loss": 0.1817,
      "num_input_tokens_seen": 52744704,
      "step": 27840
    },
    {
      "epoch": 4.542458601843544,
      "grad_norm": 0.04188871756196022,
      "learning_rate": 1.0554271613618308e-05,
      "loss": 0.1147,
      "num_input_tokens_seen": 52753584,
      "step": 27845
    },
    {
      "epoch": 4.543274329064361,
      "grad_norm": 0.42848464846611023,
      "learning_rate": 1.054626011760276e-05,
      "loss": 0.1046,
      "num_input_tokens_seen": 52762912,
      "step": 27850
    },
    {
      "epoch": 4.544090056285178,
      "grad_norm": 0.6087641716003418,
      "learning_rate": 1.0538250850535549e-05,
      "loss": 0.0642,
      "num_input_tokens_seen": 52772304,
      "step": 27855
    },
    {
      "epoch": 4.544905783505996,
      "grad_norm": 0.05405621603131294,
      "learning_rate": 1.0530243813651794e-05,
      "loss": 0.1178,
      "num_input_tokens_seen": 52782176,
      "step": 27860
    },
    {
      "epoch": 4.545721510726813,
      "grad_norm": 0.2143438756465912,
      "learning_rate": 1.0522239008186271e-05,
      "loss": 0.0382,
      "num_input_tokens_seen": 52791456,
      "step": 27865
    },
    {
      "epoch": 4.54653723794763,
      "grad_norm": 1.9793226718902588,
      "learning_rate": 1.0514236435373434e-05,
      "loss": 0.0998,
      "num_input_tokens_seen": 52800960,
      "step": 27870
    },
    {
      "epoch": 4.5473529651684474,
      "grad_norm": 0.03741812705993652,
      "learning_rate": 1.0506236096447386e-05,
      "loss": 0.0708,
      "num_input_tokens_seen": 52812304,
      "step": 27875
    },
    {
      "epoch": 4.548168692389265,
      "grad_norm": 1.7396351099014282,
      "learning_rate": 1.049823799264186e-05,
      "loss": 0.2222,
      "num_input_tokens_seen": 52820416,
      "step": 27880
    },
    {
      "epoch": 4.548984419610083,
      "grad_norm": 0.5795575380325317,
      "learning_rate": 1.049024212519028e-05,
      "loss": 0.0719,
      "num_input_tokens_seen": 52830592,
      "step": 27885
    },
    {
      "epoch": 4.5498001468309,
      "grad_norm": 0.25769826769828796,
      "learning_rate": 1.0482248495325713e-05,
      "loss": 0.2138,
      "num_input_tokens_seen": 52839840,
      "step": 27890
    },
    {
      "epoch": 4.550615874051717,
      "grad_norm": 0.8283419013023376,
      "learning_rate": 1.047425710428086e-05,
      "loss": 0.0734,
      "num_input_tokens_seen": 52849216,
      "step": 27895
    },
    {
      "epoch": 4.551431601272535,
      "grad_norm": 1.026239275932312,
      "learning_rate": 1.0466267953288114e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 52858496,
      "step": 27900
    },
    {
      "epoch": 4.552247328493352,
      "grad_norm": 0.5916745066642761,
      "learning_rate": 1.0458281043579482e-05,
      "loss": 0.1135,
      "num_input_tokens_seen": 52867392,
      "step": 27905
    },
    {
      "epoch": 4.553063055714169,
      "grad_norm": 0.12698616087436676,
      "learning_rate": 1.0450296376386657e-05,
      "loss": 0.0314,
      "num_input_tokens_seen": 52877664,
      "step": 27910
    },
    {
      "epoch": 4.553878782934986,
      "grad_norm": 0.0931742787361145,
      "learning_rate": 1.044231395294098e-05,
      "loss": 0.0847,
      "num_input_tokens_seen": 52887152,
      "step": 27915
    },
    {
      "epoch": 4.554694510155803,
      "grad_norm": 0.1089840680360794,
      "learning_rate": 1.0434333774473435e-05,
      "loss": 0.2245,
      "num_input_tokens_seen": 52895888,
      "step": 27920
    },
    {
      "epoch": 4.555510237376621,
      "grad_norm": 1.129213809967041,
      "learning_rate": 1.0426355842214657e-05,
      "loss": 0.0521,
      "num_input_tokens_seen": 52904720,
      "step": 27925
    },
    {
      "epoch": 4.5563259645974385,
      "grad_norm": 0.9051026701927185,
      "learning_rate": 1.0418380157394963e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 52915216,
      "step": 27930
    },
    {
      "epoch": 4.557141691818256,
      "grad_norm": 1.1360644102096558,
      "learning_rate": 1.0410406721244281e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 52924064,
      "step": 27935
    },
    {
      "epoch": 4.557957419039074,
      "grad_norm": 0.14964178204536438,
      "learning_rate": 1.0402435534992238e-05,
      "loss": 0.219,
      "num_input_tokens_seen": 52933472,
      "step": 27940
    },
    {
      "epoch": 4.558773146259891,
      "grad_norm": 0.20746088027954102,
      "learning_rate": 1.0394466599868071e-05,
      "loss": 0.0697,
      "num_input_tokens_seen": 52940992,
      "step": 27945
    },
    {
      "epoch": 4.559588873480708,
      "grad_norm": 0.08735840767621994,
      "learning_rate": 1.0386499917100697e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 52950432,
      "step": 27950
    },
    {
      "epoch": 4.560404600701525,
      "grad_norm": 0.11249101161956787,
      "learning_rate": 1.0378535487918692e-05,
      "loss": 0.267,
      "num_input_tokens_seen": 52959904,
      "step": 27955
    },
    {
      "epoch": 4.561220327922343,
      "grad_norm": 1.1180205345153809,
      "learning_rate": 1.037057331355025e-05,
      "loss": 0.1754,
      "num_input_tokens_seen": 52970128,
      "step": 27960
    },
    {
      "epoch": 4.56203605514316,
      "grad_norm": 0.847355306148529,
      "learning_rate": 1.0362613395223247e-05,
      "loss": 0.1002,
      "num_input_tokens_seen": 52979344,
      "step": 27965
    },
    {
      "epoch": 4.562851782363977,
      "grad_norm": 1.548225998878479,
      "learning_rate": 1.0354655734165212e-05,
      "loss": 0.1222,
      "num_input_tokens_seen": 52988464,
      "step": 27970
    },
    {
      "epoch": 4.5636675095847945,
      "grad_norm": 1.07615065574646,
      "learning_rate": 1.03467003316033e-05,
      "loss": 0.1106,
      "num_input_tokens_seen": 52998144,
      "step": 27975
    },
    {
      "epoch": 4.5644832368056125,
      "grad_norm": 0.2838824689388275,
      "learning_rate": 1.033874718876435e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 53008160,
      "step": 27980
    },
    {
      "epoch": 4.56529896402643,
      "grad_norm": 0.10779839009046555,
      "learning_rate": 1.0330796306874818e-05,
      "loss": 0.219,
      "num_input_tokens_seen": 53016832,
      "step": 27985
    },
    {
      "epoch": 4.566114691247247,
      "grad_norm": 2.3480920791625977,
      "learning_rate": 1.032284768716085e-05,
      "loss": 0.1085,
      "num_input_tokens_seen": 53026352,
      "step": 27990
    },
    {
      "epoch": 4.566930418468064,
      "grad_norm": 0.40327051281929016,
      "learning_rate": 1.0314901330848206e-05,
      "loss": 0.1515,
      "num_input_tokens_seen": 53035824,
      "step": 27995
    },
    {
      "epoch": 4.567746145688882,
      "grad_norm": 0.9669285416603088,
      "learning_rate": 1.030695723916233e-05,
      "loss": 0.0933,
      "num_input_tokens_seen": 53045856,
      "step": 28000
    },
    {
      "epoch": 4.567746145688882,
      "eval_loss": 0.14284028112888336,
      "eval_runtime": 133.2764,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 53045856,
      "step": 28000
    },
    {
      "epoch": 4.568561872909699,
      "grad_norm": 3.452141761779785,
      "learning_rate": 1.0299015413328289e-05,
      "loss": 0.3622,
      "num_input_tokens_seen": 53054368,
      "step": 28005
    },
    {
      "epoch": 4.569377600130516,
      "grad_norm": 0.9856499433517456,
      "learning_rate": 1.0291075854570809e-05,
      "loss": 0.1306,
      "num_input_tokens_seen": 53063888,
      "step": 28010
    },
    {
      "epoch": 4.570193327351333,
      "grad_norm": 1.0571246147155762,
      "learning_rate": 1.0283138564114275e-05,
      "loss": 0.2165,
      "num_input_tokens_seen": 53073968,
      "step": 28015
    },
    {
      "epoch": 4.571009054572151,
      "grad_norm": 0.7715465426445007,
      "learning_rate": 1.027520354318273e-05,
      "loss": 0.0967,
      "num_input_tokens_seen": 53083440,
      "step": 28020
    },
    {
      "epoch": 4.5718247817929685,
      "grad_norm": 0.5034065246582031,
      "learning_rate": 1.0267270792999828e-05,
      "loss": 0.1645,
      "num_input_tokens_seen": 53093776,
      "step": 28025
    },
    {
      "epoch": 4.572640509013786,
      "grad_norm": 1.014807105064392,
      "learning_rate": 1.0259340314788919e-05,
      "loss": 0.0792,
      "num_input_tokens_seen": 53103776,
      "step": 28030
    },
    {
      "epoch": 4.573456236234604,
      "grad_norm": 0.36612093448638916,
      "learning_rate": 1.0251412109772979e-05,
      "loss": 0.0598,
      "num_input_tokens_seen": 53113776,
      "step": 28035
    },
    {
      "epoch": 4.574271963455421,
      "grad_norm": 1.6570930480957031,
      "learning_rate": 1.0243486179174627e-05,
      "loss": 0.1924,
      "num_input_tokens_seen": 53122704,
      "step": 28040
    },
    {
      "epoch": 4.575087690676238,
      "grad_norm": 0.727710485458374,
      "learning_rate": 1.0235562524216158e-05,
      "loss": 0.1086,
      "num_input_tokens_seen": 53131632,
      "step": 28045
    },
    {
      "epoch": 4.575903417897055,
      "grad_norm": 0.17292913794517517,
      "learning_rate": 1.022764114611948e-05,
      "loss": 0.2146,
      "num_input_tokens_seen": 53140944,
      "step": 28050
    },
    {
      "epoch": 4.576719145117872,
      "grad_norm": 0.8807471990585327,
      "learning_rate": 1.0219722046106178e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 53149120,
      "step": 28055
    },
    {
      "epoch": 4.57753487233869,
      "grad_norm": 0.13653872907161713,
      "learning_rate": 1.0211805225397486e-05,
      "loss": 0.0994,
      "num_input_tokens_seen": 53159536,
      "step": 28060
    },
    {
      "epoch": 4.578350599559507,
      "grad_norm": 0.21963973343372345,
      "learning_rate": 1.020389068521426e-05,
      "loss": 0.3273,
      "num_input_tokens_seen": 53169216,
      "step": 28065
    },
    {
      "epoch": 4.579166326780324,
      "grad_norm": 0.6249472498893738,
      "learning_rate": 1.0195978426777039e-05,
      "loss": 0.1027,
      "num_input_tokens_seen": 53177568,
      "step": 28070
    },
    {
      "epoch": 4.5799820540011424,
      "grad_norm": 0.8065568804740906,
      "learning_rate": 1.0188068451305982e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 53187200,
      "step": 28075
    },
    {
      "epoch": 4.58079778122196,
      "grad_norm": 0.9732988476753235,
      "learning_rate": 1.0180160760020902e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 53195424,
      "step": 28080
    },
    {
      "epoch": 4.581613508442777,
      "grad_norm": 0.32744893431663513,
      "learning_rate": 1.0172255354141278e-05,
      "loss": 0.1934,
      "num_input_tokens_seen": 53203760,
      "step": 28085
    },
    {
      "epoch": 4.582429235663594,
      "grad_norm": 1.7722055912017822,
      "learning_rate": 1.0164352234886205e-05,
      "loss": 0.0958,
      "num_input_tokens_seen": 53212560,
      "step": 28090
    },
    {
      "epoch": 4.583244962884411,
      "grad_norm": 0.9475269913673401,
      "learning_rate": 1.0156451403474454e-05,
      "loss": 0.0604,
      "num_input_tokens_seen": 53224240,
      "step": 28095
    },
    {
      "epoch": 4.584060690105229,
      "grad_norm": 0.04759649932384491,
      "learning_rate": 1.0148552861124443e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 53232944,
      "step": 28100
    },
    {
      "epoch": 4.584876417326046,
      "grad_norm": 0.6950235962867737,
      "learning_rate": 1.0140656609054205e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 53242544,
      "step": 28105
    },
    {
      "epoch": 4.585692144546863,
      "grad_norm": 1.0249972343444824,
      "learning_rate": 1.0132762648481455e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 53252416,
      "step": 28110
    },
    {
      "epoch": 4.586507871767681,
      "grad_norm": 0.7396267652511597,
      "learning_rate": 1.0124870980623543e-05,
      "loss": 0.0619,
      "num_input_tokens_seen": 53263088,
      "step": 28115
    },
    {
      "epoch": 4.587323598988498,
      "grad_norm": 0.6568978428840637,
      "learning_rate": 1.0116981606697453e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 53272976,
      "step": 28120
    },
    {
      "epoch": 4.5881393262093155,
      "grad_norm": 0.5215577483177185,
      "learning_rate": 1.0109094527919838e-05,
      "loss": 0.1055,
      "num_input_tokens_seen": 53283408,
      "step": 28125
    },
    {
      "epoch": 4.588955053430133,
      "grad_norm": 0.16225750744342804,
      "learning_rate": 1.010120974550697e-05,
      "loss": 0.0384,
      "num_input_tokens_seen": 53292832,
      "step": 28130
    },
    {
      "epoch": 4.589770780650951,
      "grad_norm": 1.2988455295562744,
      "learning_rate": 1.0093327260674795e-05,
      "loss": 0.3383,
      "num_input_tokens_seen": 53302560,
      "step": 28135
    },
    {
      "epoch": 4.590586507871768,
      "grad_norm": 0.6668362021446228,
      "learning_rate": 1.0085447074638878e-05,
      "loss": 0.2376,
      "num_input_tokens_seen": 53312464,
      "step": 28140
    },
    {
      "epoch": 4.591402235092585,
      "grad_norm": 1.6703463792800903,
      "learning_rate": 1.0077569188614461e-05,
      "loss": 0.1789,
      "num_input_tokens_seen": 53321376,
      "step": 28145
    },
    {
      "epoch": 4.592217962313402,
      "grad_norm": 0.1855964958667755,
      "learning_rate": 1.0069693603816393e-05,
      "loss": 0.2072,
      "num_input_tokens_seen": 53331456,
      "step": 28150
    },
    {
      "epoch": 4.59303368953422,
      "grad_norm": 1.256590485572815,
      "learning_rate": 1.0061820321459204e-05,
      "loss": 0.1676,
      "num_input_tokens_seen": 53341648,
      "step": 28155
    },
    {
      "epoch": 4.593849416755037,
      "grad_norm": 0.11210394650697708,
      "learning_rate": 1.0053949342757038e-05,
      "loss": 0.0431,
      "num_input_tokens_seen": 53351680,
      "step": 28160
    },
    {
      "epoch": 4.594665143975854,
      "grad_norm": 1.6117424964904785,
      "learning_rate": 1.0046080668923717e-05,
      "loss": 0.2028,
      "num_input_tokens_seen": 53362192,
      "step": 28165
    },
    {
      "epoch": 4.5954808711966715,
      "grad_norm": 0.04337572306394577,
      "learning_rate": 1.003821430117267e-05,
      "loss": 0.0703,
      "num_input_tokens_seen": 53372176,
      "step": 28170
    },
    {
      "epoch": 4.5962965984174895,
      "grad_norm": 0.018622035160660744,
      "learning_rate": 1.0030350240716999e-05,
      "loss": 0.114,
      "num_input_tokens_seen": 53381968,
      "step": 28175
    },
    {
      "epoch": 4.597112325638307,
      "grad_norm": 0.9716535210609436,
      "learning_rate": 1.0022488488769449e-05,
      "loss": 0.1836,
      "num_input_tokens_seen": 53391088,
      "step": 28180
    },
    {
      "epoch": 4.597928052859124,
      "grad_norm": 0.26666611433029175,
      "learning_rate": 1.0014629046542387e-05,
      "loss": 0.3509,
      "num_input_tokens_seen": 53400848,
      "step": 28185
    },
    {
      "epoch": 4.598743780079941,
      "grad_norm": 0.5830006003379822,
      "learning_rate": 1.0006771915247842e-05,
      "loss": 0.0898,
      "num_input_tokens_seen": 53410096,
      "step": 28190
    },
    {
      "epoch": 4.599559507300759,
      "grad_norm": 0.4491601586341858,
      "learning_rate": 9.998917096097495e-06,
      "loss": 0.1224,
      "num_input_tokens_seen": 53419584,
      "step": 28195
    },
    {
      "epoch": 4.600375234521576,
      "grad_norm": 0.22073659300804138,
      "learning_rate": 9.991064590302638e-06,
      "loss": 0.2031,
      "num_input_tokens_seen": 53429232,
      "step": 28200
    },
    {
      "epoch": 4.600375234521576,
      "eval_loss": 0.14233005046844482,
      "eval_runtime": 133.2494,
      "eval_samples_per_second": 20.45,
      "eval_steps_per_second": 10.229,
      "num_input_tokens_seen": 53429232,
      "step": 28200
    },
    {
      "epoch": 4.601190961742393,
      "grad_norm": 0.9701893925666809,
      "learning_rate": 9.983214399074241e-06,
      "loss": 0.0385,
      "num_input_tokens_seen": 53438896,
      "step": 28205
    },
    {
      "epoch": 4.602006688963211,
      "grad_norm": 1.601943850517273,
      "learning_rate": 9.975366523622893e-06,
      "loss": 0.1687,
      "num_input_tokens_seen": 53448192,
      "step": 28210
    },
    {
      "epoch": 4.602822416184028,
      "grad_norm": 0.509634256362915,
      "learning_rate": 9.967520965158841e-06,
      "loss": 0.0874,
      "num_input_tokens_seen": 53458096,
      "step": 28215
    },
    {
      "epoch": 4.6036381434048455,
      "grad_norm": 0.8983921408653259,
      "learning_rate": 9.95967772489197e-06,
      "loss": 0.2313,
      "num_input_tokens_seen": 53466960,
      "step": 28220
    },
    {
      "epoch": 4.604453870625663,
      "grad_norm": 0.7938709259033203,
      "learning_rate": 9.951836804031794e-06,
      "loss": 0.0922,
      "num_input_tokens_seen": 53477456,
      "step": 28225
    },
    {
      "epoch": 4.60526959784648,
      "grad_norm": 3.039022445678711,
      "learning_rate": 9.943998203787489e-06,
      "loss": 0.0968,
      "num_input_tokens_seen": 53486656,
      "step": 28230
    },
    {
      "epoch": 4.606085325067298,
      "grad_norm": 0.7127401232719421,
      "learning_rate": 9.936161925367874e-06,
      "loss": 0.0861,
      "num_input_tokens_seen": 53494944,
      "step": 28235
    },
    {
      "epoch": 4.606901052288115,
      "grad_norm": 0.5830114483833313,
      "learning_rate": 9.928327969981386e-06,
      "loss": 0.3484,
      "num_input_tokens_seen": 53505008,
      "step": 28240
    },
    {
      "epoch": 4.607716779508932,
      "grad_norm": 0.603247880935669,
      "learning_rate": 9.920496338836135e-06,
      "loss": 0.0879,
      "num_input_tokens_seen": 53515120,
      "step": 28245
    },
    {
      "epoch": 4.60853250672975,
      "grad_norm": 0.04944638907909393,
      "learning_rate": 9.912667033139844e-06,
      "loss": 0.0416,
      "num_input_tokens_seen": 53524672,
      "step": 28250
    },
    {
      "epoch": 4.609348233950567,
      "grad_norm": 0.10340046882629395,
      "learning_rate": 9.904840054099893e-06,
      "loss": 0.0585,
      "num_input_tokens_seen": 53534608,
      "step": 28255
    },
    {
      "epoch": 4.610163961171384,
      "grad_norm": 0.3569892346858978,
      "learning_rate": 9.897015402923312e-06,
      "loss": 0.0287,
      "num_input_tokens_seen": 53543136,
      "step": 28260
    },
    {
      "epoch": 4.610979688392201,
      "grad_norm": 0.08093566447496414,
      "learning_rate": 9.889193080816744e-06,
      "loss": 0.0333,
      "num_input_tokens_seen": 53552224,
      "step": 28265
    },
    {
      "epoch": 4.6117954156130185,
      "grad_norm": 0.8848446011543274,
      "learning_rate": 9.881373088986498e-06,
      "loss": 0.1059,
      "num_input_tokens_seen": 53560752,
      "step": 28270
    },
    {
      "epoch": 4.6126111428338366,
      "grad_norm": 0.5134599804878235,
      "learning_rate": 9.873555428638523e-06,
      "loss": 0.0732,
      "num_input_tokens_seen": 53569680,
      "step": 28275
    },
    {
      "epoch": 4.613426870054654,
      "grad_norm": 1.5809922218322754,
      "learning_rate": 9.865740100978383e-06,
      "loss": 0.3528,
      "num_input_tokens_seen": 53579536,
      "step": 28280
    },
    {
      "epoch": 4.614242597275471,
      "grad_norm": 1.6054463386535645,
      "learning_rate": 9.857927107211315e-06,
      "loss": 0.2625,
      "num_input_tokens_seen": 53590336,
      "step": 28285
    },
    {
      "epoch": 4.615058324496289,
      "grad_norm": 0.24315056204795837,
      "learning_rate": 9.850116448542177e-06,
      "loss": 0.0702,
      "num_input_tokens_seen": 53599280,
      "step": 28290
    },
    {
      "epoch": 4.615874051717106,
      "grad_norm": 0.2529877722263336,
      "learning_rate": 9.842308126175457e-06,
      "loss": 0.0568,
      "num_input_tokens_seen": 53608192,
      "step": 28295
    },
    {
      "epoch": 4.616689778937923,
      "grad_norm": 0.06919069588184357,
      "learning_rate": 9.834502141315315e-06,
      "loss": 0.0342,
      "num_input_tokens_seen": 53617168,
      "step": 28300
    },
    {
      "epoch": 4.61750550615874,
      "grad_norm": 1.6217542886734009,
      "learning_rate": 9.82669849516552e-06,
      "loss": 0.1038,
      "num_input_tokens_seen": 53626912,
      "step": 28305
    },
    {
      "epoch": 4.618321233379558,
      "grad_norm": 0.1581013798713684,
      "learning_rate": 9.818897188929493e-06,
      "loss": 0.0523,
      "num_input_tokens_seen": 53637264,
      "step": 28310
    },
    {
      "epoch": 4.619136960600375,
      "grad_norm": 0.2919295132160187,
      "learning_rate": 9.811098223810309e-06,
      "loss": 0.0366,
      "num_input_tokens_seen": 53646368,
      "step": 28315
    },
    {
      "epoch": 4.6199526878211925,
      "grad_norm": 1.108205795288086,
      "learning_rate": 9.803301601010641e-06,
      "loss": 0.1036,
      "num_input_tokens_seen": 53656752,
      "step": 28320
    },
    {
      "epoch": 4.62076841504201,
      "grad_norm": 0.5474547743797302,
      "learning_rate": 9.795507321732853e-06,
      "loss": 0.051,
      "num_input_tokens_seen": 53666432,
      "step": 28325
    },
    {
      "epoch": 4.621584142262828,
      "grad_norm": 1.4456021785736084,
      "learning_rate": 9.787715387178898e-06,
      "loss": 0.3245,
      "num_input_tokens_seen": 53677312,
      "step": 28330
    },
    {
      "epoch": 4.622399869483645,
      "grad_norm": 0.937919557094574,
      "learning_rate": 9.779925798550399e-06,
      "loss": 0.2459,
      "num_input_tokens_seen": 53686848,
      "step": 28335
    },
    {
      "epoch": 4.623215596704462,
      "grad_norm": 1.2459874153137207,
      "learning_rate": 9.772138557048619e-06,
      "loss": 0.1877,
      "num_input_tokens_seen": 53696240,
      "step": 28340
    },
    {
      "epoch": 4.624031323925279,
      "grad_norm": 1.3511654138565063,
      "learning_rate": 9.764353663874426e-06,
      "loss": 0.1514,
      "num_input_tokens_seen": 53705472,
      "step": 28345
    },
    {
      "epoch": 4.624847051146097,
      "grad_norm": 1.1621536016464233,
      "learning_rate": 9.756571120228375e-06,
      "loss": 0.0763,
      "num_input_tokens_seen": 53713968,
      "step": 28350
    },
    {
      "epoch": 4.625662778366914,
      "grad_norm": 1.0436571836471558,
      "learning_rate": 9.748790927310605e-06,
      "loss": 0.1188,
      "num_input_tokens_seen": 53722864,
      "step": 28355
    },
    {
      "epoch": 4.626478505587731,
      "grad_norm": 1.2076524496078491,
      "learning_rate": 9.741013086320946e-06,
      "loss": 0.1969,
      "num_input_tokens_seen": 53732624,
      "step": 28360
    },
    {
      "epoch": 4.6272942328085485,
      "grad_norm": 1.2043691873550415,
      "learning_rate": 9.733237598458821e-06,
      "loss": 0.0846,
      "num_input_tokens_seen": 53741840,
      "step": 28365
    },
    {
      "epoch": 4.6281099600293665,
      "grad_norm": 0.3688063621520996,
      "learning_rate": 9.725464464923308e-06,
      "loss": 0.0505,
      "num_input_tokens_seen": 53751952,
      "step": 28370
    },
    {
      "epoch": 4.628925687250184,
      "grad_norm": 0.1564406007528305,
      "learning_rate": 9.717693686913123e-06,
      "loss": 0.2103,
      "num_input_tokens_seen": 53761920,
      "step": 28375
    },
    {
      "epoch": 4.629741414471001,
      "grad_norm": 0.18844014406204224,
      "learning_rate": 9.709925265626632e-06,
      "loss": 0.1203,
      "num_input_tokens_seen": 53771728,
      "step": 28380
    },
    {
      "epoch": 4.630557141691818,
      "grad_norm": 1.2988320589065552,
      "learning_rate": 9.702159202261801e-06,
      "loss": 0.2557,
      "num_input_tokens_seen": 53781040,
      "step": 28385
    },
    {
      "epoch": 4.631372868912636,
      "grad_norm": 1.1119929552078247,
      "learning_rate": 9.694395498016268e-06,
      "loss": 0.0908,
      "num_input_tokens_seen": 53791456,
      "step": 28390
    },
    {
      "epoch": 4.632188596133453,
      "grad_norm": 0.48938173055648804,
      "learning_rate": 9.686634154087298e-06,
      "loss": 0.0882,
      "num_input_tokens_seen": 53801216,
      "step": 28395
    },
    {
      "epoch": 4.63300432335427,
      "grad_norm": 1.4015018939971924,
      "learning_rate": 9.678875171671776e-06,
      "loss": 0.1147,
      "num_input_tokens_seen": 53810560,
      "step": 28400
    },
    {
      "epoch": 4.63300432335427,
      "eval_loss": 0.1426970660686493,
      "eval_runtime": 133.292,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 53810560,
      "step": 28400
    },
    {
      "epoch": 4.633820050575087,
      "grad_norm": 0.7596133947372437,
      "learning_rate": 9.671118551966246e-06,
      "loss": 0.0362,
      "num_input_tokens_seen": 53819264,
      "step": 28405
    },
    {
      "epoch": 4.634635777795905,
      "grad_norm": 0.8813177347183228,
      "learning_rate": 9.66336429616686e-06,
      "loss": 0.0626,
      "num_input_tokens_seen": 53828480,
      "step": 28410
    },
    {
      "epoch": 4.635451505016722,
      "grad_norm": 0.1265476942062378,
      "learning_rate": 9.655612405469436e-06,
      "loss": 0.1666,
      "num_input_tokens_seen": 53837616,
      "step": 28415
    },
    {
      "epoch": 4.63626723223754,
      "grad_norm": 0.06759189814329147,
      "learning_rate": 9.647862881069413e-06,
      "loss": 0.0464,
      "num_input_tokens_seen": 53847344,
      "step": 28420
    },
    {
      "epoch": 4.637082959458358,
      "grad_norm": 1.2670507431030273,
      "learning_rate": 9.640115724161855e-06,
      "loss": 0.1819,
      "num_input_tokens_seen": 53858080,
      "step": 28425
    },
    {
      "epoch": 4.637898686679175,
      "grad_norm": 0.07135065644979477,
      "learning_rate": 9.632370935941483e-06,
      "loss": 0.0498,
      "num_input_tokens_seen": 53867616,
      "step": 28430
    },
    {
      "epoch": 4.638714413899992,
      "grad_norm": 1.3745005130767822,
      "learning_rate": 9.624628517602634e-06,
      "loss": 0.1918,
      "num_input_tokens_seen": 53877520,
      "step": 28435
    },
    {
      "epoch": 4.639530141120809,
      "grad_norm": 1.135995626449585,
      "learning_rate": 9.61688847033928e-06,
      "loss": 0.137,
      "num_input_tokens_seen": 53887440,
      "step": 28440
    },
    {
      "epoch": 4.640345868341626,
      "grad_norm": 0.5617063045501709,
      "learning_rate": 9.609150795345051e-06,
      "loss": 0.185,
      "num_input_tokens_seen": 53896480,
      "step": 28445
    },
    {
      "epoch": 4.641161595562444,
      "grad_norm": 0.06447824090719223,
      "learning_rate": 9.601415493813171e-06,
      "loss": 0.0563,
      "num_input_tokens_seen": 53905328,
      "step": 28450
    },
    {
      "epoch": 4.641977322783261,
      "grad_norm": 0.3069101572036743,
      "learning_rate": 9.593682566936533e-06,
      "loss": 0.1883,
      "num_input_tokens_seen": 53914784,
      "step": 28455
    },
    {
      "epoch": 4.642793050004078,
      "grad_norm": 1.3179761171340942,
      "learning_rate": 9.58595201590766e-06,
      "loss": 0.1348,
      "num_input_tokens_seen": 53922720,
      "step": 28460
    },
    {
      "epoch": 4.643608777224896,
      "grad_norm": 0.8520383834838867,
      "learning_rate": 9.578223841918681e-06,
      "loss": 0.3075,
      "num_input_tokens_seen": 53932768,
      "step": 28465
    },
    {
      "epoch": 4.6444245044457135,
      "grad_norm": 0.2991557717323303,
      "learning_rate": 9.570498046161389e-06,
      "loss": 0.1255,
      "num_input_tokens_seen": 53942640,
      "step": 28470
    },
    {
      "epoch": 4.645240231666531,
      "grad_norm": 0.5722357630729675,
      "learning_rate": 9.562774629827206e-06,
      "loss": 0.1151,
      "num_input_tokens_seen": 53952592,
      "step": 28475
    },
    {
      "epoch": 4.646055958887348,
      "grad_norm": 0.04746527969837189,
      "learning_rate": 9.555053594107163e-06,
      "loss": 0.1773,
      "num_input_tokens_seen": 53962464,
      "step": 28480
    },
    {
      "epoch": 4.646871686108166,
      "grad_norm": 0.4513491988182068,
      "learning_rate": 9.547334940191957e-06,
      "loss": 0.0579,
      "num_input_tokens_seen": 53970576,
      "step": 28485
    },
    {
      "epoch": 4.647687413328983,
      "grad_norm": 1.0744879245758057,
      "learning_rate": 9.539618669271886e-06,
      "loss": 0.1608,
      "num_input_tokens_seen": 53979776,
      "step": 28490
    },
    {
      "epoch": 4.6485031405498,
      "grad_norm": 0.3616397976875305,
      "learning_rate": 9.531904782536904e-06,
      "loss": 0.0472,
      "num_input_tokens_seen": 53989568,
      "step": 28495
    },
    {
      "epoch": 4.649318867770617,
      "grad_norm": 1.0295714139938354,
      "learning_rate": 9.524193281176597e-06,
      "loss": 0.1424,
      "num_input_tokens_seen": 53999552,
      "step": 28500
    },
    {
      "epoch": 4.650134594991435,
      "grad_norm": 0.40572240948677063,
      "learning_rate": 9.516484166380165e-06,
      "loss": 0.1344,
      "num_input_tokens_seen": 54008912,
      "step": 28505
    },
    {
      "epoch": 4.650950322212252,
      "grad_norm": 0.8563937544822693,
      "learning_rate": 9.508777439336447e-06,
      "loss": 0.0595,
      "num_input_tokens_seen": 54019312,
      "step": 28510
    },
    {
      "epoch": 4.6517660494330695,
      "grad_norm": 0.11621597409248352,
      "learning_rate": 9.50107310123393e-06,
      "loss": 0.0698,
      "num_input_tokens_seen": 54029296,
      "step": 28515
    },
    {
      "epoch": 4.652581776653887,
      "grad_norm": 0.0884329080581665,
      "learning_rate": 9.493371153260702e-06,
      "loss": 0.0538,
      "num_input_tokens_seen": 54038448,
      "step": 28520
    },
    {
      "epoch": 4.653397503874705,
      "grad_norm": 2.0373740196228027,
      "learning_rate": 9.485671596604523e-06,
      "loss": 0.215,
      "num_input_tokens_seen": 54047136,
      "step": 28525
    },
    {
      "epoch": 4.654213231095522,
      "grad_norm": 0.11449044942855835,
      "learning_rate": 9.477974432452738e-06,
      "loss": 0.1413,
      "num_input_tokens_seen": 54056720,
      "step": 28530
    },
    {
      "epoch": 4.655028958316339,
      "grad_norm": 0.1881919503211975,
      "learning_rate": 9.470279661992356e-06,
      "loss": 0.0662,
      "num_input_tokens_seen": 54066336,
      "step": 28535
    },
    {
      "epoch": 4.655844685537156,
      "grad_norm": 1.0341953039169312,
      "learning_rate": 9.462587286410021e-06,
      "loss": 0.1353,
      "num_input_tokens_seen": 54075744,
      "step": 28540
    },
    {
      "epoch": 4.656660412757974,
      "grad_norm": 0.0564492866396904,
      "learning_rate": 9.454897306891972e-06,
      "loss": 0.0809,
      "num_input_tokens_seen": 54085712,
      "step": 28545
    },
    {
      "epoch": 4.657476139978791,
      "grad_norm": 0.13780540227890015,
      "learning_rate": 9.44720972462411e-06,
      "loss": 0.1312,
      "num_input_tokens_seen": 54095680,
      "step": 28550
    },
    {
      "epoch": 4.658291867199608,
      "grad_norm": 0.9222856163978577,
      "learning_rate": 9.439524540791964e-06,
      "loss": 0.2331,
      "num_input_tokens_seen": 54105792,
      "step": 28555
    },
    {
      "epoch": 4.659107594420425,
      "grad_norm": 0.5887706279754639,
      "learning_rate": 9.431841756580673e-06,
      "loss": 0.1543,
      "num_input_tokens_seen": 54115152,
      "step": 28560
    },
    {
      "epoch": 4.6599233216412435,
      "grad_norm": 0.7409959435462952,
      "learning_rate": 9.42416137317503e-06,
      "loss": 0.1589,
      "num_input_tokens_seen": 54124464,
      "step": 28565
    },
    {
      "epoch": 4.660739048862061,
      "grad_norm": 0.6838893294334412,
      "learning_rate": 9.416483391759437e-06,
      "loss": 0.0431,
      "num_input_tokens_seen": 54134464,
      "step": 28570
    },
    {
      "epoch": 4.661554776082878,
      "grad_norm": 0.9175777435302734,
      "learning_rate": 9.408807813517945e-06,
      "loss": 0.1022,
      "num_input_tokens_seen": 54143696,
      "step": 28575
    },
    {
      "epoch": 4.662370503303695,
      "grad_norm": 0.2705284655094147,
      "learning_rate": 9.401134639634221e-06,
      "loss": 0.1166,
      "num_input_tokens_seen": 54152848,
      "step": 28580
    },
    {
      "epoch": 4.663186230524513,
      "grad_norm": 0.058112733066082,
      "learning_rate": 9.393463871291555e-06,
      "loss": 0.1578,
      "num_input_tokens_seen": 54162736,
      "step": 28585
    },
    {
      "epoch": 4.66400195774533,
      "grad_norm": 0.48952561616897583,
      "learning_rate": 9.385795509672881e-06,
      "loss": 0.0433,
      "num_input_tokens_seen": 54171920,
      "step": 28590
    },
    {
      "epoch": 4.664817684966147,
      "grad_norm": 0.10056444257497787,
      "learning_rate": 9.378129555960771e-06,
      "loss": 0.2927,
      "num_input_tokens_seen": 54181792,
      "step": 28595
    },
    {
      "epoch": 4.665633412186965,
      "grad_norm": 1.422928810119629,
      "learning_rate": 9.370466011337392e-06,
      "loss": 0.2305,
      "num_input_tokens_seen": 54191536,
      "step": 28600
    },
    {
      "epoch": 4.665633412186965,
      "eval_loss": 0.142927348613739,
      "eval_runtime": 133.2755,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 54191536,
      "step": 28600
    },
    {
      "epoch": 4.666449139407782,
      "grad_norm": 1.659963846206665,
      "learning_rate": 9.362804876984573e-06,
      "loss": 0.1188,
      "num_input_tokens_seen": 54202224,
      "step": 28605
    },
    {
      "epoch": 4.667264866628599,
      "grad_norm": 1.3365460634231567,
      "learning_rate": 9.355146154083747e-06,
      "loss": 0.1833,
      "num_input_tokens_seen": 54212240,
      "step": 28610
    },
    {
      "epoch": 4.6680805938494165,
      "grad_norm": 0.4957411289215088,
      "learning_rate": 9.347489843815987e-06,
      "loss": 0.2791,
      "num_input_tokens_seen": 54221840,
      "step": 28615
    },
    {
      "epoch": 4.668896321070234,
      "grad_norm": 0.7718403935432434,
      "learning_rate": 9.339835947362002e-06,
      "loss": 0.072,
      "num_input_tokens_seen": 54230864,
      "step": 28620
    },
    {
      "epoch": 4.669712048291052,
      "grad_norm": 0.31628963351249695,
      "learning_rate": 9.332184465902105e-06,
      "loss": 0.0534,
      "num_input_tokens_seen": 54240544,
      "step": 28625
    },
    {
      "epoch": 4.670527775511869,
      "grad_norm": 0.22118894755840302,
      "learning_rate": 9.324535400616266e-06,
      "loss": 0.0733,
      "num_input_tokens_seen": 54250560,
      "step": 28630
    },
    {
      "epoch": 4.671343502732686,
      "grad_norm": 0.2631967067718506,
      "learning_rate": 9.31688875268405e-06,
      "loss": 0.2015,
      "num_input_tokens_seen": 54260112,
      "step": 28635
    },
    {
      "epoch": 4.672159229953504,
      "grad_norm": 0.9296298027038574,
      "learning_rate": 9.309244523284674e-06,
      "loss": 0.1194,
      "num_input_tokens_seen": 54270400,
      "step": 28640
    },
    {
      "epoch": 4.672974957174321,
      "grad_norm": 0.03458147495985031,
      "learning_rate": 9.301602713596982e-06,
      "loss": 0.1794,
      "num_input_tokens_seen": 54279280,
      "step": 28645
    },
    {
      "epoch": 4.673790684395138,
      "grad_norm": 0.05739661678671837,
      "learning_rate": 9.293963324799432e-06,
      "loss": 0.0948,
      "num_input_tokens_seen": 54288624,
      "step": 28650
    },
    {
      "epoch": 4.674606411615955,
      "grad_norm": 1.274415135383606,
      "learning_rate": 9.286326358070104e-06,
      "loss": 0.1231,
      "num_input_tokens_seen": 54297936,
      "step": 28655
    },
    {
      "epoch": 4.6754221388367725,
      "grad_norm": 0.210955411195755,
      "learning_rate": 9.278691814586729e-06,
      "loss": 0.1468,
      "num_input_tokens_seen": 54307296,
      "step": 28660
    },
    {
      "epoch": 4.6762378660575905,
      "grad_norm": 0.9195089936256409,
      "learning_rate": 9.271059695526635e-06,
      "loss": 0.1481,
      "num_input_tokens_seen": 54316848,
      "step": 28665
    },
    {
      "epoch": 4.677053593278408,
      "grad_norm": 1.1081868410110474,
      "learning_rate": 9.263430002066805e-06,
      "loss": 0.2099,
      "num_input_tokens_seen": 54325696,
      "step": 28670
    },
    {
      "epoch": 4.677869320499225,
      "grad_norm": 1.0612772703170776,
      "learning_rate": 9.25580273538382e-06,
      "loss": 0.0785,
      "num_input_tokens_seen": 54334992,
      "step": 28675
    },
    {
      "epoch": 4.678685047720043,
      "grad_norm": 0.19068272411823273,
      "learning_rate": 9.248177896653907e-06,
      "loss": 0.0378,
      "num_input_tokens_seen": 54344256,
      "step": 28680
    },
    {
      "epoch": 4.67950077494086,
      "grad_norm": 1.481374979019165,
      "learning_rate": 9.240555487052918e-06,
      "loss": 0.3372,
      "num_input_tokens_seen": 54353200,
      "step": 28685
    },
    {
      "epoch": 4.680316502161677,
      "grad_norm": 1.4918129444122314,
      "learning_rate": 9.232935507756313e-06,
      "loss": 0.2154,
      "num_input_tokens_seen": 54362944,
      "step": 28690
    },
    {
      "epoch": 4.681132229382494,
      "grad_norm": 0.5303537249565125,
      "learning_rate": 9.225317959939193e-06,
      "loss": 0.1068,
      "num_input_tokens_seen": 54372208,
      "step": 28695
    },
    {
      "epoch": 4.681947956603312,
      "grad_norm": 0.3202377259731293,
      "learning_rate": 9.217702844776287e-06,
      "loss": 0.1729,
      "num_input_tokens_seen": 54380592,
      "step": 28700
    },
    {
      "epoch": 4.682763683824129,
      "grad_norm": 1.000361680984497,
      "learning_rate": 9.210090163441929e-06,
      "loss": 0.3104,
      "num_input_tokens_seen": 54388656,
      "step": 28705
    },
    {
      "epoch": 4.6835794110449465,
      "grad_norm": 0.12165910005569458,
      "learning_rate": 9.202479917110105e-06,
      "loss": 0.0705,
      "num_input_tokens_seen": 54398256,
      "step": 28710
    },
    {
      "epoch": 4.684395138265764,
      "grad_norm": 0.21653637290000916,
      "learning_rate": 9.194872106954392e-06,
      "loss": 0.041,
      "num_input_tokens_seen": 54408608,
      "step": 28715
    },
    {
      "epoch": 4.685210865486582,
      "grad_norm": 0.7315146327018738,
      "learning_rate": 9.187266734148029e-06,
      "loss": 0.0605,
      "num_input_tokens_seen": 54418096,
      "step": 28720
    },
    {
      "epoch": 4.686026592707399,
      "grad_norm": 1.7525324821472168,
      "learning_rate": 9.179663799863849e-06,
      "loss": 0.1895,
      "num_input_tokens_seen": 54427520,
      "step": 28725
    },
    {
      "epoch": 4.686842319928216,
      "grad_norm": 0.5068126916885376,
      "learning_rate": 9.172063305274317e-06,
      "loss": 0.1257,
      "num_input_tokens_seen": 54437776,
      "step": 28730
    },
    {
      "epoch": 4.687658047149033,
      "grad_norm": 1.187021255493164,
      "learning_rate": 9.164465251551527e-06,
      "loss": 0.1917,
      "num_input_tokens_seen": 54447408,
      "step": 28735
    },
    {
      "epoch": 4.688473774369851,
      "grad_norm": 0.22371231019496918,
      "learning_rate": 9.156869639867205e-06,
      "loss": 0.0744,
      "num_input_tokens_seen": 54455824,
      "step": 28740
    },
    {
      "epoch": 4.689289501590668,
      "grad_norm": 0.07082287967205048,
      "learning_rate": 9.149276471392677e-06,
      "loss": 0.0212,
      "num_input_tokens_seen": 54465040,
      "step": 28745
    },
    {
      "epoch": 4.690105228811485,
      "grad_norm": 0.830719530582428,
      "learning_rate": 9.141685747298914e-06,
      "loss": 0.1446,
      "num_input_tokens_seen": 54475056,
      "step": 28750
    },
    {
      "epoch": 4.690920956032302,
      "grad_norm": 0.376581609249115,
      "learning_rate": 9.13409746875649e-06,
      "loss": 0.0731,
      "num_input_tokens_seen": 54485072,
      "step": 28755
    },
    {
      "epoch": 4.69173668325312,
      "grad_norm": 0.3026118278503418,
      "learning_rate": 9.12651163693562e-06,
      "loss": 0.2411,
      "num_input_tokens_seen": 54494432,
      "step": 28760
    },
    {
      "epoch": 4.692552410473938,
      "grad_norm": 0.7971454858779907,
      "learning_rate": 9.11892825300614e-06,
      "loss": 0.0612,
      "num_input_tokens_seen": 54503792,
      "step": 28765
    },
    {
      "epoch": 4.693368137694755,
      "grad_norm": 0.04323820769786835,
      "learning_rate": 9.111347318137491e-06,
      "loss": 0.0875,
      "num_input_tokens_seen": 54513680,
      "step": 28770
    },
    {
      "epoch": 4.694183864915573,
      "grad_norm": 1.2249659299850464,
      "learning_rate": 9.103768833498755e-06,
      "loss": 0.2222,
      "num_input_tokens_seen": 54524144,
      "step": 28775
    },
    {
      "epoch": 4.69499959213639,
      "grad_norm": 0.4580308496952057,
      "learning_rate": 9.096192800258639e-06,
      "loss": 0.0533,
      "num_input_tokens_seen": 54533440,
      "step": 28780
    },
    {
      "epoch": 4.695815319357207,
      "grad_norm": 0.34622323513031006,
      "learning_rate": 9.088619219585443e-06,
      "loss": 0.0638,
      "num_input_tokens_seen": 54543472,
      "step": 28785
    },
    {
      "epoch": 4.696631046578024,
      "grad_norm": 0.1488943248987198,
      "learning_rate": 9.081048092647127e-06,
      "loss": 0.1865,
      "num_input_tokens_seen": 54553392,
      "step": 28790
    },
    {
      "epoch": 4.697446773798841,
      "grad_norm": 0.41459232568740845,
      "learning_rate": 9.073479420611245e-06,
      "loss": 0.0551,
      "num_input_tokens_seen": 54562336,
      "step": 28795
    },
    {
      "epoch": 4.698262501019659,
      "grad_norm": 0.945576012134552,
      "learning_rate": 9.065913204644974e-06,
      "loss": 0.1353,
      "num_input_tokens_seen": 54572176,
      "step": 28800
    },
    {
      "epoch": 4.698262501019659,
      "eval_loss": 0.14151756465435028,
      "eval_runtime": 133.2446,
      "eval_samples_per_second": 20.451,
      "eval_steps_per_second": 10.229,
      "num_input_tokens_seen": 54572176,
      "step": 28800
    },
    {
      "epoch": 4.699078228240476,
      "grad_norm": 0.5766950845718384,
      "learning_rate": 9.058349445915135e-06,
      "loss": 0.1191,
      "num_input_tokens_seen": 54581232,
      "step": 28805
    },
    {
      "epoch": 4.6998939554612935,
      "grad_norm": 0.17359964549541473,
      "learning_rate": 9.050788145588138e-06,
      "loss": 0.0881,
      "num_input_tokens_seen": 54591328,
      "step": 28810
    },
    {
      "epoch": 4.7007096826821115,
      "grad_norm": 0.0907498151063919,
      "learning_rate": 9.043229304830039e-06,
      "loss": 0.0389,
      "num_input_tokens_seen": 54601072,
      "step": 28815
    },
    {
      "epoch": 4.701525409902929,
      "grad_norm": 0.2527492940425873,
      "learning_rate": 9.035672924806515e-06,
      "loss": 0.1962,
      "num_input_tokens_seen": 54610768,
      "step": 28820
    },
    {
      "epoch": 4.702341137123746,
      "grad_norm": 0.1291455328464508,
      "learning_rate": 9.028119006682839e-06,
      "loss": 0.0741,
      "num_input_tokens_seen": 54619712,
      "step": 28825
    },
    {
      "epoch": 4.703156864344563,
      "grad_norm": 1.0031309127807617,
      "learning_rate": 9.020567551623935e-06,
      "loss": 0.1343,
      "num_input_tokens_seen": 54629872,
      "step": 28830
    },
    {
      "epoch": 4.70397259156538,
      "grad_norm": 0.033808156847953796,
      "learning_rate": 9.013018560794318e-06,
      "loss": 0.2253,
      "num_input_tokens_seen": 54639152,
      "step": 28835
    },
    {
      "epoch": 4.704788318786198,
      "grad_norm": 1.2433655261993408,
      "learning_rate": 9.005472035358139e-06,
      "loss": 0.2191,
      "num_input_tokens_seen": 54648384,
      "step": 28840
    },
    {
      "epoch": 4.705604046007015,
      "grad_norm": 0.4621669054031372,
      "learning_rate": 8.997927976479185e-06,
      "loss": 0.0898,
      "num_input_tokens_seen": 54658512,
      "step": 28845
    },
    {
      "epoch": 4.706419773227832,
      "grad_norm": 0.11144815385341644,
      "learning_rate": 8.99038638532082e-06,
      "loss": 0.0804,
      "num_input_tokens_seen": 54667888,
      "step": 28850
    },
    {
      "epoch": 4.70723550044865,
      "grad_norm": 0.13262440264225006,
      "learning_rate": 8.982847263046065e-06,
      "loss": 0.0391,
      "num_input_tokens_seen": 54677952,
      "step": 28855
    },
    {
      "epoch": 4.7080512276694675,
      "grad_norm": 0.6966052651405334,
      "learning_rate": 8.975310610817555e-06,
      "loss": 0.2296,
      "num_input_tokens_seen": 54687072,
      "step": 28860
    },
    {
      "epoch": 4.708866954890285,
      "grad_norm": 1.3866580724716187,
      "learning_rate": 8.967776429797528e-06,
      "loss": 0.1809,
      "num_input_tokens_seen": 54696640,
      "step": 28865
    },
    {
      "epoch": 4.709682682111102,
      "grad_norm": 1.7894779443740845,
      "learning_rate": 8.960244721147842e-06,
      "loss": 0.1619,
      "num_input_tokens_seen": 54707600,
      "step": 28870
    },
    {
      "epoch": 4.71049840933192,
      "grad_norm": 0.1127617135643959,
      "learning_rate": 8.952715486029995e-06,
      "loss": 0.0576,
      "num_input_tokens_seen": 54716912,
      "step": 28875
    },
    {
      "epoch": 4.711314136552737,
      "grad_norm": 0.2789481580257416,
      "learning_rate": 8.945188725605075e-06,
      "loss": 0.1073,
      "num_input_tokens_seen": 54727248,
      "step": 28880
    },
    {
      "epoch": 4.712129863773554,
      "grad_norm": 0.04966938868165016,
      "learning_rate": 8.937664441033817e-06,
      "loss": 0.1396,
      "num_input_tokens_seen": 54735776,
      "step": 28885
    },
    {
      "epoch": 4.712945590994371,
      "grad_norm": 0.854922354221344,
      "learning_rate": 8.930142633476549e-06,
      "loss": 0.0664,
      "num_input_tokens_seen": 54744272,
      "step": 28890
    },
    {
      "epoch": 4.713761318215189,
      "grad_norm": 1.1477665901184082,
      "learning_rate": 8.92262330409323e-06,
      "loss": 0.2018,
      "num_input_tokens_seen": 54753312,
      "step": 28895
    },
    {
      "epoch": 4.714577045436006,
      "grad_norm": 0.03796755522489548,
      "learning_rate": 8.915106454043448e-06,
      "loss": 0.0958,
      "num_input_tokens_seen": 54762576,
      "step": 28900
    },
    {
      "epoch": 4.715392772656823,
      "grad_norm": 1.703497290611267,
      "learning_rate": 8.90759208448638e-06,
      "loss": 0.0771,
      "num_input_tokens_seen": 54772064,
      "step": 28905
    },
    {
      "epoch": 4.716208499877641,
      "grad_norm": 0.5539412498474121,
      "learning_rate": 8.900080196580848e-06,
      "loss": 0.2028,
      "num_input_tokens_seen": 54781216,
      "step": 28910
    },
    {
      "epoch": 4.717024227098459,
      "grad_norm": 0.05765211582183838,
      "learning_rate": 8.892570791485267e-06,
      "loss": 0.1358,
      "num_input_tokens_seen": 54792016,
      "step": 28915
    },
    {
      "epoch": 4.717839954319276,
      "grad_norm": 0.955900251865387,
      "learning_rate": 8.885063870357688e-06,
      "loss": 0.2388,
      "num_input_tokens_seen": 54801008,
      "step": 28920
    },
    {
      "epoch": 4.718655681540093,
      "grad_norm": 0.37568095326423645,
      "learning_rate": 8.87755943435578e-06,
      "loss": 0.1899,
      "num_input_tokens_seen": 54809728,
      "step": 28925
    },
    {
      "epoch": 4.71947140876091,
      "grad_norm": 0.20827512443065643,
      "learning_rate": 8.87005748463681e-06,
      "loss": 0.0431,
      "num_input_tokens_seen": 54819312,
      "step": 28930
    },
    {
      "epoch": 4.720287135981728,
      "grad_norm": 0.2542904019355774,
      "learning_rate": 8.862558022357681e-06,
      "loss": 0.1582,
      "num_input_tokens_seen": 54827232,
      "step": 28935
    },
    {
      "epoch": 4.721102863202545,
      "grad_norm": 0.45325639843940735,
      "learning_rate": 8.855061048674903e-06,
      "loss": 0.1259,
      "num_input_tokens_seen": 54836560,
      "step": 28940
    },
    {
      "epoch": 4.721918590423362,
      "grad_norm": 2.0999953746795654,
      "learning_rate": 8.847566564744595e-06,
      "loss": 0.2773,
      "num_input_tokens_seen": 54845344,
      "step": 28945
    },
    {
      "epoch": 4.72273431764418,
      "grad_norm": 0.054483357816934586,
      "learning_rate": 8.840074571722512e-06,
      "loss": 0.1238,
      "num_input_tokens_seen": 54854944,
      "step": 28950
    },
    {
      "epoch": 4.723550044864997,
      "grad_norm": 0.5989662408828735,
      "learning_rate": 8.832585070764002e-06,
      "loss": 0.0617,
      "num_input_tokens_seen": 54865600,
      "step": 28955
    },
    {
      "epoch": 4.7243657720858145,
      "grad_norm": 0.039913106709718704,
      "learning_rate": 8.825098063024045e-06,
      "loss": 0.0812,
      "num_input_tokens_seen": 54875200,
      "step": 28960
    },
    {
      "epoch": 4.725181499306632,
      "grad_norm": 0.2705205976963043,
      "learning_rate": 8.817613549657244e-06,
      "loss": 0.1391,
      "num_input_tokens_seen": 54884816,
      "step": 28965
    },
    {
      "epoch": 4.725997226527449,
      "grad_norm": 0.6146743893623352,
      "learning_rate": 8.810131531817783e-06,
      "loss": 0.2658,
      "num_input_tokens_seen": 54895152,
      "step": 28970
    },
    {
      "epoch": 4.726812953748267,
      "grad_norm": 0.6678867936134338,
      "learning_rate": 8.802652010659496e-06,
      "loss": 0.068,
      "num_input_tokens_seen": 54904608,
      "step": 28975
    },
    {
      "epoch": 4.727628680969084,
      "grad_norm": 0.7878478765487671,
      "learning_rate": 8.795174987335827e-06,
      "loss": 0.0947,
      "num_input_tokens_seen": 54913472,
      "step": 28980
    },
    {
      "epoch": 4.728444408189901,
      "grad_norm": 0.6295450329780579,
      "learning_rate": 8.787700462999807e-06,
      "loss": 0.0657,
      "num_input_tokens_seen": 54923488,
      "step": 28985
    },
    {
      "epoch": 4.729260135410719,
      "grad_norm": 0.14246848225593567,
      "learning_rate": 8.780228438804122e-06,
      "loss": 0.103,
      "num_input_tokens_seen": 54933088,
      "step": 28990
    },
    {
      "epoch": 4.730075862631536,
      "grad_norm": 0.35475918650627136,
      "learning_rate": 8.772758915901032e-06,
      "loss": 0.2549,
      "num_input_tokens_seen": 54942704,
      "step": 28995
    },
    {
      "epoch": 4.730891589852353,
      "grad_norm": 0.030219823122024536,
      "learning_rate": 8.765291895442443e-06,
      "loss": 0.0633,
      "num_input_tokens_seen": 54952896,
      "step": 29000
    },
    {
      "epoch": 4.730891589852353,
      "eval_loss": 0.14134128391742706,
      "eval_runtime": 133.3183,
      "eval_samples_per_second": 20.44,
      "eval_steps_per_second": 10.224,
      "num_input_tokens_seen": 54952896,
      "step": 29000
    },
    {
      "epoch": 4.7317073170731705,
      "grad_norm": 0.3738565444946289,
      "learning_rate": 8.75782737857987e-06,
      "loss": 0.1296,
      "num_input_tokens_seen": 54962928,
      "step": 29005
    },
    {
      "epoch": 4.732523044293988,
      "grad_norm": 0.4959174692630768,
      "learning_rate": 8.750365366464425e-06,
      "loss": 0.0897,
      "num_input_tokens_seen": 54972384,
      "step": 29010
    },
    {
      "epoch": 4.733338771514806,
      "grad_norm": 0.46501490473747253,
      "learning_rate": 8.742905860246838e-06,
      "loss": 0.1223,
      "num_input_tokens_seen": 54980384,
      "step": 29015
    },
    {
      "epoch": 4.734154498735623,
      "grad_norm": 1.4514001607894897,
      "learning_rate": 8.735448861077478e-06,
      "loss": 0.3541,
      "num_input_tokens_seen": 54989936,
      "step": 29020
    },
    {
      "epoch": 4.73497022595644,
      "grad_norm": 0.3919871747493744,
      "learning_rate": 8.727994370106288e-06,
      "loss": 0.1557,
      "num_input_tokens_seen": 54999024,
      "step": 29025
    },
    {
      "epoch": 4.735785953177258,
      "grad_norm": 0.12626419961452484,
      "learning_rate": 8.720542388482861e-06,
      "loss": 0.0419,
      "num_input_tokens_seen": 55007968,
      "step": 29030
    },
    {
      "epoch": 4.736601680398075,
      "grad_norm": 0.4181213676929474,
      "learning_rate": 8.71309291735637e-06,
      "loss": 0.2382,
      "num_input_tokens_seen": 55016432,
      "step": 29035
    },
    {
      "epoch": 4.737417407618892,
      "grad_norm": 0.6341415047645569,
      "learning_rate": 8.705645957875621e-06,
      "loss": 0.0885,
      "num_input_tokens_seen": 55026272,
      "step": 29040
    },
    {
      "epoch": 4.738233134839709,
      "grad_norm": 0.12744447588920593,
      "learning_rate": 8.698201511189048e-06,
      "loss": 0.1825,
      "num_input_tokens_seen": 55036160,
      "step": 29045
    },
    {
      "epoch": 4.739048862060527,
      "grad_norm": 0.19896970689296722,
      "learning_rate": 8.690759578444649e-06,
      "loss": 0.1956,
      "num_input_tokens_seen": 55044096,
      "step": 29050
    },
    {
      "epoch": 4.7398645892813445,
      "grad_norm": 1.9950718879699707,
      "learning_rate": 8.68332016079008e-06,
      "loss": 0.1185,
      "num_input_tokens_seen": 55053104,
      "step": 29055
    },
    {
      "epoch": 4.740680316502162,
      "grad_norm": 0.04444211721420288,
      "learning_rate": 8.6758832593726e-06,
      "loss": 0.1557,
      "num_input_tokens_seen": 55063776,
      "step": 29060
    },
    {
      "epoch": 4.741496043722979,
      "grad_norm": 1.213789701461792,
      "learning_rate": 8.668448875339053e-06,
      "loss": 0.1,
      "num_input_tokens_seen": 55072400,
      "step": 29065
    },
    {
      "epoch": 4.742311770943797,
      "grad_norm": 0.420091837644577,
      "learning_rate": 8.661017009835933e-06,
      "loss": 0.2053,
      "num_input_tokens_seen": 55081840,
      "step": 29070
    },
    {
      "epoch": 4.743127498164614,
      "grad_norm": 0.5620577335357666,
      "learning_rate": 8.653587664009311e-06,
      "loss": 0.1116,
      "num_input_tokens_seen": 55090048,
      "step": 29075
    },
    {
      "epoch": 4.743943225385431,
      "grad_norm": 0.6324814558029175,
      "learning_rate": 8.646160839004902e-06,
      "loss": 0.1793,
      "num_input_tokens_seen": 55099584,
      "step": 29080
    },
    {
      "epoch": 4.744758952606248,
      "grad_norm": 0.1461186408996582,
      "learning_rate": 8.638736535967998e-06,
      "loss": 0.0444,
      "num_input_tokens_seen": 55109472,
      "step": 29085
    },
    {
      "epoch": 4.745574679827066,
      "grad_norm": 0.11170531064271927,
      "learning_rate": 8.631314756043535e-06,
      "loss": 0.2441,
      "num_input_tokens_seen": 55119584,
      "step": 29090
    },
    {
      "epoch": 4.746390407047883,
      "grad_norm": 0.11395153403282166,
      "learning_rate": 8.62389550037603e-06,
      "loss": 0.129,
      "num_input_tokens_seen": 55128208,
      "step": 29095
    },
    {
      "epoch": 4.7472061342687,
      "grad_norm": 0.062324799597263336,
      "learning_rate": 8.616478770109646e-06,
      "loss": 0.0852,
      "num_input_tokens_seen": 55137200,
      "step": 29100
    },
    {
      "epoch": 4.7480218614895175,
      "grad_norm": 0.4200354218482971,
      "learning_rate": 8.609064566388111e-06,
      "loss": 0.2576,
      "num_input_tokens_seen": 55147552,
      "step": 29105
    },
    {
      "epoch": 4.748837588710336,
      "grad_norm": 0.4368310868740082,
      "learning_rate": 8.601652890354815e-06,
      "loss": 0.1215,
      "num_input_tokens_seen": 55156720,
      "step": 29110
    },
    {
      "epoch": 4.749653315931153,
      "grad_norm": 0.23458699882030487,
      "learning_rate": 8.594243743152705e-06,
      "loss": 0.0964,
      "num_input_tokens_seen": 55165872,
      "step": 29115
    },
    {
      "epoch": 4.75046904315197,
      "grad_norm": 0.4416273236274719,
      "learning_rate": 8.58683712592438e-06,
      "loss": 0.1241,
      "num_input_tokens_seen": 55176288,
      "step": 29120
    },
    {
      "epoch": 4.751284770372787,
      "grad_norm": 0.3450174033641815,
      "learning_rate": 8.579433039812037e-06,
      "loss": 0.2031,
      "num_input_tokens_seen": 55186096,
      "step": 29125
    },
    {
      "epoch": 4.752100497593605,
      "grad_norm": 0.7658773064613342,
      "learning_rate": 8.572031485957466e-06,
      "loss": 0.0628,
      "num_input_tokens_seen": 55194896,
      "step": 29130
    },
    {
      "epoch": 4.752916224814422,
      "grad_norm": 0.8824108242988586,
      "learning_rate": 8.564632465502084e-06,
      "loss": 0.0909,
      "num_input_tokens_seen": 55204272,
      "step": 29135
    },
    {
      "epoch": 4.753731952035239,
      "grad_norm": 1.3639782667160034,
      "learning_rate": 8.557235979586928e-06,
      "loss": 0.1252,
      "num_input_tokens_seen": 55213792,
      "step": 29140
    },
    {
      "epoch": 4.754547679256056,
      "grad_norm": 0.2671785056591034,
      "learning_rate": 8.549842029352606e-06,
      "loss": 0.1293,
      "num_input_tokens_seen": 55224560,
      "step": 29145
    },
    {
      "epoch": 4.755363406476874,
      "grad_norm": 0.2195054292678833,
      "learning_rate": 8.542450615939376e-06,
      "loss": 0.0507,
      "num_input_tokens_seen": 55235328,
      "step": 29150
    },
    {
      "epoch": 4.7561791336976915,
      "grad_norm": 1.8105216026306152,
      "learning_rate": 8.535061740487082e-06,
      "loss": 0.2619,
      "num_input_tokens_seen": 55244640,
      "step": 29155
    },
    {
      "epoch": 4.756994860918509,
      "grad_norm": 0.9732959866523743,
      "learning_rate": 8.527675404135168e-06,
      "loss": 0.1084,
      "num_input_tokens_seen": 55253904,
      "step": 29160
    },
    {
      "epoch": 4.757810588139327,
      "grad_norm": 1.1047433614730835,
      "learning_rate": 8.520291608022724e-06,
      "loss": 0.0834,
      "num_input_tokens_seen": 55261872,
      "step": 29165
    },
    {
      "epoch": 4.758626315360144,
      "grad_norm": 0.0524229072034359,
      "learning_rate": 8.512910353288398e-06,
      "loss": 0.1176,
      "num_input_tokens_seen": 55271488,
      "step": 29170
    },
    {
      "epoch": 4.759442042580961,
      "grad_norm": 0.17306751012802124,
      "learning_rate": 8.505531641070486e-06,
      "loss": 0.3074,
      "num_input_tokens_seen": 55281808,
      "step": 29175
    },
    {
      "epoch": 4.760257769801778,
      "grad_norm": 0.3370472192764282,
      "learning_rate": 8.498155472506885e-06,
      "loss": 0.0564,
      "num_input_tokens_seen": 55290896,
      "step": 29180
    },
    {
      "epoch": 4.761073497022595,
      "grad_norm": 1.7460099458694458,
      "learning_rate": 8.49078184873508e-06,
      "loss": 0.1947,
      "num_input_tokens_seen": 55300544,
      "step": 29185
    },
    {
      "epoch": 4.761889224243413,
      "grad_norm": 0.7157062888145447,
      "learning_rate": 8.483410770892188e-06,
      "loss": 0.127,
      "num_input_tokens_seen": 55309264,
      "step": 29190
    },
    {
      "epoch": 4.76270495146423,
      "grad_norm": 0.3031463623046875,
      "learning_rate": 8.476042240114909e-06,
      "loss": 0.1615,
      "num_input_tokens_seen": 55318976,
      "step": 29195
    },
    {
      "epoch": 4.7635206786850475,
      "grad_norm": 2.487164258956909,
      "learning_rate": 8.468676257539568e-06,
      "loss": 0.1857,
      "num_input_tokens_seen": 55327776,
      "step": 29200
    },
    {
      "epoch": 4.7635206786850475,
      "eval_loss": 0.14239813387393951,
      "eval_runtime": 133.35,
      "eval_samples_per_second": 20.435,
      "eval_steps_per_second": 10.221,
      "num_input_tokens_seen": 55327776,
      "step": 29200
    },
    {
      "epoch": 4.7643364059058655,
      "grad_norm": 0.04485904797911644,
      "learning_rate": 8.4613128243021e-06,
      "loss": 0.1029,
      "num_input_tokens_seen": 55337408,
      "step": 29205
    },
    {
      "epoch": 4.765152133126683,
      "grad_norm": 0.3889210522174835,
      "learning_rate": 8.453951941538028e-06,
      "loss": 0.1549,
      "num_input_tokens_seen": 55346176,
      "step": 29210
    },
    {
      "epoch": 4.7659678603475,
      "grad_norm": 0.969377338886261,
      "learning_rate": 8.446593610382495e-06,
      "loss": 0.2549,
      "num_input_tokens_seen": 55355168,
      "step": 29215
    },
    {
      "epoch": 4.766783587568317,
      "grad_norm": 1.1800689697265625,
      "learning_rate": 8.439237831970259e-06,
      "loss": 0.213,
      "num_input_tokens_seen": 55364736,
      "step": 29220
    },
    {
      "epoch": 4.767599314789135,
      "grad_norm": 0.1383642554283142,
      "learning_rate": 8.431884607435667e-06,
      "loss": 0.2802,
      "num_input_tokens_seen": 55374032,
      "step": 29225
    },
    {
      "epoch": 4.768415042009952,
      "grad_norm": 0.06888241320848465,
      "learning_rate": 8.424533937912665e-06,
      "loss": 0.0419,
      "num_input_tokens_seen": 55383072,
      "step": 29230
    },
    {
      "epoch": 4.769230769230769,
      "grad_norm": 0.30997711420059204,
      "learning_rate": 8.41718582453484e-06,
      "loss": 0.118,
      "num_input_tokens_seen": 55393520,
      "step": 29235
    },
    {
      "epoch": 4.770046496451586,
      "grad_norm": 0.03712134435772896,
      "learning_rate": 8.409840268435346e-06,
      "loss": 0.0764,
      "num_input_tokens_seen": 55402368,
      "step": 29240
    },
    {
      "epoch": 4.770862223672404,
      "grad_norm": 0.7326871156692505,
      "learning_rate": 8.402497270746976e-06,
      "loss": 0.2688,
      "num_input_tokens_seen": 55411776,
      "step": 29245
    },
    {
      "epoch": 4.771677950893221,
      "grad_norm": 0.4746428430080414,
      "learning_rate": 8.395156832602095e-06,
      "loss": 0.0186,
      "num_input_tokens_seen": 55420416,
      "step": 29250
    },
    {
      "epoch": 4.772493678114039,
      "grad_norm": 0.110906682908535,
      "learning_rate": 8.387818955132707e-06,
      "loss": 0.1505,
      "num_input_tokens_seen": 55429888,
      "step": 29255
    },
    {
      "epoch": 4.773309405334856,
      "grad_norm": 0.6929786801338196,
      "learning_rate": 8.38048363947039e-06,
      "loss": 0.0852,
      "num_input_tokens_seen": 55438480,
      "step": 29260
    },
    {
      "epoch": 4.774125132555674,
      "grad_norm": 0.29094091057777405,
      "learning_rate": 8.373150886746351e-06,
      "loss": 0.0868,
      "num_input_tokens_seen": 55447312,
      "step": 29265
    },
    {
      "epoch": 4.774940859776491,
      "grad_norm": 0.4808131456375122,
      "learning_rate": 8.365820698091397e-06,
      "loss": 0.1088,
      "num_input_tokens_seen": 55458128,
      "step": 29270
    },
    {
      "epoch": 4.775756586997308,
      "grad_norm": 0.9084393978118896,
      "learning_rate": 8.358493074635922e-06,
      "loss": 0.1142,
      "num_input_tokens_seen": 55467952,
      "step": 29275
    },
    {
      "epoch": 4.776572314218125,
      "grad_norm": 0.03830799460411072,
      "learning_rate": 8.351168017509948e-06,
      "loss": 0.0258,
      "num_input_tokens_seen": 55478144,
      "step": 29280
    },
    {
      "epoch": 4.777388041438943,
      "grad_norm": 0.8926367163658142,
      "learning_rate": 8.343845527843094e-06,
      "loss": 0.0964,
      "num_input_tokens_seen": 55487680,
      "step": 29285
    },
    {
      "epoch": 4.77820376865976,
      "grad_norm": 0.2811925709247589,
      "learning_rate": 8.336525606764566e-06,
      "loss": 0.0397,
      "num_input_tokens_seen": 55496480,
      "step": 29290
    },
    {
      "epoch": 4.779019495880577,
      "grad_norm": 1.0264506340026855,
      "learning_rate": 8.329208255403204e-06,
      "loss": 0.1491,
      "num_input_tokens_seen": 55505888,
      "step": 29295
    },
    {
      "epoch": 4.7798352231013945,
      "grad_norm": 0.5857142210006714,
      "learning_rate": 8.321893474887426e-06,
      "loss": 0.1153,
      "num_input_tokens_seen": 55514928,
      "step": 29300
    },
    {
      "epoch": 4.7806509503222125,
      "grad_norm": 0.6181142926216125,
      "learning_rate": 8.31458126634526e-06,
      "loss": 0.1397,
      "num_input_tokens_seen": 55524016,
      "step": 29305
    },
    {
      "epoch": 4.78146667754303,
      "grad_norm": 0.2084045559167862,
      "learning_rate": 8.30727163090435e-06,
      "loss": 0.1811,
      "num_input_tokens_seen": 55533920,
      "step": 29310
    },
    {
      "epoch": 4.782282404763847,
      "grad_norm": 0.6130717992782593,
      "learning_rate": 8.29996456969192e-06,
      "loss": 0.1244,
      "num_input_tokens_seen": 55543984,
      "step": 29315
    },
    {
      "epoch": 4.783098131984664,
      "grad_norm": 1.6086628437042236,
      "learning_rate": 8.292660083834818e-06,
      "loss": 0.1762,
      "num_input_tokens_seen": 55552384,
      "step": 29320
    },
    {
      "epoch": 4.783913859205482,
      "grad_norm": 0.6626279354095459,
      "learning_rate": 8.2853581744595e-06,
      "loss": 0.132,
      "num_input_tokens_seen": 55560768,
      "step": 29325
    },
    {
      "epoch": 4.784729586426299,
      "grad_norm": 0.29669734835624695,
      "learning_rate": 8.278058842691991e-06,
      "loss": 0.1281,
      "num_input_tokens_seen": 55571168,
      "step": 29330
    },
    {
      "epoch": 4.785545313647116,
      "grad_norm": 0.37230193614959717,
      "learning_rate": 8.27076208965796e-06,
      "loss": 0.1309,
      "num_input_tokens_seen": 55580384,
      "step": 29335
    },
    {
      "epoch": 4.786361040867934,
      "grad_norm": 0.43353453278541565,
      "learning_rate": 8.263467916482637e-06,
      "loss": 0.3136,
      "num_input_tokens_seen": 55590976,
      "step": 29340
    },
    {
      "epoch": 4.787176768088751,
      "grad_norm": 0.15542317926883698,
      "learning_rate": 8.256176324290885e-06,
      "loss": 0.1146,
      "num_input_tokens_seen": 55601760,
      "step": 29345
    },
    {
      "epoch": 4.7879924953095685,
      "grad_norm": 0.09644786268472672,
      "learning_rate": 8.248887314207168e-06,
      "loss": 0.0278,
      "num_input_tokens_seen": 55612448,
      "step": 29350
    },
    {
      "epoch": 4.788808222530386,
      "grad_norm": 1.1230190992355347,
      "learning_rate": 8.24160088735553e-06,
      "loss": 0.1439,
      "num_input_tokens_seen": 55622080,
      "step": 29355
    },
    {
      "epoch": 4.789623949751203,
      "grad_norm": 0.3401656448841095,
      "learning_rate": 8.234317044859629e-06,
      "loss": 0.0949,
      "num_input_tokens_seen": 55632096,
      "step": 29360
    },
    {
      "epoch": 4.790439676972021,
      "grad_norm": 0.9920913577079773,
      "learning_rate": 8.227035787842744e-06,
      "loss": 0.1229,
      "num_input_tokens_seen": 55641024,
      "step": 29365
    },
    {
      "epoch": 4.791255404192838,
      "grad_norm": 3.769155740737915,
      "learning_rate": 8.219757117427721e-06,
      "loss": 0.2821,
      "num_input_tokens_seen": 55651296,
      "step": 29370
    },
    {
      "epoch": 4.792071131413655,
      "grad_norm": 0.26400789618492126,
      "learning_rate": 8.212481034737014e-06,
      "loss": 0.1112,
      "num_input_tokens_seen": 55660240,
      "step": 29375
    },
    {
      "epoch": 4.792886858634473,
      "grad_norm": 1.4332104921340942,
      "learning_rate": 8.205207540892707e-06,
      "loss": 0.1434,
      "num_input_tokens_seen": 55669904,
      "step": 29380
    },
    {
      "epoch": 4.79370258585529,
      "grad_norm": 0.20981061458587646,
      "learning_rate": 8.197936637016442e-06,
      "loss": 0.1357,
      "num_input_tokens_seen": 55679584,
      "step": 29385
    },
    {
      "epoch": 4.794518313076107,
      "grad_norm": 0.1745203286409378,
      "learning_rate": 8.190668324229508e-06,
      "loss": 0.0297,
      "num_input_tokens_seen": 55689552,
      "step": 29390
    },
    {
      "epoch": 4.7953340402969244,
      "grad_norm": 0.36337846517562866,
      "learning_rate": 8.183402603652749e-06,
      "loss": 0.0397,
      "num_input_tokens_seen": 55700864,
      "step": 29395
    },
    {
      "epoch": 4.796149767517742,
      "grad_norm": 0.2898939251899719,
      "learning_rate": 8.176139476406635e-06,
      "loss": 0.0332,
      "num_input_tokens_seen": 55708896,
      "step": 29400
    },
    {
      "epoch": 4.796149767517742,
      "eval_loss": 0.14119859039783478,
      "eval_runtime": 133.3253,
      "eval_samples_per_second": 20.439,
      "eval_steps_per_second": 10.223,
      "num_input_tokens_seen": 55708896,
      "step": 29400
    },
    {
      "epoch": 4.79696549473856,
      "grad_norm": 3.55717134475708,
      "learning_rate": 8.16887894361125e-06,
      "loss": 0.1308,
      "num_input_tokens_seen": 55717488,
      "step": 29405
    },
    {
      "epoch": 4.797781221959377,
      "grad_norm": 0.8479077816009521,
      "learning_rate": 8.161621006386233e-06,
      "loss": 0.1415,
      "num_input_tokens_seen": 55726992,
      "step": 29410
    },
    {
      "epoch": 4.798596949180194,
      "grad_norm": 0.290691614151001,
      "learning_rate": 8.154365665850869e-06,
      "loss": 0.0621,
      "num_input_tokens_seen": 55735056,
      "step": 29415
    },
    {
      "epoch": 4.799412676401012,
      "grad_norm": 0.3151019513607025,
      "learning_rate": 8.147112923124005e-06,
      "loss": 0.1156,
      "num_input_tokens_seen": 55744816,
      "step": 29420
    },
    {
      "epoch": 4.800228403621829,
      "grad_norm": 0.07397237420082092,
      "learning_rate": 8.13986277932412e-06,
      "loss": 0.0749,
      "num_input_tokens_seen": 55754848,
      "step": 29425
    },
    {
      "epoch": 4.801044130842646,
      "grad_norm": 0.49506980180740356,
      "learning_rate": 8.132615235569277e-06,
      "loss": 0.0383,
      "num_input_tokens_seen": 55763536,
      "step": 29430
    },
    {
      "epoch": 4.801859858063463,
      "grad_norm": 1.7283759117126465,
      "learning_rate": 8.125370292977124e-06,
      "loss": 0.1903,
      "num_input_tokens_seen": 55771808,
      "step": 29435
    },
    {
      "epoch": 4.802675585284281,
      "grad_norm": 2.0296719074249268,
      "learning_rate": 8.118127952664944e-06,
      "loss": 0.2742,
      "num_input_tokens_seen": 55781216,
      "step": 29440
    },
    {
      "epoch": 4.803491312505098,
      "grad_norm": 1.9422361850738525,
      "learning_rate": 8.110888215749574e-06,
      "loss": 0.1003,
      "num_input_tokens_seen": 55790064,
      "step": 29445
    },
    {
      "epoch": 4.8043070397259156,
      "grad_norm": 0.5056707262992859,
      "learning_rate": 8.10365108334749e-06,
      "loss": 0.0729,
      "num_input_tokens_seen": 55800112,
      "step": 29450
    },
    {
      "epoch": 4.805122766946733,
      "grad_norm": 0.07925492525100708,
      "learning_rate": 8.096416556574743e-06,
      "loss": 0.0525,
      "num_input_tokens_seen": 55810272,
      "step": 29455
    },
    {
      "epoch": 4.805938494167551,
      "grad_norm": 0.030003447085618973,
      "learning_rate": 8.08918463654698e-06,
      "loss": 0.1157,
      "num_input_tokens_seen": 55819296,
      "step": 29460
    },
    {
      "epoch": 4.806754221388368,
      "grad_norm": 0.5860475897789001,
      "learning_rate": 8.081955324379458e-06,
      "loss": 0.0665,
      "num_input_tokens_seen": 55829472,
      "step": 29465
    },
    {
      "epoch": 4.807569948609185,
      "grad_norm": 1.3675225973129272,
      "learning_rate": 8.074728621187039e-06,
      "loss": 0.1406,
      "num_input_tokens_seen": 55840240,
      "step": 29470
    },
    {
      "epoch": 4.808385675830002,
      "grad_norm": 1.5117920637130737,
      "learning_rate": 8.067504528084158e-06,
      "loss": 0.165,
      "num_input_tokens_seen": 55849920,
      "step": 29475
    },
    {
      "epoch": 4.80920140305082,
      "grad_norm": 0.5618734955787659,
      "learning_rate": 8.060283046184861e-06,
      "loss": 0.3251,
      "num_input_tokens_seen": 55859408,
      "step": 29480
    },
    {
      "epoch": 4.810017130271637,
      "grad_norm": 0.06731753051280975,
      "learning_rate": 8.053064176602806e-06,
      "loss": 0.0594,
      "num_input_tokens_seen": 55869712,
      "step": 29485
    },
    {
      "epoch": 4.810832857492454,
      "grad_norm": 1.0304038524627686,
      "learning_rate": 8.045847920451216e-06,
      "loss": 0.0652,
      "num_input_tokens_seen": 55878592,
      "step": 29490
    },
    {
      "epoch": 4.8116485847132715,
      "grad_norm": 1.6500060558319092,
      "learning_rate": 8.038634278842944e-06,
      "loss": 0.1181,
      "num_input_tokens_seen": 55888352,
      "step": 29495
    },
    {
      "epoch": 4.8124643119340895,
      "grad_norm": 1.209054946899414,
      "learning_rate": 8.031423252890408e-06,
      "loss": 0.076,
      "num_input_tokens_seen": 55898160,
      "step": 29500
    },
    {
      "epoch": 4.813280039154907,
      "grad_norm": 0.08882233500480652,
      "learning_rate": 8.024214843705646e-06,
      "loss": 0.0312,
      "num_input_tokens_seen": 55907456,
      "step": 29505
    },
    {
      "epoch": 4.814095766375724,
      "grad_norm": 1.375407338142395,
      "learning_rate": 8.017009052400295e-06,
      "loss": 0.1078,
      "num_input_tokens_seen": 55916800,
      "step": 29510
    },
    {
      "epoch": 4.814911493596542,
      "grad_norm": 0.04110652208328247,
      "learning_rate": 8.00980588008557e-06,
      "loss": 0.1728,
      "num_input_tokens_seen": 55926848,
      "step": 29515
    },
    {
      "epoch": 4.815727220817359,
      "grad_norm": 1.2625988721847534,
      "learning_rate": 8.002605327872282e-06,
      "loss": 0.3137,
      "num_input_tokens_seen": 55935520,
      "step": 29520
    },
    {
      "epoch": 4.816542948038176,
      "grad_norm": 1.5978485345840454,
      "learning_rate": 7.995407396870862e-06,
      "loss": 0.18,
      "num_input_tokens_seen": 55944912,
      "step": 29525
    },
    {
      "epoch": 4.817358675258993,
      "grad_norm": 0.6754311919212341,
      "learning_rate": 7.988212088191307e-06,
      "loss": 0.1572,
      "num_input_tokens_seen": 55954112,
      "step": 29530
    },
    {
      "epoch": 4.81817440247981,
      "grad_norm": 1.3208452463150024,
      "learning_rate": 7.98101940294324e-06,
      "loss": 0.4367,
      "num_input_tokens_seen": 55964608,
      "step": 29535
    },
    {
      "epoch": 4.818990129700628,
      "grad_norm": 0.2340446561574936,
      "learning_rate": 7.973829342235847e-06,
      "loss": 0.044,
      "num_input_tokens_seen": 55974176,
      "step": 29540
    },
    {
      "epoch": 4.8198058569214455,
      "grad_norm": 0.6075187921524048,
      "learning_rate": 7.966641907177936e-06,
      "loss": 0.1602,
      "num_input_tokens_seen": 55983120,
      "step": 29545
    },
    {
      "epoch": 4.820621584142263,
      "grad_norm": 0.2879934012889862,
      "learning_rate": 7.959457098877901e-06,
      "loss": 0.0389,
      "num_input_tokens_seen": 55992048,
      "step": 29550
    },
    {
      "epoch": 4.821437311363081,
      "grad_norm": 0.5385817885398865,
      "learning_rate": 7.952274918443719e-06,
      "loss": 0.1687,
      "num_input_tokens_seen": 56001344,
      "step": 29555
    },
    {
      "epoch": 4.822253038583898,
      "grad_norm": 0.08121215552091599,
      "learning_rate": 7.945095366982983e-06,
      "loss": 0.0431,
      "num_input_tokens_seen": 56010912,
      "step": 29560
    },
    {
      "epoch": 4.823068765804715,
      "grad_norm": 0.07327644526958466,
      "learning_rate": 7.937918445602871e-06,
      "loss": 0.0546,
      "num_input_tokens_seen": 56020272,
      "step": 29565
    },
    {
      "epoch": 4.823884493025532,
      "grad_norm": 0.35531869530677795,
      "learning_rate": 7.930744155410145e-06,
      "loss": 0.1392,
      "num_input_tokens_seen": 56028864,
      "step": 29570
    },
    {
      "epoch": 4.824700220246349,
      "grad_norm": 0.5250690579414368,
      "learning_rate": 7.923572497511181e-06,
      "loss": 0.0496,
      "num_input_tokens_seen": 56038320,
      "step": 29575
    },
    {
      "epoch": 4.825515947467167,
      "grad_norm": 0.5365006327629089,
      "learning_rate": 7.916403473011927e-06,
      "loss": 0.101,
      "num_input_tokens_seen": 56047568,
      "step": 29580
    },
    {
      "epoch": 4.826331674687984,
      "grad_norm": 0.8992041349411011,
      "learning_rate": 7.909237083017953e-06,
      "loss": 0.2497,
      "num_input_tokens_seen": 56056640,
      "step": 29585
    },
    {
      "epoch": 4.827147401908801,
      "grad_norm": 2.2032601833343506,
      "learning_rate": 7.902073328634389e-06,
      "loss": 0.4419,
      "num_input_tokens_seen": 56066592,
      "step": 29590
    },
    {
      "epoch": 4.8279631291296194,
      "grad_norm": 0.33899372816085815,
      "learning_rate": 7.894912210965987e-06,
      "loss": 0.0479,
      "num_input_tokens_seen": 56075616,
      "step": 29595
    },
    {
      "epoch": 4.828778856350437,
      "grad_norm": 0.30298691987991333,
      "learning_rate": 7.887753731117075e-06,
      "loss": 0.0454,
      "num_input_tokens_seen": 56085712,
      "step": 29600
    },
    {
      "epoch": 4.828778856350437,
      "eval_loss": 0.1423065960407257,
      "eval_runtime": 133.3367,
      "eval_samples_per_second": 20.437,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 56085712,
      "step": 29600
    },
    {
      "epoch": 4.829594583571254,
      "grad_norm": 0.09858816117048264,
      "learning_rate": 7.880597890191587e-06,
      "loss": 0.1193,
      "num_input_tokens_seen": 56094528,
      "step": 29605
    },
    {
      "epoch": 4.830410310792071,
      "grad_norm": 0.5738723278045654,
      "learning_rate": 7.873444689293036e-06,
      "loss": 0.1544,
      "num_input_tokens_seen": 56104304,
      "step": 29610
    },
    {
      "epoch": 4.831226038012889,
      "grad_norm": 0.42387062311172485,
      "learning_rate": 7.866294129524548e-06,
      "loss": 0.1038,
      "num_input_tokens_seen": 56113328,
      "step": 29615
    },
    {
      "epoch": 4.832041765233706,
      "grad_norm": 0.5852455496788025,
      "learning_rate": 7.859146211988811e-06,
      "loss": 0.1134,
      "num_input_tokens_seen": 56122880,
      "step": 29620
    },
    {
      "epoch": 4.832857492454523,
      "grad_norm": 0.2589658200740814,
      "learning_rate": 7.852000937788134e-06,
      "loss": 0.1838,
      "num_input_tokens_seen": 56133248,
      "step": 29625
    },
    {
      "epoch": 4.83367321967534,
      "grad_norm": 2.041342258453369,
      "learning_rate": 7.844858308024416e-06,
      "loss": 0.1857,
      "num_input_tokens_seen": 56142928,
      "step": 29630
    },
    {
      "epoch": 4.834488946896158,
      "grad_norm": 2.0685579776763916,
      "learning_rate": 7.837718323799122e-06,
      "loss": 0.3643,
      "num_input_tokens_seen": 56152064,
      "step": 29635
    },
    {
      "epoch": 4.835304674116975,
      "grad_norm": 0.9570492506027222,
      "learning_rate": 7.83058098621334e-06,
      "loss": 0.1246,
      "num_input_tokens_seen": 56160752,
      "step": 29640
    },
    {
      "epoch": 4.8361204013377925,
      "grad_norm": 1.1158651113510132,
      "learning_rate": 7.823446296367739e-06,
      "loss": 0.1383,
      "num_input_tokens_seen": 56171184,
      "step": 29645
    },
    {
      "epoch": 4.83693612855861,
      "grad_norm": 0.17867982387542725,
      "learning_rate": 7.81631425536257e-06,
      "loss": 0.2213,
      "num_input_tokens_seen": 56181456,
      "step": 29650
    },
    {
      "epoch": 4.837751855779428,
      "grad_norm": 0.7831807732582092,
      "learning_rate": 7.809184864297689e-06,
      "loss": 0.1254,
      "num_input_tokens_seen": 56191312,
      "step": 29655
    },
    {
      "epoch": 4.838567583000245,
      "grad_norm": 1.3504540920257568,
      "learning_rate": 7.802058124272532e-06,
      "loss": 0.1409,
      "num_input_tokens_seen": 56199584,
      "step": 29660
    },
    {
      "epoch": 4.839383310221062,
      "grad_norm": 0.26965808868408203,
      "learning_rate": 7.79493403638614e-06,
      "loss": 0.0338,
      "num_input_tokens_seen": 56209904,
      "step": 29665
    },
    {
      "epoch": 4.840199037441879,
      "grad_norm": 0.14572514593601227,
      "learning_rate": 7.787812601737132e-06,
      "loss": 0.1992,
      "num_input_tokens_seen": 56219536,
      "step": 29670
    },
    {
      "epoch": 4.841014764662697,
      "grad_norm": 0.4901631772518158,
      "learning_rate": 7.780693821423715e-06,
      "loss": 0.1271,
      "num_input_tokens_seen": 56229456,
      "step": 29675
    },
    {
      "epoch": 4.841830491883514,
      "grad_norm": 0.04463391751050949,
      "learning_rate": 7.773577696543705e-06,
      "loss": 0.128,
      "num_input_tokens_seen": 56240528,
      "step": 29680
    },
    {
      "epoch": 4.842646219104331,
      "grad_norm": 1.0520994663238525,
      "learning_rate": 7.7664642281945e-06,
      "loss": 0.1714,
      "num_input_tokens_seen": 56250464,
      "step": 29685
    },
    {
      "epoch": 4.8434619463251485,
      "grad_norm": 0.36398473381996155,
      "learning_rate": 7.759353417473072e-06,
      "loss": 0.1462,
      "num_input_tokens_seen": 56260640,
      "step": 29690
    },
    {
      "epoch": 4.8442776735459665,
      "grad_norm": 1.3701847791671753,
      "learning_rate": 7.752245265476016e-06,
      "loss": 0.0966,
      "num_input_tokens_seen": 56271136,
      "step": 29695
    },
    {
      "epoch": 4.845093400766784,
      "grad_norm": 0.5451192855834961,
      "learning_rate": 7.745139773299481e-06,
      "loss": 0.13,
      "num_input_tokens_seen": 56280768,
      "step": 29700
    },
    {
      "epoch": 4.845909127987601,
      "grad_norm": 0.6094777584075928,
      "learning_rate": 7.738036942039232e-06,
      "loss": 0.1512,
      "num_input_tokens_seen": 56290240,
      "step": 29705
    },
    {
      "epoch": 4.846724855208418,
      "grad_norm": 1.7121738195419312,
      "learning_rate": 7.73093677279062e-06,
      "loss": 0.5086,
      "num_input_tokens_seen": 56299568,
      "step": 29710
    },
    {
      "epoch": 4.847540582429236,
      "grad_norm": 0.9924996495246887,
      "learning_rate": 7.72383926664857e-06,
      "loss": 0.1176,
      "num_input_tokens_seen": 56309216,
      "step": 29715
    },
    {
      "epoch": 4.848356309650053,
      "grad_norm": 0.7550828456878662,
      "learning_rate": 7.716744424707606e-06,
      "loss": 0.0792,
      "num_input_tokens_seen": 56318112,
      "step": 29720
    },
    {
      "epoch": 4.84917203687087,
      "grad_norm": 1.6703529357910156,
      "learning_rate": 7.709652248061858e-06,
      "loss": 0.1951,
      "num_input_tokens_seen": 56327056,
      "step": 29725
    },
    {
      "epoch": 4.849987764091688,
      "grad_norm": 0.2510428726673126,
      "learning_rate": 7.702562737805017e-06,
      "loss": 0.175,
      "num_input_tokens_seen": 56337264,
      "step": 29730
    },
    {
      "epoch": 4.850803491312505,
      "grad_norm": 0.08183791488409042,
      "learning_rate": 7.695475895030365e-06,
      "loss": 0.1547,
      "num_input_tokens_seen": 56345632,
      "step": 29735
    },
    {
      "epoch": 4.8516192185333225,
      "grad_norm": 0.5621039271354675,
      "learning_rate": 7.6883917208308e-06,
      "loss": 0.0714,
      "num_input_tokens_seen": 56354592,
      "step": 29740
    },
    {
      "epoch": 4.85243494575414,
      "grad_norm": 2.159942865371704,
      "learning_rate": 7.681310216298778e-06,
      "loss": 0.1863,
      "num_input_tokens_seen": 56364784,
      "step": 29745
    },
    {
      "epoch": 4.853250672974957,
      "grad_norm": 1.053968071937561,
      "learning_rate": 7.674231382526367e-06,
      "loss": 0.0815,
      "num_input_tokens_seen": 56374576,
      "step": 29750
    },
    {
      "epoch": 4.854066400195775,
      "grad_norm": 0.06348440051078796,
      "learning_rate": 7.667155220605198e-06,
      "loss": 0.0644,
      "num_input_tokens_seen": 56384048,
      "step": 29755
    },
    {
      "epoch": 4.854882127416592,
      "grad_norm": 0.3701018691062927,
      "learning_rate": 7.660081731626515e-06,
      "loss": 0.0981,
      "num_input_tokens_seen": 56394336,
      "step": 29760
    },
    {
      "epoch": 4.855697854637409,
      "grad_norm": 0.059251125901937485,
      "learning_rate": 7.653010916681141e-06,
      "loss": 0.1839,
      "num_input_tokens_seen": 56403392,
      "step": 29765
    },
    {
      "epoch": 4.856513581858227,
      "grad_norm": 0.15053033828735352,
      "learning_rate": 7.645942776859472e-06,
      "loss": 0.1423,
      "num_input_tokens_seen": 56412384,
      "step": 29770
    },
    {
      "epoch": 4.857329309079044,
      "grad_norm": 0.28753381967544556,
      "learning_rate": 7.63887731325152e-06,
      "loss": 0.1881,
      "num_input_tokens_seen": 56420896,
      "step": 29775
    },
    {
      "epoch": 4.858145036299861,
      "grad_norm": 1.8050965070724487,
      "learning_rate": 7.63181452694685e-06,
      "loss": 0.1689,
      "num_input_tokens_seen": 56429888,
      "step": 29780
    },
    {
      "epoch": 4.858960763520678,
      "grad_norm": 0.152522012591362,
      "learning_rate": 7.624754419034644e-06,
      "loss": 0.0958,
      "num_input_tokens_seen": 56438880,
      "step": 29785
    },
    {
      "epoch": 4.859776490741496,
      "grad_norm": 0.1916237771511078,
      "learning_rate": 7.6176969906036645e-06,
      "loss": 0.2064,
      "num_input_tokens_seen": 56448576,
      "step": 29790
    },
    {
      "epoch": 4.8605922179623136,
      "grad_norm": 2.075183391571045,
      "learning_rate": 7.610642242742242e-06,
      "loss": 0.325,
      "num_input_tokens_seen": 56457312,
      "step": 29795
    },
    {
      "epoch": 4.861407945183131,
      "grad_norm": 0.6478884220123291,
      "learning_rate": 7.603590176538322e-06,
      "loss": 0.0823,
      "num_input_tokens_seen": 56467376,
      "step": 29800
    },
    {
      "epoch": 4.861407945183131,
      "eval_loss": 0.14208869636058807,
      "eval_runtime": 133.2805,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 56467376,
      "step": 29800
    },
    {
      "epoch": 4.862223672403948,
      "grad_norm": 1.5558525323867798,
      "learning_rate": 7.596540793079404e-06,
      "loss": 0.2189,
      "num_input_tokens_seen": 56477680,
      "step": 29805
    },
    {
      "epoch": 4.863039399624766,
      "grad_norm": 0.27312135696411133,
      "learning_rate": 7.5894940934526125e-06,
      "loss": 0.1616,
      "num_input_tokens_seen": 56486080,
      "step": 29810
    },
    {
      "epoch": 4.863855126845583,
      "grad_norm": 0.6142357587814331,
      "learning_rate": 7.582450078744621e-06,
      "loss": 0.1302,
      "num_input_tokens_seen": 56496144,
      "step": 29815
    },
    {
      "epoch": 4.8646708540664,
      "grad_norm": 0.032631322741508484,
      "learning_rate": 7.575408750041707e-06,
      "loss": 0.0303,
      "num_input_tokens_seen": 56505984,
      "step": 29820
    },
    {
      "epoch": 4.865486581287217,
      "grad_norm": 0.18896105885505676,
      "learning_rate": 7.568370108429732e-06,
      "loss": 0.0655,
      "num_input_tokens_seen": 56515408,
      "step": 29825
    },
    {
      "epoch": 4.866302308508035,
      "grad_norm": 0.13366267085075378,
      "learning_rate": 7.561334154994154e-06,
      "loss": 0.1444,
      "num_input_tokens_seen": 56525824,
      "step": 29830
    },
    {
      "epoch": 4.867118035728852,
      "grad_norm": 0.1010451689362526,
      "learning_rate": 7.55430089081999e-06,
      "loss": 0.0511,
      "num_input_tokens_seen": 56535360,
      "step": 29835
    },
    {
      "epoch": 4.8679337629496695,
      "grad_norm": 0.13273003697395325,
      "learning_rate": 7.547270316991864e-06,
      "loss": 0.3258,
      "num_input_tokens_seen": 56544688,
      "step": 29840
    },
    {
      "epoch": 4.868749490170487,
      "grad_norm": 0.24444684386253357,
      "learning_rate": 7.5402424345939884e-06,
      "loss": 0.2309,
      "num_input_tokens_seen": 56554720,
      "step": 29845
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.2206566333770752,
      "learning_rate": 7.533217244710133e-06,
      "loss": 0.1604,
      "num_input_tokens_seen": 56563392,
      "step": 29850
    },
    {
      "epoch": 4.870380944612122,
      "grad_norm": 0.14425550401210785,
      "learning_rate": 7.52619474842369e-06,
      "loss": 0.0658,
      "num_input_tokens_seen": 56572816,
      "step": 29855
    },
    {
      "epoch": 4.871196671832939,
      "grad_norm": 0.034891728311777115,
      "learning_rate": 7.519174946817597e-06,
      "loss": 0.1573,
      "num_input_tokens_seen": 56582848,
      "step": 29860
    },
    {
      "epoch": 4.872012399053756,
      "grad_norm": 0.17506223917007446,
      "learning_rate": 7.512157840974407e-06,
      "loss": 0.0472,
      "num_input_tokens_seen": 56592640,
      "step": 29865
    },
    {
      "epoch": 4.872828126274574,
      "grad_norm": 0.2482057362794876,
      "learning_rate": 7.5051434319762496e-06,
      "loss": 0.1171,
      "num_input_tokens_seen": 56601120,
      "step": 29870
    },
    {
      "epoch": 4.873643853495391,
      "grad_norm": 1.3183329105377197,
      "learning_rate": 7.498131720904822e-06,
      "loss": 0.1491,
      "num_input_tokens_seen": 56610240,
      "step": 29875
    },
    {
      "epoch": 4.874459580716208,
      "grad_norm": 0.3890839219093323,
      "learning_rate": 7.491122708841433e-06,
      "loss": 0.2314,
      "num_input_tokens_seen": 56620048,
      "step": 29880
    },
    {
      "epoch": 4.8752753079370255,
      "grad_norm": 0.42083898186683655,
      "learning_rate": 7.4841163968669524e-06,
      "loss": 0.0479,
      "num_input_tokens_seen": 56629456,
      "step": 29885
    },
    {
      "epoch": 4.8760910351578435,
      "grad_norm": 1.065807819366455,
      "learning_rate": 7.4771127860618355e-06,
      "loss": 0.2505,
      "num_input_tokens_seen": 56637616,
      "step": 29890
    },
    {
      "epoch": 4.876906762378661,
      "grad_norm": 2.187284231185913,
      "learning_rate": 7.470111877506139e-06,
      "loss": 0.214,
      "num_input_tokens_seen": 56645936,
      "step": 29895
    },
    {
      "epoch": 4.877722489599478,
      "grad_norm": 0.6489635705947876,
      "learning_rate": 7.463113672279479e-06,
      "loss": 0.055,
      "num_input_tokens_seen": 56655008,
      "step": 29900
    },
    {
      "epoch": 4.878538216820296,
      "grad_norm": 1.1485731601715088,
      "learning_rate": 7.456118171461071e-06,
      "loss": 0.136,
      "num_input_tokens_seen": 56664032,
      "step": 29905
    },
    {
      "epoch": 4.879353944041113,
      "grad_norm": 0.038086507469415665,
      "learning_rate": 7.449125376129721e-06,
      "loss": 0.0906,
      "num_input_tokens_seen": 56673056,
      "step": 29910
    },
    {
      "epoch": 4.88016967126193,
      "grad_norm": 0.11015000939369202,
      "learning_rate": 7.442135287363788e-06,
      "loss": 0.0573,
      "num_input_tokens_seen": 56682704,
      "step": 29915
    },
    {
      "epoch": 4.880985398482747,
      "grad_norm": 0.02413581870496273,
      "learning_rate": 7.435147906241247e-06,
      "loss": 0.0785,
      "num_input_tokens_seen": 56692768,
      "step": 29920
    },
    {
      "epoch": 4.881801125703564,
      "grad_norm": 0.3972267210483551,
      "learning_rate": 7.428163233839624e-06,
      "loss": 0.0911,
      "num_input_tokens_seen": 56701248,
      "step": 29925
    },
    {
      "epoch": 4.882616852924382,
      "grad_norm": 0.10888677835464478,
      "learning_rate": 7.4211812712360525e-06,
      "loss": 0.0336,
      "num_input_tokens_seen": 56710752,
      "step": 29930
    },
    {
      "epoch": 4.883432580145199,
      "grad_norm": 0.10544852912425995,
      "learning_rate": 7.4142020195072464e-06,
      "loss": 0.0534,
      "num_input_tokens_seen": 56719504,
      "step": 29935
    },
    {
      "epoch": 4.884248307366017,
      "grad_norm": 0.6271689534187317,
      "learning_rate": 7.407225479729479e-06,
      "loss": 0.0665,
      "num_input_tokens_seen": 56728800,
      "step": 29940
    },
    {
      "epoch": 4.885064034586835,
      "grad_norm": 1.0093134641647339,
      "learning_rate": 7.400251652978632e-06,
      "loss": 0.1569,
      "num_input_tokens_seen": 56738736,
      "step": 29945
    },
    {
      "epoch": 4.885879761807652,
      "grad_norm": 0.35360974073410034,
      "learning_rate": 7.393280540330147e-06,
      "loss": 0.1257,
      "num_input_tokens_seen": 56748576,
      "step": 29950
    },
    {
      "epoch": 4.886695489028469,
      "grad_norm": 0.2386387437582016,
      "learning_rate": 7.386312142859069e-06,
      "loss": 0.1147,
      "num_input_tokens_seen": 56756896,
      "step": 29955
    },
    {
      "epoch": 4.887511216249286,
      "grad_norm": 0.1250912845134735,
      "learning_rate": 7.379346461640008e-06,
      "loss": 0.1247,
      "num_input_tokens_seen": 56767008,
      "step": 29960
    },
    {
      "epoch": 4.888326943470103,
      "grad_norm": 1.4630967378616333,
      "learning_rate": 7.372383497747149e-06,
      "loss": 0.1994,
      "num_input_tokens_seen": 56775376,
      "step": 29965
    },
    {
      "epoch": 4.889142670690921,
      "grad_norm": 0.05049682781100273,
      "learning_rate": 7.3654232522542775e-06,
      "loss": 0.0265,
      "num_input_tokens_seen": 56784688,
      "step": 29970
    },
    {
      "epoch": 4.889958397911738,
      "grad_norm": 0.5301603078842163,
      "learning_rate": 7.358465726234756e-06,
      "loss": 0.1626,
      "num_input_tokens_seen": 56793392,
      "step": 29975
    },
    {
      "epoch": 4.890774125132555,
      "grad_norm": 0.07334969192743301,
      "learning_rate": 7.351510920761512e-06,
      "loss": 0.0373,
      "num_input_tokens_seen": 56803296,
      "step": 29980
    },
    {
      "epoch": 4.891589852353373,
      "grad_norm": 0.780490517616272,
      "learning_rate": 7.344558836907067e-06,
      "loss": 0.0909,
      "num_input_tokens_seen": 56813328,
      "step": 29985
    },
    {
      "epoch": 4.8924055795741905,
      "grad_norm": 1.3333560228347778,
      "learning_rate": 7.3376094757435285e-06,
      "loss": 0.1909,
      "num_input_tokens_seen": 56823136,
      "step": 29990
    },
    {
      "epoch": 4.893221306795008,
      "grad_norm": 2.0159730911254883,
      "learning_rate": 7.330662838342561e-06,
      "loss": 0.1434,
      "num_input_tokens_seen": 56832704,
      "step": 29995
    },
    {
      "epoch": 4.894037034015825,
      "grad_norm": 0.25041449069976807,
      "learning_rate": 7.323718925775438e-06,
      "loss": 0.0687,
      "num_input_tokens_seen": 56841328,
      "step": 30000
    },
    {
      "epoch": 4.894037034015825,
      "eval_loss": 0.142201766371727,
      "eval_runtime": 133.2075,
      "eval_samples_per_second": 20.457,
      "eval_steps_per_second": 10.232,
      "num_input_tokens_seen": 56841328,
      "step": 30000
    },
    {
      "epoch": 4.894852761236643,
      "grad_norm": 0.08296425640583038,
      "learning_rate": 7.316777739112985e-06,
      "loss": 0.0956,
      "num_input_tokens_seen": 56849184,
      "step": 30005
    },
    {
      "epoch": 4.89566848845746,
      "grad_norm": 0.08642758429050446,
      "learning_rate": 7.309839279425626e-06,
      "loss": 0.0704,
      "num_input_tokens_seen": 56859440,
      "step": 30010
    },
    {
      "epoch": 4.896484215678277,
      "grad_norm": 0.34050339460372925,
      "learning_rate": 7.302903547783366e-06,
      "loss": 0.2297,
      "num_input_tokens_seen": 56869808,
      "step": 30015
    },
    {
      "epoch": 4.897299942899094,
      "grad_norm": 0.07631172984838486,
      "learning_rate": 7.2959705452557644e-06,
      "loss": 0.0846,
      "num_input_tokens_seen": 56880768,
      "step": 30020
    },
    {
      "epoch": 4.898115670119912,
      "grad_norm": 1.0914180278778076,
      "learning_rate": 7.289040272911996e-06,
      "loss": 0.0785,
      "num_input_tokens_seen": 56891360,
      "step": 30025
    },
    {
      "epoch": 4.898931397340729,
      "grad_norm": 0.21022039651870728,
      "learning_rate": 7.282112731820789e-06,
      "loss": 0.1968,
      "num_input_tokens_seen": 56901872,
      "step": 30030
    },
    {
      "epoch": 4.8997471245615465,
      "grad_norm": 0.04897032678127289,
      "learning_rate": 7.275187923050447e-06,
      "loss": 0.0723,
      "num_input_tokens_seen": 56911504,
      "step": 30035
    },
    {
      "epoch": 4.900562851782364,
      "grad_norm": 0.15136435627937317,
      "learning_rate": 7.268265847668879e-06,
      "loss": 0.0482,
      "num_input_tokens_seen": 56921408,
      "step": 30040
    },
    {
      "epoch": 4.901378579003182,
      "grad_norm": 0.3841019868850708,
      "learning_rate": 7.261346506743538e-06,
      "loss": 0.1243,
      "num_input_tokens_seen": 56931472,
      "step": 30045
    },
    {
      "epoch": 4.902194306223999,
      "grad_norm": 0.41010206937789917,
      "learning_rate": 7.254429901341486e-06,
      "loss": 0.0719,
      "num_input_tokens_seen": 56940720,
      "step": 30050
    },
    {
      "epoch": 4.903010033444816,
      "grad_norm": 0.7487492561340332,
      "learning_rate": 7.247516032529356e-06,
      "loss": 0.2031,
      "num_input_tokens_seen": 56949616,
      "step": 30055
    },
    {
      "epoch": 4.903825760665633,
      "grad_norm": 0.14788830280303955,
      "learning_rate": 7.240604901373338e-06,
      "loss": 0.0555,
      "num_input_tokens_seen": 56958384,
      "step": 30060
    },
    {
      "epoch": 4.904641487886451,
      "grad_norm": 1.2884219884872437,
      "learning_rate": 7.233696508939223e-06,
      "loss": 0.128,
      "num_input_tokens_seen": 56967952,
      "step": 30065
    },
    {
      "epoch": 4.905457215107268,
      "grad_norm": 0.24899253249168396,
      "learning_rate": 7.226790856292376e-06,
      "loss": 0.122,
      "num_input_tokens_seen": 56977872,
      "step": 30070
    },
    {
      "epoch": 4.906272942328085,
      "grad_norm": 0.7993634343147278,
      "learning_rate": 7.219887944497727e-06,
      "loss": 0.2603,
      "num_input_tokens_seen": 56987552,
      "step": 30075
    },
    {
      "epoch": 4.907088669548903,
      "grad_norm": 0.22800800204277039,
      "learning_rate": 7.2129877746198e-06,
      "loss": 0.0277,
      "num_input_tokens_seen": 56996304,
      "step": 30080
    },
    {
      "epoch": 4.9079043967697205,
      "grad_norm": 0.04589689150452614,
      "learning_rate": 7.20609034772268e-06,
      "loss": 0.1226,
      "num_input_tokens_seen": 57006448,
      "step": 30085
    },
    {
      "epoch": 4.908720123990538,
      "grad_norm": 0.032296568155288696,
      "learning_rate": 7.19919566487004e-06,
      "loss": 0.2181,
      "num_input_tokens_seen": 57015776,
      "step": 30090
    },
    {
      "epoch": 4.909535851211355,
      "grad_norm": 0.4764484465122223,
      "learning_rate": 7.192303727125132e-06,
      "loss": 0.0895,
      "num_input_tokens_seen": 57025152,
      "step": 30095
    },
    {
      "epoch": 4.910351578432172,
      "grad_norm": 0.05260420963168144,
      "learning_rate": 7.185414535550777e-06,
      "loss": 0.3402,
      "num_input_tokens_seen": 57035120,
      "step": 30100
    },
    {
      "epoch": 4.91116730565299,
      "grad_norm": 0.6795530915260315,
      "learning_rate": 7.178528091209363e-06,
      "loss": 0.1906,
      "num_input_tokens_seen": 57045648,
      "step": 30105
    },
    {
      "epoch": 4.911983032873807,
      "grad_norm": 0.3148026168346405,
      "learning_rate": 7.171644395162888e-06,
      "loss": 0.0714,
      "num_input_tokens_seen": 57056144,
      "step": 30110
    },
    {
      "epoch": 4.912798760094624,
      "grad_norm": 0.1950521469116211,
      "learning_rate": 7.164763448472881e-06,
      "loss": 0.1526,
      "num_input_tokens_seen": 57066288,
      "step": 30115
    },
    {
      "epoch": 4.913614487315442,
      "grad_norm": 0.6015404462814331,
      "learning_rate": 7.157885252200491e-06,
      "loss": 0.0469,
      "num_input_tokens_seen": 57076096,
      "step": 30120
    },
    {
      "epoch": 4.914430214536259,
      "grad_norm": 0.654499888420105,
      "learning_rate": 7.151009807406403e-06,
      "loss": 0.2117,
      "num_input_tokens_seen": 57085360,
      "step": 30125
    },
    {
      "epoch": 4.915245941757076,
      "grad_norm": 0.31173595786094666,
      "learning_rate": 7.144137115150909e-06,
      "loss": 0.0982,
      "num_input_tokens_seen": 57096048,
      "step": 30130
    },
    {
      "epoch": 4.9160616689778935,
      "grad_norm": 1.2812494039535522,
      "learning_rate": 7.1372671764938725e-06,
      "loss": 0.0754,
      "num_input_tokens_seen": 57106368,
      "step": 30135
    },
    {
      "epoch": 4.916877396198711,
      "grad_norm": 1.5895401239395142,
      "learning_rate": 7.130399992494705e-06,
      "loss": 0.2124,
      "num_input_tokens_seen": 57114512,
      "step": 30140
    },
    {
      "epoch": 4.917693123419529,
      "grad_norm": 0.9045619964599609,
      "learning_rate": 7.123535564212419e-06,
      "loss": 0.1105,
      "num_input_tokens_seen": 57124432,
      "step": 30145
    },
    {
      "epoch": 4.918508850640346,
      "grad_norm": 0.19007492065429688,
      "learning_rate": 7.116673892705611e-06,
      "loss": 0.0806,
      "num_input_tokens_seen": 57134320,
      "step": 30150
    },
    {
      "epoch": 4.919324577861163,
      "grad_norm": 0.18055307865142822,
      "learning_rate": 7.109814979032415e-06,
      "loss": 0.0851,
      "num_input_tokens_seen": 57144352,
      "step": 30155
    },
    {
      "epoch": 4.920140305081981,
      "grad_norm": 0.1354542076587677,
      "learning_rate": 7.102958824250577e-06,
      "loss": 0.2397,
      "num_input_tokens_seen": 57153424,
      "step": 30160
    },
    {
      "epoch": 4.920956032302798,
      "grad_norm": 0.2953135073184967,
      "learning_rate": 7.096105429417393e-06,
      "loss": 0.0275,
      "num_input_tokens_seen": 57162992,
      "step": 30165
    },
    {
      "epoch": 4.921771759523615,
      "grad_norm": 1.6887171268463135,
      "learning_rate": 7.0892547955897506e-06,
      "loss": 0.2794,
      "num_input_tokens_seen": 57171376,
      "step": 30170
    },
    {
      "epoch": 4.922587486744432,
      "grad_norm": 0.10389580577611923,
      "learning_rate": 7.0824069238241e-06,
      "loss": 0.0165,
      "num_input_tokens_seen": 57181088,
      "step": 30175
    },
    {
      "epoch": 4.92340321396525,
      "grad_norm": 0.2699151337146759,
      "learning_rate": 7.075561815176462e-06,
      "loss": 0.126,
      "num_input_tokens_seen": 57189936,
      "step": 30180
    },
    {
      "epoch": 4.9242189411860675,
      "grad_norm": 0.7642753720283508,
      "learning_rate": 7.068719470702445e-06,
      "loss": 0.1471,
      "num_input_tokens_seen": 57198928,
      "step": 30185
    },
    {
      "epoch": 4.925034668406885,
      "grad_norm": 0.4939224123954773,
      "learning_rate": 7.061879891457229e-06,
      "loss": 0.042,
      "num_input_tokens_seen": 57208912,
      "step": 30190
    },
    {
      "epoch": 4.925850395627702,
      "grad_norm": 0.5557674765586853,
      "learning_rate": 7.0550430784955515e-06,
      "loss": 0.1289,
      "num_input_tokens_seen": 57217840,
      "step": 30195
    },
    {
      "epoch": 4.92666612284852,
      "grad_norm": 0.5635462403297424,
      "learning_rate": 7.048209032871752e-06,
      "loss": 0.0976,
      "num_input_tokens_seen": 57227184,
      "step": 30200
    },
    {
      "epoch": 4.92666612284852,
      "eval_loss": 0.14320412278175354,
      "eval_runtime": 133.3409,
      "eval_samples_per_second": 20.436,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 57227184,
      "step": 30200
    },
    {
      "epoch": 4.927481850069337,
      "grad_norm": 0.10042384266853333,
      "learning_rate": 7.0413777556397055e-06,
      "loss": 0.0965,
      "num_input_tokens_seen": 57236192,
      "step": 30205
    },
    {
      "epoch": 4.928297577290154,
      "grad_norm": 0.16449537873268127,
      "learning_rate": 7.0345492478528925e-06,
      "loss": 0.1737,
      "num_input_tokens_seen": 57245776,
      "step": 30210
    },
    {
      "epoch": 4.929113304510971,
      "grad_norm": 0.2751343846321106,
      "learning_rate": 7.02772351056436e-06,
      "loss": 0.1535,
      "num_input_tokens_seen": 57255408,
      "step": 30215
    },
    {
      "epoch": 4.929929031731789,
      "grad_norm": 0.26217079162597656,
      "learning_rate": 7.020900544826709e-06,
      "loss": 0.0432,
      "num_input_tokens_seen": 57265296,
      "step": 30220
    },
    {
      "epoch": 4.930744758952606,
      "grad_norm": 1.996066927909851,
      "learning_rate": 7.014080351692134e-06,
      "loss": 0.2616,
      "num_input_tokens_seen": 57275024,
      "step": 30225
    },
    {
      "epoch": 4.9315604861734235,
      "grad_norm": 1.440126895904541,
      "learning_rate": 7.0072629322124024e-06,
      "loss": 0.0965,
      "num_input_tokens_seen": 57284448,
      "step": 30230
    },
    {
      "epoch": 4.932376213394241,
      "grad_norm": 0.20027311146259308,
      "learning_rate": 7.000448287438827e-06,
      "loss": 0.2162,
      "num_input_tokens_seen": 57294000,
      "step": 30235
    },
    {
      "epoch": 4.933191940615059,
      "grad_norm": 0.21727125346660614,
      "learning_rate": 6.993636418422331e-06,
      "loss": 0.0351,
      "num_input_tokens_seen": 57303536,
      "step": 30240
    },
    {
      "epoch": 4.934007667835876,
      "grad_norm": 0.2783123850822449,
      "learning_rate": 6.986827326213383e-06,
      "loss": 0.0429,
      "num_input_tokens_seen": 57311664,
      "step": 30245
    },
    {
      "epoch": 4.934823395056693,
      "grad_norm": 1.5951999425888062,
      "learning_rate": 6.9800210118620205e-06,
      "loss": 0.1138,
      "num_input_tokens_seen": 57320032,
      "step": 30250
    },
    {
      "epoch": 4.935639122277511,
      "grad_norm": 2.2817764282226562,
      "learning_rate": 6.973217476417876e-06,
      "loss": 0.3641,
      "num_input_tokens_seen": 57330288,
      "step": 30255
    },
    {
      "epoch": 4.936454849498328,
      "grad_norm": 0.40001749992370605,
      "learning_rate": 6.96641672093013e-06,
      "loss": 0.187,
      "num_input_tokens_seen": 57340432,
      "step": 30260
    },
    {
      "epoch": 4.937270576719145,
      "grad_norm": 0.26731306314468384,
      "learning_rate": 6.95961874644755e-06,
      "loss": 0.0453,
      "num_input_tokens_seen": 57350528,
      "step": 30265
    },
    {
      "epoch": 4.938086303939962,
      "grad_norm": 0.8958410024642944,
      "learning_rate": 6.952823554018476e-06,
      "loss": 0.0674,
      "num_input_tokens_seen": 57360384,
      "step": 30270
    },
    {
      "epoch": 4.938902031160779,
      "grad_norm": 0.8308261036872864,
      "learning_rate": 6.946031144690798e-06,
      "loss": 0.2888,
      "num_input_tokens_seen": 57369056,
      "step": 30275
    },
    {
      "epoch": 4.939717758381597,
      "grad_norm": 0.26347219944000244,
      "learning_rate": 6.939241519512005e-06,
      "loss": 0.1401,
      "num_input_tokens_seen": 57378672,
      "step": 30280
    },
    {
      "epoch": 4.940533485602415,
      "grad_norm": 0.037010304629802704,
      "learning_rate": 6.932454679529129e-06,
      "loss": 0.0814,
      "num_input_tokens_seen": 57388944,
      "step": 30285
    },
    {
      "epoch": 4.941349212823232,
      "grad_norm": 0.46685898303985596,
      "learning_rate": 6.925670625788791e-06,
      "loss": 0.1492,
      "num_input_tokens_seen": 57399168,
      "step": 30290
    },
    {
      "epoch": 4.94216494004405,
      "grad_norm": 0.7175585627555847,
      "learning_rate": 6.918889359337186e-06,
      "loss": 0.2984,
      "num_input_tokens_seen": 57409264,
      "step": 30295
    },
    {
      "epoch": 4.942980667264867,
      "grad_norm": 0.043527841567993164,
      "learning_rate": 6.912110881220058e-06,
      "loss": 0.048,
      "num_input_tokens_seen": 57418400,
      "step": 30300
    },
    {
      "epoch": 4.943796394485684,
      "grad_norm": 1.4914182424545288,
      "learning_rate": 6.905335192482735e-06,
      "loss": 0.1927,
      "num_input_tokens_seen": 57427824,
      "step": 30305
    },
    {
      "epoch": 4.944612121706501,
      "grad_norm": 0.5210130214691162,
      "learning_rate": 6.8985622941701275e-06,
      "loss": 0.1273,
      "num_input_tokens_seen": 57438176,
      "step": 30310
    },
    {
      "epoch": 4.945427848927318,
      "grad_norm": 0.13741575181484222,
      "learning_rate": 6.89179218732669e-06,
      "loss": 0.0485,
      "num_input_tokens_seen": 57446816,
      "step": 30315
    },
    {
      "epoch": 4.946243576148136,
      "grad_norm": 0.6284762024879456,
      "learning_rate": 6.8850248729964595e-06,
      "loss": 0.0551,
      "num_input_tokens_seen": 57456288,
      "step": 30320
    },
    {
      "epoch": 4.947059303368953,
      "grad_norm": 0.022126853466033936,
      "learning_rate": 6.8782603522230314e-06,
      "loss": 0.0993,
      "num_input_tokens_seen": 57464640,
      "step": 30325
    },
    {
      "epoch": 4.9478750305897705,
      "grad_norm": 0.08707533776760101,
      "learning_rate": 6.871498626049591e-06,
      "loss": 0.117,
      "num_input_tokens_seen": 57474272,
      "step": 30330
    },
    {
      "epoch": 4.9486907578105885,
      "grad_norm": 1.0175385475158691,
      "learning_rate": 6.8647396955188875e-06,
      "loss": 0.2538,
      "num_input_tokens_seen": 57484336,
      "step": 30335
    },
    {
      "epoch": 4.949506485031406,
      "grad_norm": 0.8256726264953613,
      "learning_rate": 6.857983561673218e-06,
      "loss": 0.0254,
      "num_input_tokens_seen": 57492688,
      "step": 30340
    },
    {
      "epoch": 4.950322212252223,
      "grad_norm": 0.328610360622406,
      "learning_rate": 6.851230225554467e-06,
      "loss": 0.0539,
      "num_input_tokens_seen": 57501680,
      "step": 30345
    },
    {
      "epoch": 4.95113793947304,
      "grad_norm": 1.3511247634887695,
      "learning_rate": 6.8444796882040946e-06,
      "loss": 0.1581,
      "num_input_tokens_seen": 57510704,
      "step": 30350
    },
    {
      "epoch": 4.951953666693858,
      "grad_norm": 0.30751222372055054,
      "learning_rate": 6.837731950663106e-06,
      "loss": 0.1139,
      "num_input_tokens_seen": 57518976,
      "step": 30355
    },
    {
      "epoch": 4.952769393914675,
      "grad_norm": 1.2966225147247314,
      "learning_rate": 6.830987013972098e-06,
      "loss": 0.1886,
      "num_input_tokens_seen": 57528896,
      "step": 30360
    },
    {
      "epoch": 4.953585121135492,
      "grad_norm": 1.5427383184432983,
      "learning_rate": 6.82424487917121e-06,
      "loss": 0.1203,
      "num_input_tokens_seen": 57537488,
      "step": 30365
    },
    {
      "epoch": 4.954400848356309,
      "grad_norm": 0.045371055603027344,
      "learning_rate": 6.8175055473001735e-06,
      "loss": 0.0556,
      "num_input_tokens_seen": 57547024,
      "step": 30370
    },
    {
      "epoch": 4.955216575577127,
      "grad_norm": 1.3987867832183838,
      "learning_rate": 6.8107690193982855e-06,
      "loss": 0.1356,
      "num_input_tokens_seen": 57556912,
      "step": 30375
    },
    {
      "epoch": 4.9560323027979445,
      "grad_norm": 0.8699240684509277,
      "learning_rate": 6.804035296504385e-06,
      "loss": 0.1579,
      "num_input_tokens_seen": 57566736,
      "step": 30380
    },
    {
      "epoch": 4.956848030018762,
      "grad_norm": 0.04943171143531799,
      "learning_rate": 6.797304379656916e-06,
      "loss": 0.0993,
      "num_input_tokens_seen": 57577104,
      "step": 30385
    },
    {
      "epoch": 4.957663757239579,
      "grad_norm": 0.26590320467948914,
      "learning_rate": 6.790576269893861e-06,
      "loss": 0.2412,
      "num_input_tokens_seen": 57587568,
      "step": 30390
    },
    {
      "epoch": 4.958479484460397,
      "grad_norm": 0.5869526267051697,
      "learning_rate": 6.783850968252772e-06,
      "loss": 0.2853,
      "num_input_tokens_seen": 57596752,
      "step": 30395
    },
    {
      "epoch": 4.959295211681214,
      "grad_norm": 1.0576030015945435,
      "learning_rate": 6.777128475770789e-06,
      "loss": 0.0957,
      "num_input_tokens_seen": 57605632,
      "step": 30400
    },
    {
      "epoch": 4.959295211681214,
      "eval_loss": 0.14259937405586243,
      "eval_runtime": 133.3105,
      "eval_samples_per_second": 20.441,
      "eval_steps_per_second": 10.224,
      "num_input_tokens_seen": 57605632,
      "step": 30400
    },
    {
      "epoch": 4.960110938902031,
      "grad_norm": 0.7712345719337463,
      "learning_rate": 6.77040879348459e-06,
      "loss": 0.2476,
      "num_input_tokens_seen": 57615824,
      "step": 30405
    },
    {
      "epoch": 4.960926666122848,
      "grad_norm": 1.0283242464065552,
      "learning_rate": 6.763691922430443e-06,
      "loss": 0.3148,
      "num_input_tokens_seen": 57625648,
      "step": 30410
    },
    {
      "epoch": 4.961742393343666,
      "grad_norm": 0.5708649754524231,
      "learning_rate": 6.756977863644178e-06,
      "loss": 0.1956,
      "num_input_tokens_seen": 57635456,
      "step": 30415
    },
    {
      "epoch": 4.962558120564483,
      "grad_norm": 0.494616836309433,
      "learning_rate": 6.7502666181611804e-06,
      "loss": 0.1227,
      "num_input_tokens_seen": 57644448,
      "step": 30420
    },
    {
      "epoch": 4.9633738477853,
      "grad_norm": 1.3329201936721802,
      "learning_rate": 6.743558187016405e-06,
      "loss": 0.1352,
      "num_input_tokens_seen": 57653984,
      "step": 30425
    },
    {
      "epoch": 4.964189575006118,
      "grad_norm": 0.776624321937561,
      "learning_rate": 6.7368525712443925e-06,
      "loss": 0.1266,
      "num_input_tokens_seen": 57663216,
      "step": 30430
    },
    {
      "epoch": 4.965005302226936,
      "grad_norm": 1.470268964767456,
      "learning_rate": 6.7301497718792155e-06,
      "loss": 0.0753,
      "num_input_tokens_seen": 57672672,
      "step": 30435
    },
    {
      "epoch": 4.965821029447753,
      "grad_norm": 0.656415581703186,
      "learning_rate": 6.723449789954544e-06,
      "loss": 0.1636,
      "num_input_tokens_seen": 57682064,
      "step": 30440
    },
    {
      "epoch": 4.96663675666857,
      "grad_norm": 0.618819534778595,
      "learning_rate": 6.716752626503586e-06,
      "loss": 0.0854,
      "num_input_tokens_seen": 57690752,
      "step": 30445
    },
    {
      "epoch": 4.967452483889387,
      "grad_norm": 0.36129823327064514,
      "learning_rate": 6.710058282559131e-06,
      "loss": 0.2526,
      "num_input_tokens_seen": 57701824,
      "step": 30450
    },
    {
      "epoch": 4.968268211110205,
      "grad_norm": 0.40610069036483765,
      "learning_rate": 6.703366759153545e-06,
      "loss": 0.093,
      "num_input_tokens_seen": 57710880,
      "step": 30455
    },
    {
      "epoch": 4.969083938331022,
      "grad_norm": 0.1416415125131607,
      "learning_rate": 6.6966780573187335e-06,
      "loss": 0.0937,
      "num_input_tokens_seen": 57720112,
      "step": 30460
    },
    {
      "epoch": 4.969899665551839,
      "grad_norm": 0.38564321398735046,
      "learning_rate": 6.689992178086174e-06,
      "loss": 0.0713,
      "num_input_tokens_seen": 57730384,
      "step": 30465
    },
    {
      "epoch": 4.970715392772657,
      "grad_norm": 0.7693479061126709,
      "learning_rate": 6.683309122486925e-06,
      "loss": 0.1395,
      "num_input_tokens_seen": 57739824,
      "step": 30470
    },
    {
      "epoch": 4.971531119993474,
      "grad_norm": 0.09481506794691086,
      "learning_rate": 6.676628891551584e-06,
      "loss": 0.2214,
      "num_input_tokens_seen": 57749376,
      "step": 30475
    },
    {
      "epoch": 4.9723468472142915,
      "grad_norm": 2.1693458557128906,
      "learning_rate": 6.6699514863103385e-06,
      "loss": 0.2348,
      "num_input_tokens_seen": 57760080,
      "step": 30480
    },
    {
      "epoch": 4.973162574435109,
      "grad_norm": 0.02847280167043209,
      "learning_rate": 6.663276907792921e-06,
      "loss": 0.2467,
      "num_input_tokens_seen": 57769360,
      "step": 30485
    },
    {
      "epoch": 4.973978301655926,
      "grad_norm": 0.9310910701751709,
      "learning_rate": 6.656605157028634e-06,
      "loss": 0.1138,
      "num_input_tokens_seen": 57778064,
      "step": 30490
    },
    {
      "epoch": 4.974794028876744,
      "grad_norm": 0.519510805606842,
      "learning_rate": 6.649936235046358e-06,
      "loss": 0.0801,
      "num_input_tokens_seen": 57787008,
      "step": 30495
    },
    {
      "epoch": 4.975609756097561,
      "grad_norm": 1.3964474201202393,
      "learning_rate": 6.643270142874508e-06,
      "loss": 0.2496,
      "num_input_tokens_seen": 57796368,
      "step": 30500
    },
    {
      "epoch": 4.976425483318378,
      "grad_norm": 0.028613148257136345,
      "learning_rate": 6.636606881541094e-06,
      "loss": 0.1766,
      "num_input_tokens_seen": 57804864,
      "step": 30505
    },
    {
      "epoch": 4.977241210539196,
      "grad_norm": 0.4702196717262268,
      "learning_rate": 6.629946452073662e-06,
      "loss": 0.0598,
      "num_input_tokens_seen": 57814160,
      "step": 30510
    },
    {
      "epoch": 4.978056937760013,
      "grad_norm": 0.5903986692428589,
      "learning_rate": 6.6232888554993375e-06,
      "loss": 0.1187,
      "num_input_tokens_seen": 57823136,
      "step": 30515
    },
    {
      "epoch": 4.97887266498083,
      "grad_norm": 0.4037543535232544,
      "learning_rate": 6.616634092844817e-06,
      "loss": 0.3096,
      "num_input_tokens_seen": 57832640,
      "step": 30520
    },
    {
      "epoch": 4.9796883922016475,
      "grad_norm": 0.2423114776611328,
      "learning_rate": 6.609982165136331e-06,
      "loss": 0.1183,
      "num_input_tokens_seen": 57841408,
      "step": 30525
    },
    {
      "epoch": 4.9805041194224655,
      "grad_norm": 0.2125813513994217,
      "learning_rate": 6.603333073399706e-06,
      "loss": 0.06,
      "num_input_tokens_seen": 57850624,
      "step": 30530
    },
    {
      "epoch": 4.981319846643283,
      "grad_norm": 0.44395968317985535,
      "learning_rate": 6.596686818660308e-06,
      "loss": 0.0947,
      "num_input_tokens_seen": 57861008,
      "step": 30535
    },
    {
      "epoch": 4.9821355738641,
      "grad_norm": 1.3051291704177856,
      "learning_rate": 6.590043401943066e-06,
      "loss": 0.2187,
      "num_input_tokens_seen": 57870624,
      "step": 30540
    },
    {
      "epoch": 4.982951301084917,
      "grad_norm": 0.4525304138660431,
      "learning_rate": 6.583402824272494e-06,
      "loss": 0.1208,
      "num_input_tokens_seen": 57880672,
      "step": 30545
    },
    {
      "epoch": 4.983767028305735,
      "grad_norm": 0.7460183501243591,
      "learning_rate": 6.576765086672634e-06,
      "loss": 0.0615,
      "num_input_tokens_seen": 57890128,
      "step": 30550
    },
    {
      "epoch": 4.984582755526552,
      "grad_norm": 0.11456378549337387,
      "learning_rate": 6.57013019016712e-06,
      "loss": 0.068,
      "num_input_tokens_seen": 57899760,
      "step": 30555
    },
    {
      "epoch": 4.985398482747369,
      "grad_norm": 1.4181017875671387,
      "learning_rate": 6.563498135779142e-06,
      "loss": 0.1682,
      "num_input_tokens_seen": 57908624,
      "step": 30560
    },
    {
      "epoch": 4.986214209968186,
      "grad_norm": 0.9270327091217041,
      "learning_rate": 6.556868924531431e-06,
      "loss": 0.1817,
      "num_input_tokens_seen": 57918112,
      "step": 30565
    },
    {
      "epoch": 4.987029937189004,
      "grad_norm": 0.18474751710891724,
      "learning_rate": 6.550242557446304e-06,
      "loss": 0.0315,
      "num_input_tokens_seen": 57928864,
      "step": 30570
    },
    {
      "epoch": 4.9878456644098215,
      "grad_norm": 0.13399994373321533,
      "learning_rate": 6.543619035545634e-06,
      "loss": 0.0638,
      "num_input_tokens_seen": 57938576,
      "step": 30575
    },
    {
      "epoch": 4.988661391630639,
      "grad_norm": 1.19684636592865,
      "learning_rate": 6.53699835985084e-06,
      "loss": 0.246,
      "num_input_tokens_seen": 57948272,
      "step": 30580
    },
    {
      "epoch": 4.989477118851456,
      "grad_norm": 0.42064404487609863,
      "learning_rate": 6.530380531382927e-06,
      "loss": 0.1461,
      "num_input_tokens_seen": 57957456,
      "step": 30585
    },
    {
      "epoch": 4.990292846072274,
      "grad_norm": 0.4268341362476349,
      "learning_rate": 6.523765551162433e-06,
      "loss": 0.0608,
      "num_input_tokens_seen": 57967600,
      "step": 30590
    },
    {
      "epoch": 4.991108573293091,
      "grad_norm": 0.41341593861579895,
      "learning_rate": 6.517153420209476e-06,
      "loss": 0.1899,
      "num_input_tokens_seen": 57977424,
      "step": 30595
    },
    {
      "epoch": 4.991924300513908,
      "grad_norm": 0.5718939304351807,
      "learning_rate": 6.510544139543739e-06,
      "loss": 0.1694,
      "num_input_tokens_seen": 57987472,
      "step": 30600
    },
    {
      "epoch": 4.991924300513908,
      "eval_loss": 0.1417924463748932,
      "eval_runtime": 133.3773,
      "eval_samples_per_second": 20.431,
      "eval_steps_per_second": 10.219,
      "num_input_tokens_seen": 57987472,
      "step": 30600
    },
    {
      "epoch": 4.992740027734725,
      "grad_norm": 0.2393948882818222,
      "learning_rate": 6.503937710184452e-06,
      "loss": 0.2103,
      "num_input_tokens_seen": 57996720,
      "step": 30605
    },
    {
      "epoch": 4.993555754955543,
      "grad_norm": 1.5970405340194702,
      "learning_rate": 6.4973341331503954e-06,
      "loss": 0.1184,
      "num_input_tokens_seen": 58005568,
      "step": 30610
    },
    {
      "epoch": 4.99437148217636,
      "grad_norm": 1.9484847784042358,
      "learning_rate": 6.490733409459942e-06,
      "loss": 0.1714,
      "num_input_tokens_seen": 58015360,
      "step": 30615
    },
    {
      "epoch": 4.995187209397177,
      "grad_norm": 0.14042864739894867,
      "learning_rate": 6.484135540130995e-06,
      "loss": 0.1584,
      "num_input_tokens_seen": 58024480,
      "step": 30620
    },
    {
      "epoch": 4.9960029366179945,
      "grad_norm": 1.5303629636764526,
      "learning_rate": 6.4775405261810364e-06,
      "loss": 0.1764,
      "num_input_tokens_seen": 58033104,
      "step": 30625
    },
    {
      "epoch": 4.996818663838813,
      "grad_norm": 1.0816657543182373,
      "learning_rate": 6.470948368627092e-06,
      "loss": 0.2709,
      "num_input_tokens_seen": 58043456,
      "step": 30630
    },
    {
      "epoch": 4.99763439105963,
      "grad_norm": 0.7844672799110413,
      "learning_rate": 6.464359068485756e-06,
      "loss": 0.2909,
      "num_input_tokens_seen": 58052752,
      "step": 30635
    },
    {
      "epoch": 4.998450118280447,
      "grad_norm": 0.07036247849464417,
      "learning_rate": 6.457772626773195e-06,
      "loss": 0.0237,
      "num_input_tokens_seen": 58061056,
      "step": 30640
    },
    {
      "epoch": 4.999265845501265,
      "grad_norm": 0.13484428822994232,
      "learning_rate": 6.451189044505104e-06,
      "loss": 0.1905,
      "num_input_tokens_seen": 58071584,
      "step": 30645
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.18548861145973206,
      "learning_rate": 6.44460832269676e-06,
      "loss": 0.0907,
      "num_input_tokens_seen": 58080480,
      "step": 30650
    },
    {
      "epoch": 5.000815727220817,
      "grad_norm": 0.5746557712554932,
      "learning_rate": 6.438030462363001e-06,
      "loss": 0.0475,
      "num_input_tokens_seen": 58090336,
      "step": 30655
    },
    {
      "epoch": 5.001631454441635,
      "grad_norm": 0.7319589853286743,
      "learning_rate": 6.431455464518205e-06,
      "loss": 0.1164,
      "num_input_tokens_seen": 58099552,
      "step": 30660
    },
    {
      "epoch": 5.002447181662452,
      "grad_norm": 0.4630686342716217,
      "learning_rate": 6.424883330176326e-06,
      "loss": 0.1657,
      "num_input_tokens_seen": 58109376,
      "step": 30665
    },
    {
      "epoch": 5.003262908883269,
      "grad_norm": 0.4496445655822754,
      "learning_rate": 6.418314060350864e-06,
      "loss": 0.0611,
      "num_input_tokens_seen": 58119536,
      "step": 30670
    },
    {
      "epoch": 5.0040786361040865,
      "grad_norm": 0.19391830265522003,
      "learning_rate": 6.4117476560548895e-06,
      "loss": 0.0555,
      "num_input_tokens_seen": 58129616,
      "step": 30675
    },
    {
      "epoch": 5.004894363324905,
      "grad_norm": 0.38217365741729736,
      "learning_rate": 6.405184118301016e-06,
      "loss": 0.0931,
      "num_input_tokens_seen": 58138848,
      "step": 30680
    },
    {
      "epoch": 5.005710090545722,
      "grad_norm": 0.1507139503955841,
      "learning_rate": 6.398623448101434e-06,
      "loss": 0.0585,
      "num_input_tokens_seen": 58148512,
      "step": 30685
    },
    {
      "epoch": 5.006525817766539,
      "grad_norm": 1.5247154235839844,
      "learning_rate": 6.392065646467871e-06,
      "loss": 0.1297,
      "num_input_tokens_seen": 58157600,
      "step": 30690
    },
    {
      "epoch": 5.007341544987356,
      "grad_norm": 2.2261006832122803,
      "learning_rate": 6.385510714411632e-06,
      "loss": 0.1158,
      "num_input_tokens_seen": 58168528,
      "step": 30695
    },
    {
      "epoch": 5.008157272208174,
      "grad_norm": 0.6662259101867676,
      "learning_rate": 6.378958652943559e-06,
      "loss": 0.1489,
      "num_input_tokens_seen": 58177632,
      "step": 30700
    },
    {
      "epoch": 5.008972999428991,
      "grad_norm": 0.04912377893924713,
      "learning_rate": 6.3724094630740776e-06,
      "loss": 0.0402,
      "num_input_tokens_seen": 58187488,
      "step": 30705
    },
    {
      "epoch": 5.009788726649808,
      "grad_norm": 0.8196302652359009,
      "learning_rate": 6.365863145813136e-06,
      "loss": 0.3171,
      "num_input_tokens_seen": 58196432,
      "step": 30710
    },
    {
      "epoch": 5.010604453870625,
      "grad_norm": 0.328769326210022,
      "learning_rate": 6.359319702170269e-06,
      "loss": 0.0286,
      "num_input_tokens_seen": 58205072,
      "step": 30715
    },
    {
      "epoch": 5.011420181091443,
      "grad_norm": 0.6087225079536438,
      "learning_rate": 6.352779133154566e-06,
      "loss": 0.0513,
      "num_input_tokens_seen": 58213488,
      "step": 30720
    },
    {
      "epoch": 5.0122359083122605,
      "grad_norm": 0.38762781023979187,
      "learning_rate": 6.346241439774648e-06,
      "loss": 0.1574,
      "num_input_tokens_seen": 58223456,
      "step": 30725
    },
    {
      "epoch": 5.013051635533078,
      "grad_norm": 0.06824149191379547,
      "learning_rate": 6.339706623038716e-06,
      "loss": 0.0731,
      "num_input_tokens_seen": 58232768,
      "step": 30730
    },
    {
      "epoch": 5.013867362753895,
      "grad_norm": 0.3901393413543701,
      "learning_rate": 6.333174683954532e-06,
      "loss": 0.0666,
      "num_input_tokens_seen": 58242496,
      "step": 30735
    },
    {
      "epoch": 5.014683089974713,
      "grad_norm": 0.5383016467094421,
      "learning_rate": 6.326645623529387e-06,
      "loss": 0.0681,
      "num_input_tokens_seen": 58252064,
      "step": 30740
    },
    {
      "epoch": 5.01549881719553,
      "grad_norm": 0.18299968540668488,
      "learning_rate": 6.320119442770156e-06,
      "loss": 0.042,
      "num_input_tokens_seen": 58261008,
      "step": 30745
    },
    {
      "epoch": 5.016314544416347,
      "grad_norm": 0.6419057846069336,
      "learning_rate": 6.313596142683254e-06,
      "loss": 0.0691,
      "num_input_tokens_seen": 58270912,
      "step": 30750
    },
    {
      "epoch": 5.017130271637164,
      "grad_norm": 0.3501022458076477,
      "learning_rate": 6.307075724274647e-06,
      "loss": 0.2523,
      "num_input_tokens_seen": 58280784,
      "step": 30755
    },
    {
      "epoch": 5.017945998857982,
      "grad_norm": 0.0806104987859726,
      "learning_rate": 6.300558188549882e-06,
      "loss": 0.0752,
      "num_input_tokens_seen": 58290768,
      "step": 30760
    },
    {
      "epoch": 5.018761726078799,
      "grad_norm": 1.0174272060394287,
      "learning_rate": 6.29404353651403e-06,
      "loss": 0.128,
      "num_input_tokens_seen": 58301584,
      "step": 30765
    },
    {
      "epoch": 5.0195774532996165,
      "grad_norm": 0.040664322674274445,
      "learning_rate": 6.287531769171737e-06,
      "loss": 0.1036,
      "num_input_tokens_seen": 58310304,
      "step": 30770
    },
    {
      "epoch": 5.020393180520434,
      "grad_norm": 1.4493659734725952,
      "learning_rate": 6.2810228875272045e-06,
      "loss": 0.2128,
      "num_input_tokens_seen": 58320224,
      "step": 30775
    },
    {
      "epoch": 5.021208907741252,
      "grad_norm": 0.40882185101509094,
      "learning_rate": 6.274516892584179e-06,
      "loss": 0.2386,
      "num_input_tokens_seen": 58329056,
      "step": 30780
    },
    {
      "epoch": 5.022024634962069,
      "grad_norm": 0.30906230211257935,
      "learning_rate": 6.268013785345969e-06,
      "loss": 0.3311,
      "num_input_tokens_seen": 58338800,
      "step": 30785
    },
    {
      "epoch": 5.022840362182886,
      "grad_norm": 1.3934099674224854,
      "learning_rate": 6.26151356681543e-06,
      "loss": 0.2614,
      "num_input_tokens_seen": 58348160,
      "step": 30790
    },
    {
      "epoch": 5.023656089403703,
      "grad_norm": 0.15094643831253052,
      "learning_rate": 6.255016237994981e-06,
      "loss": 0.0245,
      "num_input_tokens_seen": 58359328,
      "step": 30795
    },
    {
      "epoch": 5.024471816624521,
      "grad_norm": 0.1248619556427002,
      "learning_rate": 6.248521799886603e-06,
      "loss": 0.0476,
      "num_input_tokens_seen": 58367056,
      "step": 30800
    },
    {
      "epoch": 5.024471816624521,
      "eval_loss": 0.1420515775680542,
      "eval_runtime": 133.2438,
      "eval_samples_per_second": 20.451,
      "eval_steps_per_second": 10.229,
      "num_input_tokens_seen": 58367056,
      "step": 30800
    },
    {
      "epoch": 5.025287543845338,
      "grad_norm": 0.07999234646558762,
      "learning_rate": 6.242030253491798e-06,
      "loss": 0.1389,
      "num_input_tokens_seen": 58376928,
      "step": 30805
    },
    {
      "epoch": 5.026103271066155,
      "grad_norm": 0.5163329839706421,
      "learning_rate": 6.235541599811656e-06,
      "loss": 0.1672,
      "num_input_tokens_seen": 58385504,
      "step": 30810
    },
    {
      "epoch": 5.026918998286972,
      "grad_norm": 0.4247353971004486,
      "learning_rate": 6.229055839846814e-06,
      "loss": 0.1125,
      "num_input_tokens_seen": 58395568,
      "step": 30815
    },
    {
      "epoch": 5.02773472550779,
      "grad_norm": 0.3797476291656494,
      "learning_rate": 6.222572974597455e-06,
      "loss": 0.0578,
      "num_input_tokens_seen": 58403744,
      "step": 30820
    },
    {
      "epoch": 5.028550452728608,
      "grad_norm": 0.08760766685009003,
      "learning_rate": 6.216093005063306e-06,
      "loss": 0.1041,
      "num_input_tokens_seen": 58412720,
      "step": 30825
    },
    {
      "epoch": 5.029366179949425,
      "grad_norm": 1.240206003189087,
      "learning_rate": 6.209615932243678e-06,
      "loss": 0.2286,
      "num_input_tokens_seen": 58422128,
      "step": 30830
    },
    {
      "epoch": 5.030181907170242,
      "grad_norm": 0.6646883487701416,
      "learning_rate": 6.203141757137399e-06,
      "loss": 0.0944,
      "num_input_tokens_seen": 58429632,
      "step": 30835
    },
    {
      "epoch": 5.03099763439106,
      "grad_norm": 0.1822856217622757,
      "learning_rate": 6.196670480742886e-06,
      "loss": 0.0497,
      "num_input_tokens_seen": 58439344,
      "step": 30840
    },
    {
      "epoch": 5.031813361611877,
      "grad_norm": 0.38867756724357605,
      "learning_rate": 6.190202104058074e-06,
      "loss": 0.191,
      "num_input_tokens_seen": 58449488,
      "step": 30845
    },
    {
      "epoch": 5.032629088832694,
      "grad_norm": 1.6812915802001953,
      "learning_rate": 6.183736628080475e-06,
      "loss": 0.1987,
      "num_input_tokens_seen": 58458496,
      "step": 30850
    },
    {
      "epoch": 5.033444816053512,
      "grad_norm": 1.0123635530471802,
      "learning_rate": 6.177274053807155e-06,
      "loss": 0.1892,
      "num_input_tokens_seen": 58467600,
      "step": 30855
    },
    {
      "epoch": 5.034260543274329,
      "grad_norm": 0.38123536109924316,
      "learning_rate": 6.170814382234713e-06,
      "loss": 0.0775,
      "num_input_tokens_seen": 58477056,
      "step": 30860
    },
    {
      "epoch": 5.035076270495146,
      "grad_norm": 1.3090564012527466,
      "learning_rate": 6.16435761435932e-06,
      "loss": 0.2213,
      "num_input_tokens_seen": 58487232,
      "step": 30865
    },
    {
      "epoch": 5.0358919977159635,
      "grad_norm": 1.2883068323135376,
      "learning_rate": 6.157903751176681e-06,
      "loss": 0.0761,
      "num_input_tokens_seen": 58496592,
      "step": 30870
    },
    {
      "epoch": 5.0367077249367815,
      "grad_norm": 0.10645325481891632,
      "learning_rate": 6.151452793682066e-06,
      "loss": 0.146,
      "num_input_tokens_seen": 58506960,
      "step": 30875
    },
    {
      "epoch": 5.037523452157599,
      "grad_norm": 1.7566983699798584,
      "learning_rate": 6.145004742870305e-06,
      "loss": 0.155,
      "num_input_tokens_seen": 58515920,
      "step": 30880
    },
    {
      "epoch": 5.038339179378416,
      "grad_norm": 0.11250407993793488,
      "learning_rate": 6.138559599735752e-06,
      "loss": 0.1284,
      "num_input_tokens_seen": 58526176,
      "step": 30885
    },
    {
      "epoch": 5.039154906599233,
      "grad_norm": 0.8721154928207397,
      "learning_rate": 6.132117365272344e-06,
      "loss": 0.1763,
      "num_input_tokens_seen": 58535264,
      "step": 30890
    },
    {
      "epoch": 5.039970633820051,
      "grad_norm": 1.1484849452972412,
      "learning_rate": 6.125678040473545e-06,
      "loss": 0.1819,
      "num_input_tokens_seen": 58544560,
      "step": 30895
    },
    {
      "epoch": 5.040786361040868,
      "grad_norm": 0.11620600521564484,
      "learning_rate": 6.1192416263323755e-06,
      "loss": 0.0865,
      "num_input_tokens_seen": 58553312,
      "step": 30900
    },
    {
      "epoch": 5.041602088261685,
      "grad_norm": 0.6299225687980652,
      "learning_rate": 6.112808123841424e-06,
      "loss": 0.097,
      "num_input_tokens_seen": 58562560,
      "step": 30905
    },
    {
      "epoch": 5.042417815482502,
      "grad_norm": 0.22408413887023926,
      "learning_rate": 6.106377533992805e-06,
      "loss": 0.1032,
      "num_input_tokens_seen": 58571904,
      "step": 30910
    },
    {
      "epoch": 5.04323354270332,
      "grad_norm": 0.8343403935432434,
      "learning_rate": 6.099949857778204e-06,
      "loss": 0.1733,
      "num_input_tokens_seen": 58581296,
      "step": 30915
    },
    {
      "epoch": 5.0440492699241375,
      "grad_norm": 0.30404162406921387,
      "learning_rate": 6.093525096188852e-06,
      "loss": 0.1684,
      "num_input_tokens_seen": 58591344,
      "step": 30920
    },
    {
      "epoch": 5.044864997144955,
      "grad_norm": 0.477071613073349,
      "learning_rate": 6.087103250215518e-06,
      "loss": 0.2057,
      "num_input_tokens_seen": 58601184,
      "step": 30925
    },
    {
      "epoch": 5.045680724365772,
      "grad_norm": 0.15110909938812256,
      "learning_rate": 6.080684320848537e-06,
      "loss": 0.047,
      "num_input_tokens_seen": 58610480,
      "step": 30930
    },
    {
      "epoch": 5.04649645158659,
      "grad_norm": 0.45370879769325256,
      "learning_rate": 6.074268309077794e-06,
      "loss": 0.189,
      "num_input_tokens_seen": 58620400,
      "step": 30935
    },
    {
      "epoch": 5.047312178807407,
      "grad_norm": 0.45816296339035034,
      "learning_rate": 6.067855215892709e-06,
      "loss": 0.1461,
      "num_input_tokens_seen": 58629744,
      "step": 30940
    },
    {
      "epoch": 5.048127906028224,
      "grad_norm": 0.5242000222206116,
      "learning_rate": 6.061445042282271e-06,
      "loss": 0.0912,
      "num_input_tokens_seen": 58640608,
      "step": 30945
    },
    {
      "epoch": 5.048943633249041,
      "grad_norm": 1.2390155792236328,
      "learning_rate": 6.055037789234999e-06,
      "loss": 0.1884,
      "num_input_tokens_seen": 58649984,
      "step": 30950
    },
    {
      "epoch": 5.049759360469859,
      "grad_norm": 0.39416107535362244,
      "learning_rate": 6.048633457738975e-06,
      "loss": 0.0497,
      "num_input_tokens_seen": 58659168,
      "step": 30955
    },
    {
      "epoch": 5.050575087690676,
      "grad_norm": 0.3767966628074646,
      "learning_rate": 6.042232048781837e-06,
      "loss": 0.0608,
      "num_input_tokens_seen": 58668336,
      "step": 30960
    },
    {
      "epoch": 5.051390814911493,
      "grad_norm": 1.6506789922714233,
      "learning_rate": 6.035833563350757e-06,
      "loss": 0.1428,
      "num_input_tokens_seen": 58679424,
      "step": 30965
    },
    {
      "epoch": 5.052206542132311,
      "grad_norm": 0.33534669876098633,
      "learning_rate": 6.0294380024324525e-06,
      "loss": 0.06,
      "num_input_tokens_seen": 58688832,
      "step": 30970
    },
    {
      "epoch": 5.053022269353129,
      "grad_norm": 0.9842577576637268,
      "learning_rate": 6.023045367013213e-06,
      "loss": 0.1992,
      "num_input_tokens_seen": 58699200,
      "step": 30975
    },
    {
      "epoch": 5.053837996573946,
      "grad_norm": 0.5176402926445007,
      "learning_rate": 6.016655658078851e-06,
      "loss": 0.0979,
      "num_input_tokens_seen": 58708832,
      "step": 30980
    },
    {
      "epoch": 5.054653723794763,
      "grad_norm": 1.8387277126312256,
      "learning_rate": 6.010268876614753e-06,
      "loss": 0.2004,
      "num_input_tokens_seen": 58718160,
      "step": 30985
    },
    {
      "epoch": 5.05546945101558,
      "grad_norm": 0.6327978372573853,
      "learning_rate": 6.0038850236058266e-06,
      "loss": 0.1044,
      "num_input_tokens_seen": 58727504,
      "step": 30990
    },
    {
      "epoch": 5.056285178236398,
      "grad_norm": 0.39929527044296265,
      "learning_rate": 5.997504100036549e-06,
      "loss": 0.0675,
      "num_input_tokens_seen": 58737664,
      "step": 30995
    },
    {
      "epoch": 5.057100905457215,
      "grad_norm": 0.4899643659591675,
      "learning_rate": 5.991126106890949e-06,
      "loss": 0.1684,
      "num_input_tokens_seen": 58746720,
      "step": 31000
    },
    {
      "epoch": 5.057100905457215,
      "eval_loss": 0.14255774021148682,
      "eval_runtime": 133.2664,
      "eval_samples_per_second": 20.448,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 58746720,
      "step": 31000
    },
    {
      "epoch": 5.057916632678032,
      "grad_norm": 0.2753770649433136,
      "learning_rate": 5.984751045152576e-06,
      "loss": 0.2352,
      "num_input_tokens_seen": 58755504,
      "step": 31005
    },
    {
      "epoch": 5.058732359898849,
      "grad_norm": 0.3791089951992035,
      "learning_rate": 5.978378915804553e-06,
      "loss": 0.2947,
      "num_input_tokens_seen": 58765520,
      "step": 31010
    },
    {
      "epoch": 5.059548087119667,
      "grad_norm": 0.5137808918952942,
      "learning_rate": 5.972009719829547e-06,
      "loss": 0.162,
      "num_input_tokens_seen": 58775392,
      "step": 31015
    },
    {
      "epoch": 5.0603638143404845,
      "grad_norm": 0.4258744716644287,
      "learning_rate": 5.965643458209755e-06,
      "loss": 0.1047,
      "num_input_tokens_seen": 58785024,
      "step": 31020
    },
    {
      "epoch": 5.061179541561302,
      "grad_norm": 0.2177211493253708,
      "learning_rate": 5.95928013192695e-06,
      "loss": 0.0309,
      "num_input_tokens_seen": 58793280,
      "step": 31025
    },
    {
      "epoch": 5.06199526878212,
      "grad_norm": 0.07463420927524567,
      "learning_rate": 5.952919741962423e-06,
      "loss": 0.1506,
      "num_input_tokens_seen": 58802240,
      "step": 31030
    },
    {
      "epoch": 5.062810996002937,
      "grad_norm": 0.17710484564304352,
      "learning_rate": 5.946562289297042e-06,
      "loss": 0.1557,
      "num_input_tokens_seen": 58812704,
      "step": 31035
    },
    {
      "epoch": 5.063626723223754,
      "grad_norm": 0.7911497950553894,
      "learning_rate": 5.9402077749111855e-06,
      "loss": 0.245,
      "num_input_tokens_seen": 58820640,
      "step": 31040
    },
    {
      "epoch": 5.064442450444571,
      "grad_norm": 2.2887661457061768,
      "learning_rate": 5.933856199784821e-06,
      "loss": 0.2238,
      "num_input_tokens_seen": 58830176,
      "step": 31045
    },
    {
      "epoch": 5.065258177665389,
      "grad_norm": 0.5741758346557617,
      "learning_rate": 5.927507564897419e-06,
      "loss": 0.183,
      "num_input_tokens_seen": 58839632,
      "step": 31050
    },
    {
      "epoch": 5.066073904886206,
      "grad_norm": 0.350072979927063,
      "learning_rate": 5.9211618712280395e-06,
      "loss": 0.0211,
      "num_input_tokens_seen": 58849376,
      "step": 31055
    },
    {
      "epoch": 5.066889632107023,
      "grad_norm": 1.1206090450286865,
      "learning_rate": 5.914819119755255e-06,
      "loss": 0.1308,
      "num_input_tokens_seen": 58858640,
      "step": 31060
    },
    {
      "epoch": 5.0677053593278405,
      "grad_norm": 0.10675591230392456,
      "learning_rate": 5.908479311457205e-06,
      "loss": 0.0395,
      "num_input_tokens_seen": 58868416,
      "step": 31065
    },
    {
      "epoch": 5.0685210865486585,
      "grad_norm": 0.14132127165794373,
      "learning_rate": 5.902142447311559e-06,
      "loss": 0.0547,
      "num_input_tokens_seen": 58877888,
      "step": 31070
    },
    {
      "epoch": 5.069336813769476,
      "grad_norm": 0.11189424991607666,
      "learning_rate": 5.895808528295546e-06,
      "loss": 0.0874,
      "num_input_tokens_seen": 58887808,
      "step": 31075
    },
    {
      "epoch": 5.070152540990293,
      "grad_norm": 0.45630979537963867,
      "learning_rate": 5.889477555385941e-06,
      "loss": 0.0957,
      "num_input_tokens_seen": 58897424,
      "step": 31080
    },
    {
      "epoch": 5.07096826821111,
      "grad_norm": 0.0887465626001358,
      "learning_rate": 5.883149529559051e-06,
      "loss": 0.043,
      "num_input_tokens_seen": 58906512,
      "step": 31085
    },
    {
      "epoch": 5.071783995431928,
      "grad_norm": 0.0543363094329834,
      "learning_rate": 5.876824451790738e-06,
      "loss": 0.0653,
      "num_input_tokens_seen": 58915904,
      "step": 31090
    },
    {
      "epoch": 5.072599722652745,
      "grad_norm": 0.25197556614875793,
      "learning_rate": 5.87050232305642e-06,
      "loss": 0.2509,
      "num_input_tokens_seen": 58925664,
      "step": 31095
    },
    {
      "epoch": 5.073415449873562,
      "grad_norm": 0.563656747341156,
      "learning_rate": 5.864183144331034e-06,
      "loss": 0.1516,
      "num_input_tokens_seen": 58935920,
      "step": 31100
    },
    {
      "epoch": 5.074231177094379,
      "grad_norm": 0.868454098701477,
      "learning_rate": 5.857866916589089e-06,
      "loss": 0.1832,
      "num_input_tokens_seen": 58946064,
      "step": 31105
    },
    {
      "epoch": 5.075046904315197,
      "grad_norm": 0.1492270678281784,
      "learning_rate": 5.8515536408046216e-06,
      "loss": 0.1353,
      "num_input_tokens_seen": 58955712,
      "step": 31110
    },
    {
      "epoch": 5.0758626315360145,
      "grad_norm": 0.28556931018829346,
      "learning_rate": 5.845243317951208e-06,
      "loss": 0.138,
      "num_input_tokens_seen": 58964832,
      "step": 31115
    },
    {
      "epoch": 5.076678358756832,
      "grad_norm": 1.6939390897750854,
      "learning_rate": 5.838935949001997e-06,
      "loss": 0.1658,
      "num_input_tokens_seen": 58973840,
      "step": 31120
    },
    {
      "epoch": 5.077494085977649,
      "grad_norm": 0.09482202678918839,
      "learning_rate": 5.8326315349296476e-06,
      "loss": 0.0611,
      "num_input_tokens_seen": 58982896,
      "step": 31125
    },
    {
      "epoch": 5.078309813198467,
      "grad_norm": 0.6152809262275696,
      "learning_rate": 5.826330076706396e-06,
      "loss": 0.0451,
      "num_input_tokens_seen": 58991472,
      "step": 31130
    },
    {
      "epoch": 5.079125540419284,
      "grad_norm": 0.1700243055820465,
      "learning_rate": 5.820031575303988e-06,
      "loss": 0.2408,
      "num_input_tokens_seen": 59000736,
      "step": 31135
    },
    {
      "epoch": 5.079941267640101,
      "grad_norm": 0.32907459139823914,
      "learning_rate": 5.813736031693745e-06,
      "loss": 0.0711,
      "num_input_tokens_seen": 59009568,
      "step": 31140
    },
    {
      "epoch": 5.080756994860918,
      "grad_norm": 1.4061627388000488,
      "learning_rate": 5.807443446846522e-06,
      "loss": 0.0917,
      "num_input_tokens_seen": 59020032,
      "step": 31145
    },
    {
      "epoch": 5.081572722081736,
      "grad_norm": 0.9919489026069641,
      "learning_rate": 5.801153821732699e-06,
      "loss": 0.1661,
      "num_input_tokens_seen": 59029056,
      "step": 31150
    },
    {
      "epoch": 5.082388449302553,
      "grad_norm": 0.4620792269706726,
      "learning_rate": 5.794867157322229e-06,
      "loss": 0.1438,
      "num_input_tokens_seen": 59038384,
      "step": 31155
    },
    {
      "epoch": 5.08320417652337,
      "grad_norm": 0.7535696029663086,
      "learning_rate": 5.788583454584593e-06,
      "loss": 0.1281,
      "num_input_tokens_seen": 59047760,
      "step": 31160
    },
    {
      "epoch": 5.0840199037441876,
      "grad_norm": 0.5632907748222351,
      "learning_rate": 5.7823027144888075e-06,
      "loss": 0.1115,
      "num_input_tokens_seen": 59056080,
      "step": 31165
    },
    {
      "epoch": 5.084835630965006,
      "grad_norm": 2.1748127937316895,
      "learning_rate": 5.776024938003455e-06,
      "loss": 0.3028,
      "num_input_tokens_seen": 59065184,
      "step": 31170
    },
    {
      "epoch": 5.085651358185823,
      "grad_norm": 0.32575559616088867,
      "learning_rate": 5.7697501260966345e-06,
      "loss": 0.0664,
      "num_input_tokens_seen": 59076048,
      "step": 31175
    },
    {
      "epoch": 5.08646708540664,
      "grad_norm": 0.11420565098524094,
      "learning_rate": 5.7634782797360145e-06,
      "loss": 0.1097,
      "num_input_tokens_seen": 59085856,
      "step": 31180
    },
    {
      "epoch": 5.087282812627457,
      "grad_norm": 0.3856383264064789,
      "learning_rate": 5.757209399888777e-06,
      "loss": 0.0802,
      "num_input_tokens_seen": 59094464,
      "step": 31185
    },
    {
      "epoch": 5.088098539848275,
      "grad_norm": 0.19563129544258118,
      "learning_rate": 5.750943487521679e-06,
      "loss": 0.0587,
      "num_input_tokens_seen": 59104384,
      "step": 31190
    },
    {
      "epoch": 5.088914267069092,
      "grad_norm": 0.29282042384147644,
      "learning_rate": 5.744680543600986e-06,
      "loss": 0.2262,
      "num_input_tokens_seen": 59114048,
      "step": 31195
    },
    {
      "epoch": 5.089729994289909,
      "grad_norm": 0.07667087018489838,
      "learning_rate": 5.738420569092537e-06,
      "loss": 0.0495,
      "num_input_tokens_seen": 59124272,
      "step": 31200
    },
    {
      "epoch": 5.089729994289909,
      "eval_loss": 0.1413251906633377,
      "eval_runtime": 133.3306,
      "eval_samples_per_second": 20.438,
      "eval_steps_per_second": 10.223,
      "num_input_tokens_seen": 59124272,
      "step": 31200
    },
    {
      "epoch": 5.090545721510727,
      "grad_norm": 1.0827852487564087,
      "learning_rate": 5.732163564961684e-06,
      "loss": 0.1658,
      "num_input_tokens_seen": 59134464,
      "step": 31205
    },
    {
      "epoch": 5.091361448731544,
      "grad_norm": 0.06992926448583603,
      "learning_rate": 5.725909532173354e-06,
      "loss": 0.1685,
      "num_input_tokens_seen": 59143552,
      "step": 31210
    },
    {
      "epoch": 5.0921771759523615,
      "grad_norm": 0.13747629523277283,
      "learning_rate": 5.719658471691977e-06,
      "loss": 0.22,
      "num_input_tokens_seen": 59153600,
      "step": 31215
    },
    {
      "epoch": 5.092992903173179,
      "grad_norm": 0.5720354318618774,
      "learning_rate": 5.71341038448156e-06,
      "loss": 0.2253,
      "num_input_tokens_seen": 59163648,
      "step": 31220
    },
    {
      "epoch": 5.093808630393997,
      "grad_norm": 0.19113023579120636,
      "learning_rate": 5.707165271505635e-06,
      "loss": 0.0963,
      "num_input_tokens_seen": 59173536,
      "step": 31225
    },
    {
      "epoch": 5.094624357614814,
      "grad_norm": 0.08316461741924286,
      "learning_rate": 5.700923133727271e-06,
      "loss": 0.0533,
      "num_input_tokens_seen": 59183440,
      "step": 31230
    },
    {
      "epoch": 5.095440084835631,
      "grad_norm": 1.0895212888717651,
      "learning_rate": 5.694683972109083e-06,
      "loss": 0.1484,
      "num_input_tokens_seen": 59192816,
      "step": 31235
    },
    {
      "epoch": 5.096255812056448,
      "grad_norm": 0.7573392391204834,
      "learning_rate": 5.688447787613241e-06,
      "loss": 0.0629,
      "num_input_tokens_seen": 59201952,
      "step": 31240
    },
    {
      "epoch": 5.097071539277266,
      "grad_norm": 0.9284729957580566,
      "learning_rate": 5.6822145812014285e-06,
      "loss": 0.1194,
      "num_input_tokens_seen": 59211536,
      "step": 31245
    },
    {
      "epoch": 5.097887266498083,
      "grad_norm": 0.9982293248176575,
      "learning_rate": 5.675984353834896e-06,
      "loss": 0.1537,
      "num_input_tokens_seen": 59220752,
      "step": 31250
    },
    {
      "epoch": 5.0987029937189,
      "grad_norm": 0.40911245346069336,
      "learning_rate": 5.66975710647441e-06,
      "loss": 0.0871,
      "num_input_tokens_seen": 59230976,
      "step": 31255
    },
    {
      "epoch": 5.0995187209397175,
      "grad_norm": 0.4579318165779114,
      "learning_rate": 5.663532840080304e-06,
      "loss": 0.0928,
      "num_input_tokens_seen": 59241008,
      "step": 31260
    },
    {
      "epoch": 5.1003344481605355,
      "grad_norm": 0.8929829597473145,
      "learning_rate": 5.6573115556124325e-06,
      "loss": 0.1271,
      "num_input_tokens_seen": 59250080,
      "step": 31265
    },
    {
      "epoch": 5.101150175381353,
      "grad_norm": 0.03078165091574192,
      "learning_rate": 5.651093254030185e-06,
      "loss": 0.1282,
      "num_input_tokens_seen": 59258800,
      "step": 31270
    },
    {
      "epoch": 5.10196590260217,
      "grad_norm": 0.5059190988540649,
      "learning_rate": 5.644877936292514e-06,
      "loss": 0.2232,
      "num_input_tokens_seen": 59267792,
      "step": 31275
    },
    {
      "epoch": 5.102781629822987,
      "grad_norm": 0.48210442066192627,
      "learning_rate": 5.638665603357901e-06,
      "loss": 0.1269,
      "num_input_tokens_seen": 59276960,
      "step": 31280
    },
    {
      "epoch": 5.103597357043805,
      "grad_norm": 0.09358509629964828,
      "learning_rate": 5.632456256184357e-06,
      "loss": 0.1002,
      "num_input_tokens_seen": 59286576,
      "step": 31285
    },
    {
      "epoch": 5.104413084264622,
      "grad_norm": 0.8048520684242249,
      "learning_rate": 5.626249895729452e-06,
      "loss": 0.0795,
      "num_input_tokens_seen": 59294752,
      "step": 31290
    },
    {
      "epoch": 5.105228811485439,
      "grad_norm": 2.7675764560699463,
      "learning_rate": 5.620046522950273e-06,
      "loss": 0.2566,
      "num_input_tokens_seen": 59304304,
      "step": 31295
    },
    {
      "epoch": 5.106044538706256,
      "grad_norm": 1.871424674987793,
      "learning_rate": 5.613846138803464e-06,
      "loss": 0.1949,
      "num_input_tokens_seen": 59313520,
      "step": 31300
    },
    {
      "epoch": 5.106860265927074,
      "grad_norm": 0.15106363594532013,
      "learning_rate": 5.607648744245206e-06,
      "loss": 0.0266,
      "num_input_tokens_seen": 59323376,
      "step": 31305
    },
    {
      "epoch": 5.1076759931478914,
      "grad_norm": 0.23449909687042236,
      "learning_rate": 5.601454340231207e-06,
      "loss": 0.1063,
      "num_input_tokens_seen": 59332976,
      "step": 31310
    },
    {
      "epoch": 5.108491720368709,
      "grad_norm": 0.6924178600311279,
      "learning_rate": 5.595262927716724e-06,
      "loss": 0.0635,
      "num_input_tokens_seen": 59342336,
      "step": 31315
    },
    {
      "epoch": 5.109307447589526,
      "grad_norm": 0.6864960193634033,
      "learning_rate": 5.589074507656561e-06,
      "loss": 0.1369,
      "num_input_tokens_seen": 59352688,
      "step": 31320
    },
    {
      "epoch": 5.110123174810344,
      "grad_norm": 0.2712657153606415,
      "learning_rate": 5.582889081005044e-06,
      "loss": 0.209,
      "num_input_tokens_seen": 59363088,
      "step": 31325
    },
    {
      "epoch": 5.110938902031161,
      "grad_norm": 1.3347699642181396,
      "learning_rate": 5.5767066487160316e-06,
      "loss": 0.2234,
      "num_input_tokens_seen": 59371904,
      "step": 31330
    },
    {
      "epoch": 5.111754629251978,
      "grad_norm": 0.20370706915855408,
      "learning_rate": 5.570527211742949e-06,
      "loss": 0.1723,
      "num_input_tokens_seen": 59381600,
      "step": 31335
    },
    {
      "epoch": 5.112570356472795,
      "grad_norm": 0.24175061285495758,
      "learning_rate": 5.564350771038731e-06,
      "loss": 0.1523,
      "num_input_tokens_seen": 59390704,
      "step": 31340
    },
    {
      "epoch": 5.113386083693613,
      "grad_norm": 0.6857295036315918,
      "learning_rate": 5.558177327555875e-06,
      "loss": 0.2292,
      "num_input_tokens_seen": 59400208,
      "step": 31345
    },
    {
      "epoch": 5.11420181091443,
      "grad_norm": 1.4146878719329834,
      "learning_rate": 5.552006882246388e-06,
      "loss": 0.2806,
      "num_input_tokens_seen": 59410384,
      "step": 31350
    },
    {
      "epoch": 5.115017538135247,
      "grad_norm": 2.008143424987793,
      "learning_rate": 5.545839436061839e-06,
      "loss": 0.1573,
      "num_input_tokens_seen": 59420400,
      "step": 31355
    },
    {
      "epoch": 5.1158332653560645,
      "grad_norm": 1.7152334451675415,
      "learning_rate": 5.539674989953331e-06,
      "loss": 0.2717,
      "num_input_tokens_seen": 59430688,
      "step": 31360
    },
    {
      "epoch": 5.1166489925768825,
      "grad_norm": 3.3237857818603516,
      "learning_rate": 5.533513544871488e-06,
      "loss": 0.2733,
      "num_input_tokens_seen": 59440400,
      "step": 31365
    },
    {
      "epoch": 5.1174647197977,
      "grad_norm": 0.08056993037462234,
      "learning_rate": 5.527355101766493e-06,
      "loss": 0.1318,
      "num_input_tokens_seen": 59449248,
      "step": 31370
    },
    {
      "epoch": 5.118280447018517,
      "grad_norm": 0.6845649480819702,
      "learning_rate": 5.521199661588044e-06,
      "loss": 0.1702,
      "num_input_tokens_seen": 59457712,
      "step": 31375
    },
    {
      "epoch": 5.119096174239334,
      "grad_norm": 0.38527315855026245,
      "learning_rate": 5.5150472252853944e-06,
      "loss": 0.1217,
      "num_input_tokens_seen": 59467184,
      "step": 31380
    },
    {
      "epoch": 5.119911901460152,
      "grad_norm": 1.1010944843292236,
      "learning_rate": 5.50889779380733e-06,
      "loss": 0.164,
      "num_input_tokens_seen": 59476496,
      "step": 31385
    },
    {
      "epoch": 5.120727628680969,
      "grad_norm": 0.7866716980934143,
      "learning_rate": 5.5027513681021605e-06,
      "loss": 0.2478,
      "num_input_tokens_seen": 59486064,
      "step": 31390
    },
    {
      "epoch": 5.121543355901786,
      "grad_norm": 0.27820566296577454,
      "learning_rate": 5.4966079491177545e-06,
      "loss": 0.0912,
      "num_input_tokens_seen": 59495552,
      "step": 31395
    },
    {
      "epoch": 5.122359083122603,
      "grad_norm": 0.4324628710746765,
      "learning_rate": 5.490467537801491e-06,
      "loss": 0.0825,
      "num_input_tokens_seen": 59504688,
      "step": 31400
    },
    {
      "epoch": 5.122359083122603,
      "eval_loss": 0.14067649841308594,
      "eval_runtime": 133.3146,
      "eval_samples_per_second": 20.44,
      "eval_steps_per_second": 10.224,
      "num_input_tokens_seen": 59504688,
      "step": 31400
    },
    {
      "epoch": 5.123174810343421,
      "grad_norm": 0.7231076955795288,
      "learning_rate": 5.484330135100313e-06,
      "loss": 0.042,
      "num_input_tokens_seen": 59513408,
      "step": 31405
    },
    {
      "epoch": 5.1239905375642385,
      "grad_norm": 0.1335248053073883,
      "learning_rate": 5.4781957419606785e-06,
      "loss": 0.2343,
      "num_input_tokens_seen": 59523504,
      "step": 31410
    },
    {
      "epoch": 5.124806264785056,
      "grad_norm": 0.1810694932937622,
      "learning_rate": 5.472064359328577e-06,
      "loss": 0.0327,
      "num_input_tokens_seen": 59533552,
      "step": 31415
    },
    {
      "epoch": 5.125621992005874,
      "grad_norm": 0.04057419300079346,
      "learning_rate": 5.4659359881495565e-06,
      "loss": 0.148,
      "num_input_tokens_seen": 59543600,
      "step": 31420
    },
    {
      "epoch": 5.126437719226691,
      "grad_norm": 0.06053910776972771,
      "learning_rate": 5.4598106293686916e-06,
      "loss": 0.0513,
      "num_input_tokens_seen": 59553024,
      "step": 31425
    },
    {
      "epoch": 5.127253446447508,
      "grad_norm": 1.443156361579895,
      "learning_rate": 5.45368828393058e-06,
      "loss": 0.285,
      "num_input_tokens_seen": 59560880,
      "step": 31430
    },
    {
      "epoch": 5.128069173668325,
      "grad_norm": 0.23352082073688507,
      "learning_rate": 5.44756895277937e-06,
      "loss": 0.0873,
      "num_input_tokens_seen": 59570096,
      "step": 31435
    },
    {
      "epoch": 5.128884900889143,
      "grad_norm": 0.10538055002689362,
      "learning_rate": 5.441452636858746e-06,
      "loss": 0.3134,
      "num_input_tokens_seen": 59578672,
      "step": 31440
    },
    {
      "epoch": 5.12970062810996,
      "grad_norm": 0.9899752736091614,
      "learning_rate": 5.435339337111905e-06,
      "loss": 0.1369,
      "num_input_tokens_seen": 59588112,
      "step": 31445
    },
    {
      "epoch": 5.130516355330777,
      "grad_norm": 0.14701955020427704,
      "learning_rate": 5.42922905448161e-06,
      "loss": 0.0458,
      "num_input_tokens_seen": 59596896,
      "step": 31450
    },
    {
      "epoch": 5.1313320825515945,
      "grad_norm": 0.5242257118225098,
      "learning_rate": 5.423121789910129e-06,
      "loss": 0.2477,
      "num_input_tokens_seen": 59604960,
      "step": 31455
    },
    {
      "epoch": 5.1321478097724125,
      "grad_norm": 0.5017930269241333,
      "learning_rate": 5.417017544339287e-06,
      "loss": 0.0775,
      "num_input_tokens_seen": 59612240,
      "step": 31460
    },
    {
      "epoch": 5.13296353699323,
      "grad_norm": 0.24694229662418365,
      "learning_rate": 5.410916318710443e-06,
      "loss": 0.1324,
      "num_input_tokens_seen": 59622288,
      "step": 31465
    },
    {
      "epoch": 5.133779264214047,
      "grad_norm": 0.04171046242117882,
      "learning_rate": 5.404818113964466e-06,
      "loss": 0.1932,
      "num_input_tokens_seen": 59631296,
      "step": 31470
    },
    {
      "epoch": 5.134594991434864,
      "grad_norm": 0.07311860471963882,
      "learning_rate": 5.398722931041792e-06,
      "loss": 0.1431,
      "num_input_tokens_seen": 59641664,
      "step": 31475
    },
    {
      "epoch": 5.135410718655682,
      "grad_norm": 0.3250354826450348,
      "learning_rate": 5.392630770882367e-06,
      "loss": 0.2477,
      "num_input_tokens_seen": 59652416,
      "step": 31480
    },
    {
      "epoch": 5.136226445876499,
      "grad_norm": 0.07375846803188324,
      "learning_rate": 5.3865416344256705e-06,
      "loss": 0.1381,
      "num_input_tokens_seen": 59661840,
      "step": 31485
    },
    {
      "epoch": 5.137042173097316,
      "grad_norm": 1.0309395790100098,
      "learning_rate": 5.380455522610742e-06,
      "loss": 0.1846,
      "num_input_tokens_seen": 59670032,
      "step": 31490
    },
    {
      "epoch": 5.137857900318133,
      "grad_norm": 1.389022946357727,
      "learning_rate": 5.374372436376116e-06,
      "loss": 0.1585,
      "num_input_tokens_seen": 59679264,
      "step": 31495
    },
    {
      "epoch": 5.138673627538951,
      "grad_norm": 0.44190099835395813,
      "learning_rate": 5.368292376659895e-06,
      "loss": 0.1574,
      "num_input_tokens_seen": 59688640,
      "step": 31500
    },
    {
      "epoch": 5.139489354759768,
      "grad_norm": 0.7629227638244629,
      "learning_rate": 5.362215344399701e-06,
      "loss": 0.1597,
      "num_input_tokens_seen": 59697216,
      "step": 31505
    },
    {
      "epoch": 5.1403050819805856,
      "grad_norm": 0.7614008188247681,
      "learning_rate": 5.356141340532678e-06,
      "loss": 0.0894,
      "num_input_tokens_seen": 59706672,
      "step": 31510
    },
    {
      "epoch": 5.141120809201403,
      "grad_norm": 0.5112019777297974,
      "learning_rate": 5.350070365995522e-06,
      "loss": 0.0344,
      "num_input_tokens_seen": 59716320,
      "step": 31515
    },
    {
      "epoch": 5.141936536422221,
      "grad_norm": 0.5115207433700562,
      "learning_rate": 5.344002421724459e-06,
      "loss": 0.2873,
      "num_input_tokens_seen": 59726560,
      "step": 31520
    },
    {
      "epoch": 5.142752263643038,
      "grad_norm": 0.4668665826320648,
      "learning_rate": 5.337937508655228e-06,
      "loss": 0.0528,
      "num_input_tokens_seen": 59735776,
      "step": 31525
    },
    {
      "epoch": 5.143567990863855,
      "grad_norm": 0.9363847374916077,
      "learning_rate": 5.331875627723126e-06,
      "loss": 0.1251,
      "num_input_tokens_seen": 59744512,
      "step": 31530
    },
    {
      "epoch": 5.144383718084672,
      "grad_norm": 0.1625945121049881,
      "learning_rate": 5.325816779862963e-06,
      "loss": 0.1595,
      "num_input_tokens_seen": 59754400,
      "step": 31535
    },
    {
      "epoch": 5.14519944530549,
      "grad_norm": 0.09168302267789841,
      "learning_rate": 5.319760966009102e-06,
      "loss": 0.1514,
      "num_input_tokens_seen": 59763456,
      "step": 31540
    },
    {
      "epoch": 5.146015172526307,
      "grad_norm": 1.6249825954437256,
      "learning_rate": 5.3137081870954096e-06,
      "loss": 0.2146,
      "num_input_tokens_seen": 59773648,
      "step": 31545
    },
    {
      "epoch": 5.146830899747124,
      "grad_norm": 0.4800652265548706,
      "learning_rate": 5.307658444055313e-06,
      "loss": 0.2164,
      "num_input_tokens_seen": 59783504,
      "step": 31550
    },
    {
      "epoch": 5.1476466269679415,
      "grad_norm": 0.10638198256492615,
      "learning_rate": 5.301611737821749e-06,
      "loss": 0.1349,
      "num_input_tokens_seen": 59793456,
      "step": 31555
    },
    {
      "epoch": 5.1484623541887595,
      "grad_norm": 0.9040229916572571,
      "learning_rate": 5.295568069327206e-06,
      "loss": 0.0426,
      "num_input_tokens_seen": 59802400,
      "step": 31560
    },
    {
      "epoch": 5.149278081409577,
      "grad_norm": 0.19000348448753357,
      "learning_rate": 5.289527439503683e-06,
      "loss": 0.1045,
      "num_input_tokens_seen": 59811552,
      "step": 31565
    },
    {
      "epoch": 5.150093808630394,
      "grad_norm": 0.2863769233226776,
      "learning_rate": 5.28348984928273e-06,
      "loss": 0.088,
      "num_input_tokens_seen": 59820560,
      "step": 31570
    },
    {
      "epoch": 5.150909535851211,
      "grad_norm": 0.6241189241409302,
      "learning_rate": 5.27745529959541e-06,
      "loss": 0.1798,
      "num_input_tokens_seen": 59830960,
      "step": 31575
    },
    {
      "epoch": 5.151725263072029,
      "grad_norm": 0.3773447871208191,
      "learning_rate": 5.271423791372335e-06,
      "loss": 0.1589,
      "num_input_tokens_seen": 59838784,
      "step": 31580
    },
    {
      "epoch": 5.152540990292846,
      "grad_norm": 0.17781585454940796,
      "learning_rate": 5.26539532554364e-06,
      "loss": 0.027,
      "num_input_tokens_seen": 59848432,
      "step": 31585
    },
    {
      "epoch": 5.153356717513663,
      "grad_norm": 1.4129319190979004,
      "learning_rate": 5.25936990303898e-06,
      "loss": 0.1519,
      "num_input_tokens_seen": 59857616,
      "step": 31590
    },
    {
      "epoch": 5.154172444734481,
      "grad_norm": 0.4614119827747345,
      "learning_rate": 5.253347524787555e-06,
      "loss": 0.1064,
      "num_input_tokens_seen": 59867232,
      "step": 31595
    },
    {
      "epoch": 5.154988171955298,
      "grad_norm": 0.4691604673862457,
      "learning_rate": 5.2473281917181035e-06,
      "loss": 0.1022,
      "num_input_tokens_seen": 59875840,
      "step": 31600
    },
    {
      "epoch": 5.154988171955298,
      "eval_loss": 0.14016859233379364,
      "eval_runtime": 133.341,
      "eval_samples_per_second": 20.436,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 59875840,
      "step": 31600
    },
    {
      "epoch": 5.1558038991761155,
      "grad_norm": 0.23704537749290466,
      "learning_rate": 5.241311904758864e-06,
      "loss": 0.0328,
      "num_input_tokens_seen": 59885616,
      "step": 31605
    },
    {
      "epoch": 5.156619626396933,
      "grad_norm": 0.29082170128822327,
      "learning_rate": 5.23529866483764e-06,
      "loss": 0.076,
      "num_input_tokens_seen": 59893936,
      "step": 31610
    },
    {
      "epoch": 5.157435353617751,
      "grad_norm": 0.2149481326341629,
      "learning_rate": 5.229288472881732e-06,
      "loss": 0.2379,
      "num_input_tokens_seen": 59902272,
      "step": 31615
    },
    {
      "epoch": 5.158251080838568,
      "grad_norm": 0.17985667288303375,
      "learning_rate": 5.2232813298180025e-06,
      "loss": 0.1411,
      "num_input_tokens_seen": 59911136,
      "step": 31620
    },
    {
      "epoch": 5.159066808059385,
      "grad_norm": 0.1434367597103119,
      "learning_rate": 5.217277236572824e-06,
      "loss": 0.0855,
      "num_input_tokens_seen": 59920448,
      "step": 31625
    },
    {
      "epoch": 5.159882535280202,
      "grad_norm": 0.39936646819114685,
      "learning_rate": 5.211276194072093e-06,
      "loss": 0.147,
      "num_input_tokens_seen": 59929680,
      "step": 31630
    },
    {
      "epoch": 5.16069826250102,
      "grad_norm": 0.12884128093719482,
      "learning_rate": 5.205278203241254e-06,
      "loss": 0.1176,
      "num_input_tokens_seen": 59939648,
      "step": 31635
    },
    {
      "epoch": 5.161513989721837,
      "grad_norm": 0.72426438331604,
      "learning_rate": 5.199283265005278e-06,
      "loss": 0.1234,
      "num_input_tokens_seen": 59949296,
      "step": 31640
    },
    {
      "epoch": 5.162329716942654,
      "grad_norm": 0.3044857382774353,
      "learning_rate": 5.193291380288648e-06,
      "loss": 0.2492,
      "num_input_tokens_seen": 59959392,
      "step": 31645
    },
    {
      "epoch": 5.163145444163471,
      "grad_norm": 0.2849677801132202,
      "learning_rate": 5.1873025500153995e-06,
      "loss": 0.1877,
      "num_input_tokens_seen": 59969664,
      "step": 31650
    },
    {
      "epoch": 5.1639611713842895,
      "grad_norm": 0.6799843907356262,
      "learning_rate": 5.181316775109071e-06,
      "loss": 0.103,
      "num_input_tokens_seen": 59978960,
      "step": 31655
    },
    {
      "epoch": 5.164776898605107,
      "grad_norm": 0.07712294161319733,
      "learning_rate": 5.1753340564927564e-06,
      "loss": 0.0863,
      "num_input_tokens_seen": 59987280,
      "step": 31660
    },
    {
      "epoch": 5.165592625825924,
      "grad_norm": 0.15950743854045868,
      "learning_rate": 5.169354395089068e-06,
      "loss": 0.115,
      "num_input_tokens_seen": 59996880,
      "step": 31665
    },
    {
      "epoch": 5.166408353046741,
      "grad_norm": 0.6460190415382385,
      "learning_rate": 5.1633777918201346e-06,
      "loss": 0.0911,
      "num_input_tokens_seen": 60007056,
      "step": 31670
    },
    {
      "epoch": 5.167224080267559,
      "grad_norm": 0.8234176635742188,
      "learning_rate": 5.157404247607625e-06,
      "loss": 0.0953,
      "num_input_tokens_seen": 60015392,
      "step": 31675
    },
    {
      "epoch": 5.168039807488376,
      "grad_norm": 1.5848013162612915,
      "learning_rate": 5.1514337633727454e-06,
      "loss": 0.1619,
      "num_input_tokens_seen": 60023488,
      "step": 31680
    },
    {
      "epoch": 5.168855534709193,
      "grad_norm": 0.06890374422073364,
      "learning_rate": 5.145466340036206e-06,
      "loss": 0.0793,
      "num_input_tokens_seen": 60033440,
      "step": 31685
    },
    {
      "epoch": 5.16967126193001,
      "grad_norm": 0.23146040737628937,
      "learning_rate": 5.139501978518274e-06,
      "loss": 0.0414,
      "num_input_tokens_seen": 60044128,
      "step": 31690
    },
    {
      "epoch": 5.170486989150828,
      "grad_norm": 0.3809041976928711,
      "learning_rate": 5.133540679738716e-06,
      "loss": 0.1508,
      "num_input_tokens_seen": 60052240,
      "step": 31695
    },
    {
      "epoch": 5.171302716371645,
      "grad_norm": 1.0873597860336304,
      "learning_rate": 5.127582444616838e-06,
      "loss": 0.1311,
      "num_input_tokens_seen": 60061104,
      "step": 31700
    },
    {
      "epoch": 5.1721184435924625,
      "grad_norm": 0.259143590927124,
      "learning_rate": 5.121627274071486e-06,
      "loss": 0.1357,
      "num_input_tokens_seen": 60071232,
      "step": 31705
    },
    {
      "epoch": 5.17293417081328,
      "grad_norm": 2.548157215118408,
      "learning_rate": 5.115675169021009e-06,
      "loss": 0.1757,
      "num_input_tokens_seen": 60080720,
      "step": 31710
    },
    {
      "epoch": 5.173749898034098,
      "grad_norm": 0.22435453534126282,
      "learning_rate": 5.1097261303832994e-06,
      "loss": 0.094,
      "num_input_tokens_seen": 60090528,
      "step": 31715
    },
    {
      "epoch": 5.174565625254915,
      "grad_norm": 1.4654040336608887,
      "learning_rate": 5.103780159075788e-06,
      "loss": 0.1995,
      "num_input_tokens_seen": 60099552,
      "step": 31720
    },
    {
      "epoch": 5.175381352475732,
      "grad_norm": 0.5618850588798523,
      "learning_rate": 5.0978372560154e-06,
      "loss": 0.1045,
      "num_input_tokens_seen": 60107968,
      "step": 31725
    },
    {
      "epoch": 5.176197079696549,
      "grad_norm": 0.7849770188331604,
      "learning_rate": 5.091897422118619e-06,
      "loss": 0.0762,
      "num_input_tokens_seen": 60117568,
      "step": 31730
    },
    {
      "epoch": 5.177012806917367,
      "grad_norm": 1.431793451309204,
      "learning_rate": 5.0859606583014305e-06,
      "loss": 0.0698,
      "num_input_tokens_seen": 60127040,
      "step": 31735
    },
    {
      "epoch": 5.177828534138184,
      "grad_norm": 0.44561392068862915,
      "learning_rate": 5.080026965479365e-06,
      "loss": 0.0466,
      "num_input_tokens_seen": 60136864,
      "step": 31740
    },
    {
      "epoch": 5.178644261359001,
      "grad_norm": 0.2576200067996979,
      "learning_rate": 5.074096344567475e-06,
      "loss": 0.1052,
      "num_input_tokens_seen": 60145216,
      "step": 31745
    },
    {
      "epoch": 5.1794599885798185,
      "grad_norm": 1.0073914527893066,
      "learning_rate": 5.0681687964803294e-06,
      "loss": 0.2292,
      "num_input_tokens_seen": 60153680,
      "step": 31750
    },
    {
      "epoch": 5.1802757158006365,
      "grad_norm": 0.47863030433654785,
      "learning_rate": 5.06224432213204e-06,
      "loss": 0.0428,
      "num_input_tokens_seen": 60161392,
      "step": 31755
    },
    {
      "epoch": 5.181091443021454,
      "grad_norm": 0.11106991022825241,
      "learning_rate": 5.056322922436224e-06,
      "loss": 0.1491,
      "num_input_tokens_seen": 60170416,
      "step": 31760
    },
    {
      "epoch": 5.181907170242271,
      "grad_norm": 0.5546286106109619,
      "learning_rate": 5.0504045983060465e-06,
      "loss": 0.1333,
      "num_input_tokens_seen": 60179280,
      "step": 31765
    },
    {
      "epoch": 5.182722897463089,
      "grad_norm": 0.42580047249794006,
      "learning_rate": 5.044489350654183e-06,
      "loss": 0.0859,
      "num_input_tokens_seen": 60187824,
      "step": 31770
    },
    {
      "epoch": 5.183538624683906,
      "grad_norm": 0.7929852604866028,
      "learning_rate": 5.038577180392831e-06,
      "loss": 0.0418,
      "num_input_tokens_seen": 60197104,
      "step": 31775
    },
    {
      "epoch": 5.184354351904723,
      "grad_norm": 1.3883925676345825,
      "learning_rate": 5.032668088433729e-06,
      "loss": 0.1099,
      "num_input_tokens_seen": 60206256,
      "step": 31780
    },
    {
      "epoch": 5.18517007912554,
      "grad_norm": 1.3036495447158813,
      "learning_rate": 5.02676207568814e-06,
      "loss": 0.2121,
      "num_input_tokens_seen": 60215968,
      "step": 31785
    },
    {
      "epoch": 5.185985806346358,
      "grad_norm": 0.348116010427475,
      "learning_rate": 5.02085914306683e-06,
      "loss": 0.0524,
      "num_input_tokens_seen": 60226240,
      "step": 31790
    },
    {
      "epoch": 5.186801533567175,
      "grad_norm": 0.4533537030220032,
      "learning_rate": 5.014959291480123e-06,
      "loss": 0.1906,
      "num_input_tokens_seen": 60237312,
      "step": 31795
    },
    {
      "epoch": 5.1876172607879925,
      "grad_norm": 0.10977702587842941,
      "learning_rate": 5.009062521837835e-06,
      "loss": 0.107,
      "num_input_tokens_seen": 60247360,
      "step": 31800
    },
    {
      "epoch": 5.1876172607879925,
      "eval_loss": 0.14101256430149078,
      "eval_runtime": 133.2701,
      "eval_samples_per_second": 20.447,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 60247360,
      "step": 31800
    },
    {
      "epoch": 5.18843298800881,
      "grad_norm": 1.4767862558364868,
      "learning_rate": 5.003168835049324e-06,
      "loss": 0.2605,
      "num_input_tokens_seen": 60257120,
      "step": 31805
    },
    {
      "epoch": 5.189248715229628,
      "grad_norm": 0.475212424993515,
      "learning_rate": 4.997278232023483e-06,
      "loss": 0.1822,
      "num_input_tokens_seen": 60266112,
      "step": 31810
    },
    {
      "epoch": 5.190064442450445,
      "grad_norm": 0.805801272392273,
      "learning_rate": 4.9913907136687036e-06,
      "loss": 0.1999,
      "num_input_tokens_seen": 60274928,
      "step": 31815
    },
    {
      "epoch": 5.190880169671262,
      "grad_norm": 0.7553549408912659,
      "learning_rate": 4.985506280892918e-06,
      "loss": 0.222,
      "num_input_tokens_seen": 60284368,
      "step": 31820
    },
    {
      "epoch": 5.191695896892079,
      "grad_norm": 0.1915169507265091,
      "learning_rate": 4.979624934603589e-06,
      "loss": 0.075,
      "num_input_tokens_seen": 60293360,
      "step": 31825
    },
    {
      "epoch": 5.192511624112897,
      "grad_norm": 0.8077946305274963,
      "learning_rate": 4.97374667570768e-06,
      "loss": 0.1097,
      "num_input_tokens_seen": 60302992,
      "step": 31830
    },
    {
      "epoch": 5.193327351333714,
      "grad_norm": 0.10112566500902176,
      "learning_rate": 4.967871505111704e-06,
      "loss": 0.1295,
      "num_input_tokens_seen": 60312512,
      "step": 31835
    },
    {
      "epoch": 5.194143078554531,
      "grad_norm": 1.1516934633255005,
      "learning_rate": 4.961999423721686e-06,
      "loss": 0.247,
      "num_input_tokens_seen": 60321936,
      "step": 31840
    },
    {
      "epoch": 5.194958805775348,
      "grad_norm": 0.07122356444597244,
      "learning_rate": 4.956130432443159e-06,
      "loss": 0.1159,
      "num_input_tokens_seen": 60331792,
      "step": 31845
    },
    {
      "epoch": 5.195774532996166,
      "grad_norm": 0.2664964199066162,
      "learning_rate": 4.950264532181215e-06,
      "loss": 0.1892,
      "num_input_tokens_seen": 60339872,
      "step": 31850
    },
    {
      "epoch": 5.196590260216984,
      "grad_norm": 0.07088261842727661,
      "learning_rate": 4.944401723840433e-06,
      "loss": 0.0978,
      "num_input_tokens_seen": 60349952,
      "step": 31855
    },
    {
      "epoch": 5.197405987437801,
      "grad_norm": 0.3018292784690857,
      "learning_rate": 4.938542008324942e-06,
      "loss": 0.0968,
      "num_input_tokens_seen": 60358896,
      "step": 31860
    },
    {
      "epoch": 5.198221714658618,
      "grad_norm": 0.7476277351379395,
      "learning_rate": 4.9326853865383855e-06,
      "loss": 0.1764,
      "num_input_tokens_seen": 60367584,
      "step": 31865
    },
    {
      "epoch": 5.199037441879436,
      "grad_norm": 0.8413105010986328,
      "learning_rate": 4.926831859383918e-06,
      "loss": 0.2298,
      "num_input_tokens_seen": 60377360,
      "step": 31870
    },
    {
      "epoch": 5.199853169100253,
      "grad_norm": 1.496875524520874,
      "learning_rate": 4.92098142776424e-06,
      "loss": 0.1274,
      "num_input_tokens_seen": 60386288,
      "step": 31875
    },
    {
      "epoch": 5.20066889632107,
      "grad_norm": 0.666090726852417,
      "learning_rate": 4.91513409258155e-06,
      "loss": 0.0985,
      "num_input_tokens_seen": 60396352,
      "step": 31880
    },
    {
      "epoch": 5.201484623541887,
      "grad_norm": 0.2884186804294586,
      "learning_rate": 4.909289854737581e-06,
      "loss": 0.0904,
      "num_input_tokens_seen": 60404848,
      "step": 31885
    },
    {
      "epoch": 5.202300350762705,
      "grad_norm": 0.2937859296798706,
      "learning_rate": 4.903448715133602e-06,
      "loss": 0.0955,
      "num_input_tokens_seen": 60414384,
      "step": 31890
    },
    {
      "epoch": 5.203116077983522,
      "grad_norm": 0.731503427028656,
      "learning_rate": 4.897610674670372e-06,
      "loss": 0.0899,
      "num_input_tokens_seen": 60423136,
      "step": 31895
    },
    {
      "epoch": 5.2039318052043395,
      "grad_norm": 1.6063475608825684,
      "learning_rate": 4.8917757342482e-06,
      "loss": 0.2616,
      "num_input_tokens_seen": 60431536,
      "step": 31900
    },
    {
      "epoch": 5.204747532425157,
      "grad_norm": 1.1840925216674805,
      "learning_rate": 4.885943894766909e-06,
      "loss": 0.0944,
      "num_input_tokens_seen": 60441136,
      "step": 31905
    },
    {
      "epoch": 5.205563259645975,
      "grad_norm": 0.22306108474731445,
      "learning_rate": 4.880115157125842e-06,
      "loss": 0.1408,
      "num_input_tokens_seen": 60451984,
      "step": 31910
    },
    {
      "epoch": 5.206378986866792,
      "grad_norm": 0.1626429259777069,
      "learning_rate": 4.874289522223857e-06,
      "loss": 0.0919,
      "num_input_tokens_seen": 60461872,
      "step": 31915
    },
    {
      "epoch": 5.207194714087609,
      "grad_norm": 0.7593790888786316,
      "learning_rate": 4.868466990959339e-06,
      "loss": 0.1252,
      "num_input_tokens_seen": 60470720,
      "step": 31920
    },
    {
      "epoch": 5.208010441308426,
      "grad_norm": 1.5254167318344116,
      "learning_rate": 4.8626475642301964e-06,
      "loss": 0.2676,
      "num_input_tokens_seen": 60479776,
      "step": 31925
    },
    {
      "epoch": 5.208826168529244,
      "grad_norm": 0.4837275445461273,
      "learning_rate": 4.856831242933871e-06,
      "loss": 0.2108,
      "num_input_tokens_seen": 60491088,
      "step": 31930
    },
    {
      "epoch": 5.209641895750061,
      "grad_norm": 0.2928425073623657,
      "learning_rate": 4.851018027967294e-06,
      "loss": 0.1039,
      "num_input_tokens_seen": 60500288,
      "step": 31935
    },
    {
      "epoch": 5.210457622970878,
      "grad_norm": 2.6353135108947754,
      "learning_rate": 4.845207920226946e-06,
      "loss": 0.2066,
      "num_input_tokens_seen": 60510080,
      "step": 31940
    },
    {
      "epoch": 5.211273350191696,
      "grad_norm": 0.9092556238174438,
      "learning_rate": 4.839400920608825e-06,
      "loss": 0.09,
      "num_input_tokens_seen": 60519760,
      "step": 31945
    },
    {
      "epoch": 5.2120890774125135,
      "grad_norm": 1.9086233377456665,
      "learning_rate": 4.83359703000843e-06,
      "loss": 0.2053,
      "num_input_tokens_seen": 60529264,
      "step": 31950
    },
    {
      "epoch": 5.212904804633331,
      "grad_norm": 0.3807205855846405,
      "learning_rate": 4.827796249320804e-06,
      "loss": 0.3813,
      "num_input_tokens_seen": 60539312,
      "step": 31955
    },
    {
      "epoch": 5.213720531854148,
      "grad_norm": 0.9849512577056885,
      "learning_rate": 4.82199857944049e-06,
      "loss": 0.1039,
      "num_input_tokens_seen": 60547440,
      "step": 31960
    },
    {
      "epoch": 5.214536259074965,
      "grad_norm": 0.18261510133743286,
      "learning_rate": 4.8162040212615695e-06,
      "loss": 0.0475,
      "num_input_tokens_seen": 60557040,
      "step": 31965
    },
    {
      "epoch": 5.215351986295783,
      "grad_norm": 0.7444178462028503,
      "learning_rate": 4.810412575677639e-06,
      "loss": 0.1404,
      "num_input_tokens_seen": 60566880,
      "step": 31970
    },
    {
      "epoch": 5.2161677135166,
      "grad_norm": 1.8530020713806152,
      "learning_rate": 4.804624243581801e-06,
      "loss": 0.1398,
      "num_input_tokens_seen": 60576112,
      "step": 31975
    },
    {
      "epoch": 5.216983440737417,
      "grad_norm": 0.8126489520072937,
      "learning_rate": 4.798839025866703e-06,
      "loss": 0.1303,
      "num_input_tokens_seen": 60586192,
      "step": 31980
    },
    {
      "epoch": 5.217799167958235,
      "grad_norm": 0.38581547141075134,
      "learning_rate": 4.793056923424491e-06,
      "loss": 0.0981,
      "num_input_tokens_seen": 60594224,
      "step": 31985
    },
    {
      "epoch": 5.218614895179052,
      "grad_norm": 0.5374927520751953,
      "learning_rate": 4.78727793714683e-06,
      "loss": 0.2154,
      "num_input_tokens_seen": 60603120,
      "step": 31990
    },
    {
      "epoch": 5.219430622399869,
      "grad_norm": 0.7407779097557068,
      "learning_rate": 4.7815020679249285e-06,
      "loss": 0.1384,
      "num_input_tokens_seen": 60612208,
      "step": 31995
    },
    {
      "epoch": 5.220246349620687,
      "grad_norm": 1.2055379152297974,
      "learning_rate": 4.775729316649483e-06,
      "loss": 0.0413,
      "num_input_tokens_seen": 60622464,
      "step": 32000
    },
    {
      "epoch": 5.220246349620687,
      "eval_loss": 0.14088134467601776,
      "eval_runtime": 133.3561,
      "eval_samples_per_second": 20.434,
      "eval_steps_per_second": 10.221,
      "num_input_tokens_seen": 60622464,
      "step": 32000
    },
    {
      "epoch": 5.221062076841505,
      "grad_norm": 0.10502181202173233,
      "learning_rate": 4.769959684210728e-06,
      "loss": 0.0275,
      "num_input_tokens_seen": 60631664,
      "step": 32005
    },
    {
      "epoch": 5.221877804062322,
      "grad_norm": 0.3528323173522949,
      "learning_rate": 4.764193171498426e-06,
      "loss": 0.1814,
      "num_input_tokens_seen": 60642336,
      "step": 32010
    },
    {
      "epoch": 5.222693531283139,
      "grad_norm": 0.2950824499130249,
      "learning_rate": 4.75842977940183e-06,
      "loss": 0.0517,
      "num_input_tokens_seen": 60651504,
      "step": 32015
    },
    {
      "epoch": 5.223509258503956,
      "grad_norm": 0.8686140775680542,
      "learning_rate": 4.752669508809729e-06,
      "loss": 0.084,
      "num_input_tokens_seen": 60661968,
      "step": 32020
    },
    {
      "epoch": 5.224324985724774,
      "grad_norm": 0.5218603014945984,
      "learning_rate": 4.746912360610445e-06,
      "loss": 0.0499,
      "num_input_tokens_seen": 60671952,
      "step": 32025
    },
    {
      "epoch": 5.225140712945591,
      "grad_norm": 2.246122121810913,
      "learning_rate": 4.741158335691781e-06,
      "loss": 0.1137,
      "num_input_tokens_seen": 60681280,
      "step": 32030
    },
    {
      "epoch": 5.225956440166408,
      "grad_norm": 0.4609055817127228,
      "learning_rate": 4.7354074349410994e-06,
      "loss": 0.0919,
      "num_input_tokens_seen": 60690368,
      "step": 32035
    },
    {
      "epoch": 5.226772167387225,
      "grad_norm": 0.05803927034139633,
      "learning_rate": 4.729659659245245e-06,
      "loss": 0.113,
      "num_input_tokens_seen": 60701152,
      "step": 32040
    },
    {
      "epoch": 5.227587894608043,
      "grad_norm": 0.8818814158439636,
      "learning_rate": 4.723915009490601e-06,
      "loss": 0.0516,
      "num_input_tokens_seen": 60709904,
      "step": 32045
    },
    {
      "epoch": 5.2284036218288605,
      "grad_norm": 0.09246750921010971,
      "learning_rate": 4.718173486563077e-06,
      "loss": 0.1851,
      "num_input_tokens_seen": 60719664,
      "step": 32050
    },
    {
      "epoch": 5.229219349049678,
      "grad_norm": 0.44604310393333435,
      "learning_rate": 4.71243509134808e-06,
      "loss": 0.061,
      "num_input_tokens_seen": 60728976,
      "step": 32055
    },
    {
      "epoch": 5.230035076270495,
      "grad_norm": 1.3048865795135498,
      "learning_rate": 4.706699824730532e-06,
      "loss": 0.3618,
      "num_input_tokens_seen": 60738880,
      "step": 32060
    },
    {
      "epoch": 5.230850803491313,
      "grad_norm": 0.28284305334091187,
      "learning_rate": 4.700967687594901e-06,
      "loss": 0.0976,
      "num_input_tokens_seen": 60749488,
      "step": 32065
    },
    {
      "epoch": 5.23166653071213,
      "grad_norm": 0.7967010140419006,
      "learning_rate": 4.69523868082514e-06,
      "loss": 0.0343,
      "num_input_tokens_seen": 60760032,
      "step": 32070
    },
    {
      "epoch": 5.232482257932947,
      "grad_norm": 1.5278661251068115,
      "learning_rate": 4.689512805304747e-06,
      "loss": 0.0829,
      "num_input_tokens_seen": 60769680,
      "step": 32075
    },
    {
      "epoch": 5.233297985153764,
      "grad_norm": 1.2248728275299072,
      "learning_rate": 4.683790061916707e-06,
      "loss": 0.1659,
      "num_input_tokens_seen": 60779360,
      "step": 32080
    },
    {
      "epoch": 5.234113712374582,
      "grad_norm": 0.6677896976470947,
      "learning_rate": 4.678070451543551e-06,
      "loss": 0.1187,
      "num_input_tokens_seen": 60789632,
      "step": 32085
    },
    {
      "epoch": 5.234929439595399,
      "grad_norm": 0.9946677684783936,
      "learning_rate": 4.6723539750673204e-06,
      "loss": 0.0686,
      "num_input_tokens_seen": 60798656,
      "step": 32090
    },
    {
      "epoch": 5.2357451668162165,
      "grad_norm": 0.1616320163011551,
      "learning_rate": 4.666640633369551e-06,
      "loss": 0.1003,
      "num_input_tokens_seen": 60807408,
      "step": 32095
    },
    {
      "epoch": 5.236560894037034,
      "grad_norm": 0.5250091552734375,
      "learning_rate": 4.660930427331323e-06,
      "loss": 0.2677,
      "num_input_tokens_seen": 60817264,
      "step": 32100
    },
    {
      "epoch": 5.237376621257852,
      "grad_norm": 0.3815121054649353,
      "learning_rate": 4.6552233578332244e-06,
      "loss": 0.1048,
      "num_input_tokens_seen": 60828096,
      "step": 32105
    },
    {
      "epoch": 5.238192348478669,
      "grad_norm": 0.33330854773521423,
      "learning_rate": 4.649519425755347e-06,
      "loss": 0.0829,
      "num_input_tokens_seen": 60837536,
      "step": 32110
    },
    {
      "epoch": 5.239008075699486,
      "grad_norm": 1.2080810070037842,
      "learning_rate": 4.64381863197732e-06,
      "loss": 0.1304,
      "num_input_tokens_seen": 60846256,
      "step": 32115
    },
    {
      "epoch": 5.239823802920303,
      "grad_norm": 0.6869412660598755,
      "learning_rate": 4.638120977378269e-06,
      "loss": 0.0769,
      "num_input_tokens_seen": 60856288,
      "step": 32120
    },
    {
      "epoch": 5.240639530141121,
      "grad_norm": 0.6661149859428406,
      "learning_rate": 4.632426462836848e-06,
      "loss": 0.3048,
      "num_input_tokens_seen": 60865696,
      "step": 32125
    },
    {
      "epoch": 5.241455257361938,
      "grad_norm": 1.5697565078735352,
      "learning_rate": 4.626735089231224e-06,
      "loss": 0.1403,
      "num_input_tokens_seen": 60875632,
      "step": 32130
    },
    {
      "epoch": 5.242270984582755,
      "grad_norm": 1.4082298278808594,
      "learning_rate": 4.621046857439068e-06,
      "loss": 0.23,
      "num_input_tokens_seen": 60884976,
      "step": 32135
    },
    {
      "epoch": 5.243086711803572,
      "grad_norm": 0.08299555629491806,
      "learning_rate": 4.615361768337587e-06,
      "loss": 0.1287,
      "num_input_tokens_seen": 60894976,
      "step": 32140
    },
    {
      "epoch": 5.2439024390243905,
      "grad_norm": 1.344712734222412,
      "learning_rate": 4.6096798228034946e-06,
      "loss": 0.2492,
      "num_input_tokens_seen": 60904288,
      "step": 32145
    },
    {
      "epoch": 5.244718166245208,
      "grad_norm": 0.12663087248802185,
      "learning_rate": 4.604001021713008e-06,
      "loss": 0.0799,
      "num_input_tokens_seen": 60913504,
      "step": 32150
    },
    {
      "epoch": 5.245533893466025,
      "grad_norm": 0.3822021186351776,
      "learning_rate": 4.598325365941883e-06,
      "loss": 0.0887,
      "num_input_tokens_seen": 60922640,
      "step": 32155
    },
    {
      "epoch": 5.246349620686843,
      "grad_norm": 0.3615717589855194,
      "learning_rate": 4.5926528563653645e-06,
      "loss": 0.0631,
      "num_input_tokens_seen": 60931808,
      "step": 32160
    },
    {
      "epoch": 5.24716534790766,
      "grad_norm": 0.370208740234375,
      "learning_rate": 4.5869834938582295e-06,
      "loss": 0.0556,
      "num_input_tokens_seen": 60942096,
      "step": 32165
    },
    {
      "epoch": 5.247981075128477,
      "grad_norm": 0.4481102228164673,
      "learning_rate": 4.581317279294772e-06,
      "loss": 0.2149,
      "num_input_tokens_seen": 60951168,
      "step": 32170
    },
    {
      "epoch": 5.248796802349294,
      "grad_norm": 0.06834748387336731,
      "learning_rate": 4.57565421354878e-06,
      "loss": 0.1691,
      "num_input_tokens_seen": 60959504,
      "step": 32175
    },
    {
      "epoch": 5.249612529570112,
      "grad_norm": 0.40332987904548645,
      "learning_rate": 4.569994297493579e-06,
      "loss": 0.3087,
      "num_input_tokens_seen": 60968320,
      "step": 32180
    },
    {
      "epoch": 5.250428256790929,
      "grad_norm": 1.0031298398971558,
      "learning_rate": 4.564337532002002e-06,
      "loss": 0.2192,
      "num_input_tokens_seen": 60978800,
      "step": 32185
    },
    {
      "epoch": 5.251243984011746,
      "grad_norm": 1.6818673610687256,
      "learning_rate": 4.55868391794638e-06,
      "loss": 0.2072,
      "num_input_tokens_seen": 60988496,
      "step": 32190
    },
    {
      "epoch": 5.2520597112325635,
      "grad_norm": 0.8687321543693542,
      "learning_rate": 4.553033456198588e-06,
      "loss": 0.1483,
      "num_input_tokens_seen": 60997776,
      "step": 32195
    },
    {
      "epoch": 5.252875438453382,
      "grad_norm": 0.6491759419441223,
      "learning_rate": 4.54738614762999e-06,
      "loss": 0.0824,
      "num_input_tokens_seen": 61006768,
      "step": 32200
    },
    {
      "epoch": 5.252875438453382,
      "eval_loss": 0.1405024528503418,
      "eval_runtime": 133.2952,
      "eval_samples_per_second": 20.443,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 61006768,
      "step": 32200
    },
    {
      "epoch": 5.253691165674199,
      "grad_norm": 1.0954149961471558,
      "learning_rate": 4.541741993111465e-06,
      "loss": 0.2028,
      "num_input_tokens_seen": 61015360,
      "step": 32205
    },
    {
      "epoch": 5.254506892895016,
      "grad_norm": 0.796089768409729,
      "learning_rate": 4.536100993513423e-06,
      "loss": 0.1073,
      "num_input_tokens_seen": 61023536,
      "step": 32210
    },
    {
      "epoch": 5.255322620115833,
      "grad_norm": 1.5150067806243896,
      "learning_rate": 4.530463149705768e-06,
      "loss": 0.1048,
      "num_input_tokens_seen": 61032496,
      "step": 32215
    },
    {
      "epoch": 5.256138347336651,
      "grad_norm": 1.9599360227584839,
      "learning_rate": 4.524828462557934e-06,
      "loss": 0.0992,
      "num_input_tokens_seen": 61042336,
      "step": 32220
    },
    {
      "epoch": 5.256954074557468,
      "grad_norm": 0.04673568531870842,
      "learning_rate": 4.5191969329388625e-06,
      "loss": 0.0517,
      "num_input_tokens_seen": 61052160,
      "step": 32225
    },
    {
      "epoch": 5.257769801778285,
      "grad_norm": 1.4695532321929932,
      "learning_rate": 4.5135685617169965e-06,
      "loss": 0.1346,
      "num_input_tokens_seen": 61062272,
      "step": 32230
    },
    {
      "epoch": 5.258585528999102,
      "grad_norm": 0.868217945098877,
      "learning_rate": 4.507943349760313e-06,
      "loss": 0.0665,
      "num_input_tokens_seen": 61072096,
      "step": 32235
    },
    {
      "epoch": 5.25940125621992,
      "grad_norm": 2.344630479812622,
      "learning_rate": 4.502321297936277e-06,
      "loss": 0.1837,
      "num_input_tokens_seen": 61081840,
      "step": 32240
    },
    {
      "epoch": 5.2602169834407375,
      "grad_norm": 0.79212886095047,
      "learning_rate": 4.496702407111888e-06,
      "loss": 0.2038,
      "num_input_tokens_seen": 61092224,
      "step": 32245
    },
    {
      "epoch": 5.261032710661555,
      "grad_norm": 1.295959234237671,
      "learning_rate": 4.491086678153653e-06,
      "loss": 0.1314,
      "num_input_tokens_seen": 61101072,
      "step": 32250
    },
    {
      "epoch": 5.261848437882372,
      "grad_norm": 1.5923168659210205,
      "learning_rate": 4.485474111927579e-06,
      "loss": 0.1925,
      "num_input_tokens_seen": 61109776,
      "step": 32255
    },
    {
      "epoch": 5.26266416510319,
      "grad_norm": 1.5447477102279663,
      "learning_rate": 4.479864709299197e-06,
      "loss": 0.2964,
      "num_input_tokens_seen": 61120816,
      "step": 32260
    },
    {
      "epoch": 5.263479892324007,
      "grad_norm": 0.8330258727073669,
      "learning_rate": 4.474258471133555e-06,
      "loss": 0.0746,
      "num_input_tokens_seen": 61129808,
      "step": 32265
    },
    {
      "epoch": 5.264295619544824,
      "grad_norm": 0.24417191743850708,
      "learning_rate": 4.4686553982952014e-06,
      "loss": 0.0815,
      "num_input_tokens_seen": 61139472,
      "step": 32270
    },
    {
      "epoch": 5.265111346765641,
      "grad_norm": 0.6784769296646118,
      "learning_rate": 4.463055491648191e-06,
      "loss": 0.1737,
      "num_input_tokens_seen": 61148576,
      "step": 32275
    },
    {
      "epoch": 5.265927073986459,
      "grad_norm": 0.18755413591861725,
      "learning_rate": 4.457458752056112e-06,
      "loss": 0.1303,
      "num_input_tokens_seen": 61158144,
      "step": 32280
    },
    {
      "epoch": 5.266742801207276,
      "grad_norm": 0.7226209044456482,
      "learning_rate": 4.451865180382042e-06,
      "loss": 0.0912,
      "num_input_tokens_seen": 61167408,
      "step": 32285
    },
    {
      "epoch": 5.2675585284280935,
      "grad_norm": 0.962560772895813,
      "learning_rate": 4.4462747774885936e-06,
      "loss": 0.198,
      "num_input_tokens_seen": 61177632,
      "step": 32290
    },
    {
      "epoch": 5.268374255648911,
      "grad_norm": 0.1788845807313919,
      "learning_rate": 4.440687544237859e-06,
      "loss": 0.0916,
      "num_input_tokens_seen": 61188048,
      "step": 32295
    },
    {
      "epoch": 5.269189982869729,
      "grad_norm": 0.6186056137084961,
      "learning_rate": 4.435103481491471e-06,
      "loss": 0.089,
      "num_input_tokens_seen": 61195424,
      "step": 32300
    },
    {
      "epoch": 5.270005710090546,
      "grad_norm": 0.49426713585853577,
      "learning_rate": 4.429522590110569e-06,
      "loss": 0.0319,
      "num_input_tokens_seen": 61205376,
      "step": 32305
    },
    {
      "epoch": 5.270821437311363,
      "grad_norm": 0.04155900329351425,
      "learning_rate": 4.423944870955779e-06,
      "loss": 0.0395,
      "num_input_tokens_seen": 61215488,
      "step": 32310
    },
    {
      "epoch": 5.27163716453218,
      "grad_norm": 0.5678252577781677,
      "learning_rate": 4.418370324887272e-06,
      "loss": 0.0476,
      "num_input_tokens_seen": 61224736,
      "step": 32315
    },
    {
      "epoch": 5.272452891752998,
      "grad_norm": 0.31630662083625793,
      "learning_rate": 4.412798952764699e-06,
      "loss": 0.0449,
      "num_input_tokens_seen": 61234032,
      "step": 32320
    },
    {
      "epoch": 5.273268618973815,
      "grad_norm": 0.49291718006134033,
      "learning_rate": 4.407230755447245e-06,
      "loss": 0.0585,
      "num_input_tokens_seen": 61244336,
      "step": 32325
    },
    {
      "epoch": 5.274084346194632,
      "grad_norm": 0.30107617378234863,
      "learning_rate": 4.401665733793598e-06,
      "loss": 0.0644,
      "num_input_tokens_seen": 61253744,
      "step": 32330
    },
    {
      "epoch": 5.27490007341545,
      "grad_norm": 0.32902660965919495,
      "learning_rate": 4.3961038886619425e-06,
      "loss": 0.1362,
      "num_input_tokens_seen": 61263168,
      "step": 32335
    },
    {
      "epoch": 5.275715800636267,
      "grad_norm": 0.22086971998214722,
      "learning_rate": 4.39054522091e-06,
      "loss": 0.1337,
      "num_input_tokens_seen": 61273456,
      "step": 32340
    },
    {
      "epoch": 5.276531527857085,
      "grad_norm": 0.8604934215545654,
      "learning_rate": 4.384989731394979e-06,
      "loss": 0.2428,
      "num_input_tokens_seen": 61282528,
      "step": 32345
    },
    {
      "epoch": 5.277347255077902,
      "grad_norm": 0.10939616709947586,
      "learning_rate": 4.379437420973598e-06,
      "loss": 0.047,
      "num_input_tokens_seen": 61290704,
      "step": 32350
    },
    {
      "epoch": 5.27816298229872,
      "grad_norm": 0.07868442684412003,
      "learning_rate": 4.373888290502107e-06,
      "loss": 0.1114,
      "num_input_tokens_seen": 61301312,
      "step": 32355
    },
    {
      "epoch": 5.278978709519537,
      "grad_norm": 1.506834626197815,
      "learning_rate": 4.36834234083624e-06,
      "loss": 0.2336,
      "num_input_tokens_seen": 61311936,
      "step": 32360
    },
    {
      "epoch": 5.279794436740354,
      "grad_norm": 0.7261633276939392,
      "learning_rate": 4.362799572831258e-06,
      "loss": 0.0977,
      "num_input_tokens_seen": 61321808,
      "step": 32365
    },
    {
      "epoch": 5.280610163961171,
      "grad_norm": 0.39299121499061584,
      "learning_rate": 4.35725998734193e-06,
      "loss": 0.0892,
      "num_input_tokens_seen": 61330784,
      "step": 32370
    },
    {
      "epoch": 5.281425891181989,
      "grad_norm": 0.3141215145587921,
      "learning_rate": 4.3517235852225195e-06,
      "loss": 0.041,
      "num_input_tokens_seen": 61339008,
      "step": 32375
    },
    {
      "epoch": 5.282241618402806,
      "grad_norm": 0.9602063298225403,
      "learning_rate": 4.346190367326822e-06,
      "loss": 0.1028,
      "num_input_tokens_seen": 61348656,
      "step": 32380
    },
    {
      "epoch": 5.283057345623623,
      "grad_norm": 0.0980956107378006,
      "learning_rate": 4.340660334508115e-06,
      "loss": 0.0921,
      "num_input_tokens_seen": 61357712,
      "step": 32385
    },
    {
      "epoch": 5.2838730728444405,
      "grad_norm": 0.30261915922164917,
      "learning_rate": 4.335133487619206e-06,
      "loss": 0.2473,
      "num_input_tokens_seen": 61367504,
      "step": 32390
    },
    {
      "epoch": 5.2846888000652585,
      "grad_norm": 0.055845748633146286,
      "learning_rate": 4.329609827512409e-06,
      "loss": 0.1266,
      "num_input_tokens_seen": 61376896,
      "step": 32395
    },
    {
      "epoch": 5.285504527286076,
      "grad_norm": 0.0779387354850769,
      "learning_rate": 4.324089355039531e-06,
      "loss": 0.0257,
      "num_input_tokens_seen": 61386992,
      "step": 32400
    },
    {
      "epoch": 5.285504527286076,
      "eval_loss": 0.14069148898124695,
      "eval_runtime": 133.2894,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 61386992,
      "step": 32400
    },
    {
      "epoch": 5.286320254506893,
      "grad_norm": 0.1785631626844406,
      "learning_rate": 4.3185720710519075e-06,
      "loss": 0.0796,
      "num_input_tokens_seen": 61397072,
      "step": 32405
    },
    {
      "epoch": 5.28713598172771,
      "grad_norm": 1.3563661575317383,
      "learning_rate": 4.3130579764003724e-06,
      "loss": 0.1795,
      "num_input_tokens_seen": 61406992,
      "step": 32410
    },
    {
      "epoch": 5.287951708948528,
      "grad_norm": 1.7863965034484863,
      "learning_rate": 4.307547071935267e-06,
      "loss": 0.0879,
      "num_input_tokens_seen": 61415760,
      "step": 32415
    },
    {
      "epoch": 5.288767436169345,
      "grad_norm": 0.790156900882721,
      "learning_rate": 4.302039358506435e-06,
      "loss": 0.1323,
      "num_input_tokens_seen": 61424736,
      "step": 32420
    },
    {
      "epoch": 5.289583163390162,
      "grad_norm": 0.22151748836040497,
      "learning_rate": 4.296534836963245e-06,
      "loss": 0.1413,
      "num_input_tokens_seen": 61433856,
      "step": 32425
    },
    {
      "epoch": 5.290398890610979,
      "grad_norm": 0.2141256183385849,
      "learning_rate": 4.291033508154555e-06,
      "loss": 0.0773,
      "num_input_tokens_seen": 61443472,
      "step": 32430
    },
    {
      "epoch": 5.291214617831797,
      "grad_norm": 0.13299037516117096,
      "learning_rate": 4.285535372928748e-06,
      "loss": 0.0653,
      "num_input_tokens_seen": 61453200,
      "step": 32435
    },
    {
      "epoch": 5.2920303450526145,
      "grad_norm": 1.4612592458724976,
      "learning_rate": 4.280040432133695e-06,
      "loss": 0.1717,
      "num_input_tokens_seen": 61463040,
      "step": 32440
    },
    {
      "epoch": 5.292846072273432,
      "grad_norm": 1.4059569835662842,
      "learning_rate": 4.274548686616789e-06,
      "loss": 0.2129,
      "num_input_tokens_seen": 61473184,
      "step": 32445
    },
    {
      "epoch": 5.293661799494249,
      "grad_norm": 0.4452126622200012,
      "learning_rate": 4.2690601372249364e-06,
      "loss": 0.0975,
      "num_input_tokens_seen": 61481920,
      "step": 32450
    },
    {
      "epoch": 5.294477526715067,
      "grad_norm": 1.0499776601791382,
      "learning_rate": 4.263574784804525e-06,
      "loss": 0.2031,
      "num_input_tokens_seen": 61492464,
      "step": 32455
    },
    {
      "epoch": 5.295293253935884,
      "grad_norm": 0.03803417459130287,
      "learning_rate": 4.258092630201479e-06,
      "loss": 0.2203,
      "num_input_tokens_seen": 61501280,
      "step": 32460
    },
    {
      "epoch": 5.296108981156701,
      "grad_norm": 0.12647885084152222,
      "learning_rate": 4.252613674261202e-06,
      "loss": 0.22,
      "num_input_tokens_seen": 61510416,
      "step": 32465
    },
    {
      "epoch": 5.296924708377518,
      "grad_norm": 0.2613542675971985,
      "learning_rate": 4.2471379178286224e-06,
      "loss": 0.1968,
      "num_input_tokens_seen": 61519616,
      "step": 32470
    },
    {
      "epoch": 5.297740435598336,
      "grad_norm": 0.8229423761367798,
      "learning_rate": 4.241665361748181e-06,
      "loss": 0.0686,
      "num_input_tokens_seen": 61528560,
      "step": 32475
    },
    {
      "epoch": 5.298556162819153,
      "grad_norm": 0.7059274315834045,
      "learning_rate": 4.2361960068637994e-06,
      "loss": 0.1303,
      "num_input_tokens_seen": 61537008,
      "step": 32480
    },
    {
      "epoch": 5.2993718900399704,
      "grad_norm": 1.168643593788147,
      "learning_rate": 4.230729854018933e-06,
      "loss": 0.1086,
      "num_input_tokens_seen": 61545344,
      "step": 32485
    },
    {
      "epoch": 5.300187617260788,
      "grad_norm": 0.6204383969306946,
      "learning_rate": 4.225266904056521e-06,
      "loss": 0.06,
      "num_input_tokens_seen": 61555296,
      "step": 32490
    },
    {
      "epoch": 5.301003344481606,
      "grad_norm": 0.031215498223900795,
      "learning_rate": 4.21980715781903e-06,
      "loss": 0.0365,
      "num_input_tokens_seen": 61565344,
      "step": 32495
    },
    {
      "epoch": 5.301819071702423,
      "grad_norm": 1.9409401416778564,
      "learning_rate": 4.214350616148416e-06,
      "loss": 0.263,
      "num_input_tokens_seen": 61576320,
      "step": 32500
    },
    {
      "epoch": 5.30263479892324,
      "grad_norm": 0.10468736290931702,
      "learning_rate": 4.20889727988614e-06,
      "loss": 0.1278,
      "num_input_tokens_seen": 61586352,
      "step": 32505
    },
    {
      "epoch": 5.303450526144058,
      "grad_norm": 1.1004210710525513,
      "learning_rate": 4.20344714987318e-06,
      "loss": 0.1597,
      "num_input_tokens_seen": 61595296,
      "step": 32510
    },
    {
      "epoch": 5.304266253364875,
      "grad_norm": 0.21351319551467896,
      "learning_rate": 4.198000226950022e-06,
      "loss": 0.0576,
      "num_input_tokens_seen": 61606000,
      "step": 32515
    },
    {
      "epoch": 5.305081980585692,
      "grad_norm": 0.6089025139808655,
      "learning_rate": 4.192556511956635e-06,
      "loss": 0.0359,
      "num_input_tokens_seen": 61614256,
      "step": 32520
    },
    {
      "epoch": 5.305897707806509,
      "grad_norm": 0.27026140689849854,
      "learning_rate": 4.18711600573252e-06,
      "loss": 0.2139,
      "num_input_tokens_seen": 61624544,
      "step": 32525
    },
    {
      "epoch": 5.306713435027326,
      "grad_norm": 0.3346690833568573,
      "learning_rate": 4.181678709116671e-06,
      "loss": 0.1852,
      "num_input_tokens_seen": 61634144,
      "step": 32530
    },
    {
      "epoch": 5.307529162248144,
      "grad_norm": 0.9056082963943481,
      "learning_rate": 4.1762446229475785e-06,
      "loss": 0.0208,
      "num_input_tokens_seen": 61643744,
      "step": 32535
    },
    {
      "epoch": 5.3083448894689615,
      "grad_norm": 0.5754855275154114,
      "learning_rate": 4.17081374806326e-06,
      "loss": 0.042,
      "num_input_tokens_seen": 61653120,
      "step": 32540
    },
    {
      "epoch": 5.309160616689779,
      "grad_norm": 0.7270680665969849,
      "learning_rate": 4.165386085301212e-06,
      "loss": 0.1651,
      "num_input_tokens_seen": 61662288,
      "step": 32545
    },
    {
      "epoch": 5.309976343910597,
      "grad_norm": 0.8273679614067078,
      "learning_rate": 4.1599616354984525e-06,
      "loss": 0.1368,
      "num_input_tokens_seen": 61671760,
      "step": 32550
    },
    {
      "epoch": 5.310792071131414,
      "grad_norm": 0.12646707892417908,
      "learning_rate": 4.154540399491508e-06,
      "loss": 0.1773,
      "num_input_tokens_seen": 61681152,
      "step": 32555
    },
    {
      "epoch": 5.311607798352231,
      "grad_norm": 0.20028430223464966,
      "learning_rate": 4.149122378116394e-06,
      "loss": 0.101,
      "num_input_tokens_seen": 61690672,
      "step": 32560
    },
    {
      "epoch": 5.312423525573048,
      "grad_norm": 0.217369943857193,
      "learning_rate": 4.14370757220863e-06,
      "loss": 0.2256,
      "num_input_tokens_seen": 61700192,
      "step": 32565
    },
    {
      "epoch": 5.313239252793866,
      "grad_norm": 0.12090373039245605,
      "learning_rate": 4.138295982603263e-06,
      "loss": 0.0995,
      "num_input_tokens_seen": 61709872,
      "step": 32570
    },
    {
      "epoch": 5.314054980014683,
      "grad_norm": 0.4677494466304779,
      "learning_rate": 4.132887610134814e-06,
      "loss": 0.1004,
      "num_input_tokens_seen": 61720304,
      "step": 32575
    },
    {
      "epoch": 5.3148707072355,
      "grad_norm": 0.26302963495254517,
      "learning_rate": 4.127482455637335e-06,
      "loss": 0.1358,
      "num_input_tokens_seen": 61730064,
      "step": 32580
    },
    {
      "epoch": 5.3156864344563175,
      "grad_norm": 0.4542093575000763,
      "learning_rate": 4.1220805199443545e-06,
      "loss": 0.0581,
      "num_input_tokens_seen": 61739808,
      "step": 32585
    },
    {
      "epoch": 5.3165021616771355,
      "grad_norm": 0.1567968726158142,
      "learning_rate": 4.116681803888925e-06,
      "loss": 0.054,
      "num_input_tokens_seen": 61750240,
      "step": 32590
    },
    {
      "epoch": 5.317317888897953,
      "grad_norm": 1.2261879444122314,
      "learning_rate": 4.111286308303605e-06,
      "loss": 0.1649,
      "num_input_tokens_seen": 61759920,
      "step": 32595
    },
    {
      "epoch": 5.31813361611877,
      "grad_norm": 1.085145115852356,
      "learning_rate": 4.105894034020433e-06,
      "loss": 0.314,
      "num_input_tokens_seen": 61770000,
      "step": 32600
    },
    {
      "epoch": 5.31813361611877,
      "eval_loss": 0.14071525633335114,
      "eval_runtime": 133.2381,
      "eval_samples_per_second": 20.452,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 61770000,
      "step": 32600
    },
    {
      "epoch": 5.318949343339587,
      "grad_norm": 1.5519332885742188,
      "learning_rate": 4.100504981870975e-06,
      "loss": 0.1763,
      "num_input_tokens_seen": 61780112,
      "step": 32605
    },
    {
      "epoch": 5.319765070560405,
      "grad_norm": 0.6825606822967529,
      "learning_rate": 4.0951191526862915e-06,
      "loss": 0.1495,
      "num_input_tokens_seen": 61790672,
      "step": 32610
    },
    {
      "epoch": 5.320580797781222,
      "grad_norm": 0.23897315561771393,
      "learning_rate": 4.089736547296938e-06,
      "loss": 0.0908,
      "num_input_tokens_seen": 61801120,
      "step": 32615
    },
    {
      "epoch": 5.321396525002039,
      "grad_norm": 0.4134387969970703,
      "learning_rate": 4.08435716653299e-06,
      "loss": 0.1232,
      "num_input_tokens_seen": 61810896,
      "step": 32620
    },
    {
      "epoch": 5.322212252222856,
      "grad_norm": 0.28881826996803284,
      "learning_rate": 4.0789810112240005e-06,
      "loss": 0.131,
      "num_input_tokens_seen": 61821088,
      "step": 32625
    },
    {
      "epoch": 5.323027979443674,
      "grad_norm": 0.34098753333091736,
      "learning_rate": 4.073608082199057e-06,
      "loss": 0.0859,
      "num_input_tokens_seen": 61829312,
      "step": 32630
    },
    {
      "epoch": 5.3238437066644915,
      "grad_norm": 0.2406250238418579,
      "learning_rate": 4.068238380286718e-06,
      "loss": 0.1111,
      "num_input_tokens_seen": 61839888,
      "step": 32635
    },
    {
      "epoch": 5.324659433885309,
      "grad_norm": 2.227555274963379,
      "learning_rate": 4.062871906315072e-06,
      "loss": 0.2551,
      "num_input_tokens_seen": 61850080,
      "step": 32640
    },
    {
      "epoch": 5.325475161106126,
      "grad_norm": 1.574413537979126,
      "learning_rate": 4.057508661111686e-06,
      "loss": 0.2468,
      "num_input_tokens_seen": 61859808,
      "step": 32645
    },
    {
      "epoch": 5.326290888326944,
      "grad_norm": 1.1653467416763306,
      "learning_rate": 4.052148645503648e-06,
      "loss": 0.2148,
      "num_input_tokens_seen": 61869440,
      "step": 32650
    },
    {
      "epoch": 5.327106615547761,
      "grad_norm": 0.7664217352867126,
      "learning_rate": 4.046791860317531e-06,
      "loss": 0.139,
      "num_input_tokens_seen": 61878512,
      "step": 32655
    },
    {
      "epoch": 5.327922342768578,
      "grad_norm": 0.14480696618556976,
      "learning_rate": 4.041438306379431e-06,
      "loss": 0.232,
      "num_input_tokens_seen": 61887328,
      "step": 32660
    },
    {
      "epoch": 5.328738069989395,
      "grad_norm": 0.2878437638282776,
      "learning_rate": 4.036087984514916e-06,
      "loss": 0.1823,
      "num_input_tokens_seen": 61897328,
      "step": 32665
    },
    {
      "epoch": 5.329553797210213,
      "grad_norm": 1.0595310926437378,
      "learning_rate": 4.030740895549084e-06,
      "loss": 0.1122,
      "num_input_tokens_seen": 61906064,
      "step": 32670
    },
    {
      "epoch": 5.33036952443103,
      "grad_norm": 0.1785901039838791,
      "learning_rate": 4.025397040306531e-06,
      "loss": 0.0689,
      "num_input_tokens_seen": 61914976,
      "step": 32675
    },
    {
      "epoch": 5.331185251651847,
      "grad_norm": 0.9548571705818176,
      "learning_rate": 4.0200564196113285e-06,
      "loss": 0.1215,
      "num_input_tokens_seen": 61925216,
      "step": 32680
    },
    {
      "epoch": 5.332000978872665,
      "grad_norm": 0.054345984011888504,
      "learning_rate": 4.014719034287079e-06,
      "loss": 0.0896,
      "num_input_tokens_seen": 61934688,
      "step": 32685
    },
    {
      "epoch": 5.332816706093483,
      "grad_norm": 1.1059792041778564,
      "learning_rate": 4.0093848851568775e-06,
      "loss": 0.1299,
      "num_input_tokens_seen": 61943376,
      "step": 32690
    },
    {
      "epoch": 5.3336324333143,
      "grad_norm": 1.4203555583953857,
      "learning_rate": 4.004053973043304e-06,
      "loss": 0.2019,
      "num_input_tokens_seen": 61954544,
      "step": 32695
    },
    {
      "epoch": 5.334448160535117,
      "grad_norm": 0.9311079382896423,
      "learning_rate": 3.998726298768465e-06,
      "loss": 0.1479,
      "num_input_tokens_seen": 61965232,
      "step": 32700
    },
    {
      "epoch": 5.335263887755934,
      "grad_norm": 0.8325363397598267,
      "learning_rate": 3.99340186315395e-06,
      "loss": 0.1723,
      "num_input_tokens_seen": 61975680,
      "step": 32705
    },
    {
      "epoch": 5.336079614976752,
      "grad_norm": 1.794826865196228,
      "learning_rate": 3.988080667020849e-06,
      "loss": 0.12,
      "num_input_tokens_seen": 61985152,
      "step": 32710
    },
    {
      "epoch": 5.336895342197569,
      "grad_norm": 0.05266950652003288,
      "learning_rate": 3.982762711189766e-06,
      "loss": 0.1238,
      "num_input_tokens_seen": 61994896,
      "step": 32715
    },
    {
      "epoch": 5.337711069418386,
      "grad_norm": 1.2123372554779053,
      "learning_rate": 3.977447996480785e-06,
      "loss": 0.326,
      "num_input_tokens_seen": 62003952,
      "step": 32720
    },
    {
      "epoch": 5.338526796639204,
      "grad_norm": 1.8618732690811157,
      "learning_rate": 3.97213652371351e-06,
      "loss": 0.1164,
      "num_input_tokens_seen": 62012944,
      "step": 32725
    },
    {
      "epoch": 5.339342523860021,
      "grad_norm": 0.06918603926897049,
      "learning_rate": 3.966828293707042e-06,
      "loss": 0.1203,
      "num_input_tokens_seen": 62022704,
      "step": 32730
    },
    {
      "epoch": 5.3401582510808385,
      "grad_norm": 0.3553656041622162,
      "learning_rate": 3.961523307279963e-06,
      "loss": 0.0624,
      "num_input_tokens_seen": 62030976,
      "step": 32735
    },
    {
      "epoch": 5.340973978301656,
      "grad_norm": 0.18959940969944,
      "learning_rate": 3.956221565250382e-06,
      "loss": 0.0917,
      "num_input_tokens_seen": 62040816,
      "step": 32740
    },
    {
      "epoch": 5.341789705522474,
      "grad_norm": 1.8469829559326172,
      "learning_rate": 3.950923068435883e-06,
      "loss": 0.1691,
      "num_input_tokens_seen": 62049712,
      "step": 32745
    },
    {
      "epoch": 5.342605432743291,
      "grad_norm": 1.7174665927886963,
      "learning_rate": 3.945627817653566e-06,
      "loss": 0.1665,
      "num_input_tokens_seen": 62058720,
      "step": 32750
    },
    {
      "epoch": 5.343421159964108,
      "grad_norm": 1.3003838062286377,
      "learning_rate": 3.9403358137200335e-06,
      "loss": 0.1981,
      "num_input_tokens_seen": 62068384,
      "step": 32755
    },
    {
      "epoch": 5.344236887184925,
      "grad_norm": 0.521844744682312,
      "learning_rate": 3.9350470574513605e-06,
      "loss": 0.0476,
      "num_input_tokens_seen": 62078224,
      "step": 32760
    },
    {
      "epoch": 5.345052614405743,
      "grad_norm": 0.30628666281700134,
      "learning_rate": 3.9297615496631525e-06,
      "loss": 0.0422,
      "num_input_tokens_seen": 62087360,
      "step": 32765
    },
    {
      "epoch": 5.34586834162656,
      "grad_norm": 0.9806636571884155,
      "learning_rate": 3.924479291170505e-06,
      "loss": 0.0651,
      "num_input_tokens_seen": 62097120,
      "step": 32770
    },
    {
      "epoch": 5.346684068847377,
      "grad_norm": 0.07161558419466019,
      "learning_rate": 3.919200282788002e-06,
      "loss": 0.0821,
      "num_input_tokens_seen": 62106784,
      "step": 32775
    },
    {
      "epoch": 5.3474997960681945,
      "grad_norm": 1.505690097808838,
      "learning_rate": 3.913924525329726e-06,
      "loss": 0.0888,
      "num_input_tokens_seen": 62117616,
      "step": 32780
    },
    {
      "epoch": 5.3483155232890125,
      "grad_norm": 0.3675982356071472,
      "learning_rate": 3.908652019609279e-06,
      "loss": 0.1543,
      "num_input_tokens_seen": 62128016,
      "step": 32785
    },
    {
      "epoch": 5.34913125050983,
      "grad_norm": 1.2420765161514282,
      "learning_rate": 3.9033827664397364e-06,
      "loss": 0.1454,
      "num_input_tokens_seen": 62136672,
      "step": 32790
    },
    {
      "epoch": 5.349946977730647,
      "grad_norm": 0.7707218527793884,
      "learning_rate": 3.898116766633694e-06,
      "loss": 0.1284,
      "num_input_tokens_seen": 62146000,
      "step": 32795
    },
    {
      "epoch": 5.350762704951464,
      "grad_norm": 1.0771148204803467,
      "learning_rate": 3.8928540210032225e-06,
      "loss": 0.1222,
      "num_input_tokens_seen": 62154640,
      "step": 32800
    },
    {
      "epoch": 5.350762704951464,
      "eval_loss": 0.14029406011104584,
      "eval_runtime": 133.3045,
      "eval_samples_per_second": 20.442,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 62154640,
      "step": 32800
    },
    {
      "epoch": 5.351578432172282,
      "grad_norm": 0.16951341927051544,
      "learning_rate": 3.887594530359909e-06,
      "loss": 0.1125,
      "num_input_tokens_seen": 62164352,
      "step": 32805
    },
    {
      "epoch": 5.352394159393099,
      "grad_norm": 1.112378478050232,
      "learning_rate": 3.88233829551484e-06,
      "loss": 0.0638,
      "num_input_tokens_seen": 62173280,
      "step": 32810
    },
    {
      "epoch": 5.353209886613916,
      "grad_norm": 1.2404729127883911,
      "learning_rate": 3.877085317278581e-06,
      "loss": 0.1673,
      "num_input_tokens_seen": 62183344,
      "step": 32815
    },
    {
      "epoch": 5.354025613834733,
      "grad_norm": 0.6030818819999695,
      "learning_rate": 3.87183559646122e-06,
      "loss": 0.1092,
      "num_input_tokens_seen": 62192896,
      "step": 32820
    },
    {
      "epoch": 5.354841341055551,
      "grad_norm": 0.6684694290161133,
      "learning_rate": 3.866589133872317e-06,
      "loss": 0.138,
      "num_input_tokens_seen": 62202288,
      "step": 32825
    },
    {
      "epoch": 5.3556570682763684,
      "grad_norm": 0.6790863275527954,
      "learning_rate": 3.861345930320948e-06,
      "loss": 0.0972,
      "num_input_tokens_seen": 62212192,
      "step": 32830
    },
    {
      "epoch": 5.356472795497186,
      "grad_norm": 1.3350054025650024,
      "learning_rate": 3.856105986615688e-06,
      "loss": 0.2553,
      "num_input_tokens_seen": 62221200,
      "step": 32835
    },
    {
      "epoch": 5.357288522718003,
      "grad_norm": 0.15433108806610107,
      "learning_rate": 3.850869303564589e-06,
      "loss": 0.1703,
      "num_input_tokens_seen": 62231248,
      "step": 32840
    },
    {
      "epoch": 5.358104249938821,
      "grad_norm": 0.783310055732727,
      "learning_rate": 3.845635881975226e-06,
      "loss": 0.1502,
      "num_input_tokens_seen": 62240416,
      "step": 32845
    },
    {
      "epoch": 5.358919977159638,
      "grad_norm": 1.5347096920013428,
      "learning_rate": 3.840405722654647e-06,
      "loss": 0.0817,
      "num_input_tokens_seen": 62249696,
      "step": 32850
    },
    {
      "epoch": 5.359735704380455,
      "grad_norm": 1.1742736101150513,
      "learning_rate": 3.835178826409419e-06,
      "loss": 0.2146,
      "num_input_tokens_seen": 62259312,
      "step": 32855
    },
    {
      "epoch": 5.360551431601272,
      "grad_norm": 0.0760822668671608,
      "learning_rate": 3.8299551940455895e-06,
      "loss": 0.0492,
      "num_input_tokens_seen": 62269568,
      "step": 32860
    },
    {
      "epoch": 5.36136715882209,
      "grad_norm": 0.539496123790741,
      "learning_rate": 3.824734826368703e-06,
      "loss": 0.0929,
      "num_input_tokens_seen": 62279040,
      "step": 32865
    },
    {
      "epoch": 5.362182886042907,
      "grad_norm": 0.042892154306173325,
      "learning_rate": 3.819517724183813e-06,
      "loss": 0.0954,
      "num_input_tokens_seen": 62288944,
      "step": 32870
    },
    {
      "epoch": 5.362998613263724,
      "grad_norm": 0.8936642408370972,
      "learning_rate": 3.8143038882954648e-06,
      "loss": 0.0962,
      "num_input_tokens_seen": 62299776,
      "step": 32875
    },
    {
      "epoch": 5.3638143404845415,
      "grad_norm": 0.27042439579963684,
      "learning_rate": 3.8090933195076867e-06,
      "loss": 0.0406,
      "num_input_tokens_seen": 62308464,
      "step": 32880
    },
    {
      "epoch": 5.3646300677053596,
      "grad_norm": 2.178851842880249,
      "learning_rate": 3.8038860186240198e-06,
      "loss": 0.0989,
      "num_input_tokens_seen": 62318400,
      "step": 32885
    },
    {
      "epoch": 5.365445794926177,
      "grad_norm": 0.9919695854187012,
      "learning_rate": 3.7986819864475026e-06,
      "loss": 0.1895,
      "num_input_tokens_seen": 62327808,
      "step": 32890
    },
    {
      "epoch": 5.366261522146994,
      "grad_norm": 0.8715391755104065,
      "learning_rate": 3.793481223780651e-06,
      "loss": 0.0616,
      "num_input_tokens_seen": 62336688,
      "step": 32895
    },
    {
      "epoch": 5.367077249367812,
      "grad_norm": 0.51410311460495,
      "learning_rate": 3.788283731425496e-06,
      "loss": 0.187,
      "num_input_tokens_seen": 62346480,
      "step": 32900
    },
    {
      "epoch": 5.367892976588629,
      "grad_norm": 2.090265989303589,
      "learning_rate": 3.7830895101835488e-06,
      "loss": 0.215,
      "num_input_tokens_seen": 62357360,
      "step": 32905
    },
    {
      "epoch": 5.368708703809446,
      "grad_norm": 1.6905087232589722,
      "learning_rate": 3.7778985608558274e-06,
      "loss": 0.0966,
      "num_input_tokens_seen": 62366864,
      "step": 32910
    },
    {
      "epoch": 5.369524431030263,
      "grad_norm": 0.3743521273136139,
      "learning_rate": 3.7727108842428443e-06,
      "loss": 0.057,
      "num_input_tokens_seen": 62375936,
      "step": 32915
    },
    {
      "epoch": 5.370340158251081,
      "grad_norm": 0.11455821245908737,
      "learning_rate": 3.7675264811446065e-06,
      "loss": 0.1418,
      "num_input_tokens_seen": 62385008,
      "step": 32920
    },
    {
      "epoch": 5.371155885471898,
      "grad_norm": 0.4435218870639801,
      "learning_rate": 3.7623453523605994e-06,
      "loss": 0.2229,
      "num_input_tokens_seen": 62394528,
      "step": 32925
    },
    {
      "epoch": 5.3719716126927155,
      "grad_norm": 0.31227362155914307,
      "learning_rate": 3.757167498689834e-06,
      "loss": 0.2688,
      "num_input_tokens_seen": 62403408,
      "step": 32930
    },
    {
      "epoch": 5.372787339913533,
      "grad_norm": 0.06548987329006195,
      "learning_rate": 3.7519929209307914e-06,
      "loss": 0.0845,
      "num_input_tokens_seen": 62412624,
      "step": 32935
    },
    {
      "epoch": 5.373603067134351,
      "grad_norm": 0.4334385097026825,
      "learning_rate": 3.746821619881463e-06,
      "loss": 0.1448,
      "num_input_tokens_seen": 62422416,
      "step": 32940
    },
    {
      "epoch": 5.374418794355168,
      "grad_norm": 0.0699172392487526,
      "learning_rate": 3.74165359633932e-06,
      "loss": 0.146,
      "num_input_tokens_seen": 62430816,
      "step": 32945
    },
    {
      "epoch": 5.375234521575985,
      "grad_norm": 0.17754092812538147,
      "learning_rate": 3.736488851101341e-06,
      "loss": 0.1511,
      "num_input_tokens_seen": 62440720,
      "step": 32950
    },
    {
      "epoch": 5.376050248796802,
      "grad_norm": 0.2533603012561798,
      "learning_rate": 3.7313273849640035e-06,
      "loss": 0.0539,
      "num_input_tokens_seen": 62450160,
      "step": 32955
    },
    {
      "epoch": 5.37686597601762,
      "grad_norm": 0.1563243567943573,
      "learning_rate": 3.7261691987232533e-06,
      "loss": 0.0801,
      "num_input_tokens_seen": 62461536,
      "step": 32960
    },
    {
      "epoch": 5.377681703238437,
      "grad_norm": 0.1340327113866806,
      "learning_rate": 3.7210142931745575e-06,
      "loss": 0.0479,
      "num_input_tokens_seen": 62470464,
      "step": 32965
    },
    {
      "epoch": 5.378497430459254,
      "grad_norm": 0.3411330580711365,
      "learning_rate": 3.7158626691128712e-06,
      "loss": 0.111,
      "num_input_tokens_seen": 62480368,
      "step": 32970
    },
    {
      "epoch": 5.3793131576800715,
      "grad_norm": 0.1835961639881134,
      "learning_rate": 3.710714327332629e-06,
      "loss": 0.0639,
      "num_input_tokens_seen": 62491056,
      "step": 32975
    },
    {
      "epoch": 5.3801288849008895,
      "grad_norm": 0.17736975848674774,
      "learning_rate": 3.7055692686277815e-06,
      "loss": 0.0479,
      "num_input_tokens_seen": 62502224,
      "step": 32980
    },
    {
      "epoch": 5.380944612121707,
      "grad_norm": 0.21450011432170868,
      "learning_rate": 3.70042749379175e-06,
      "loss": 0.0534,
      "num_input_tokens_seen": 62512192,
      "step": 32985
    },
    {
      "epoch": 5.381760339342524,
      "grad_norm": 0.06199850142002106,
      "learning_rate": 3.6952890036174693e-06,
      "loss": 0.046,
      "num_input_tokens_seen": 62521952,
      "step": 32990
    },
    {
      "epoch": 5.382576066563341,
      "grad_norm": 0.10544704645872116,
      "learning_rate": 3.690153798897353e-06,
      "loss": 0.0808,
      "num_input_tokens_seen": 62531968,
      "step": 32995
    },
    {
      "epoch": 5.383391793784159,
      "grad_norm": 0.8271560072898865,
      "learning_rate": 3.6850218804233225e-06,
      "loss": 0.1365,
      "num_input_tokens_seen": 62541664,
      "step": 33000
    },
    {
      "epoch": 5.383391793784159,
      "eval_loss": 0.14066888391971588,
      "eval_runtime": 133.303,
      "eval_samples_per_second": 20.442,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 62541664,
      "step": 33000
    },
    {
      "epoch": 5.384207521004976,
      "grad_norm": 1.4110770225524902,
      "learning_rate": 3.679893248986779e-06,
      "loss": 0.1913,
      "num_input_tokens_seen": 62551008,
      "step": 33005
    },
    {
      "epoch": 5.385023248225793,
      "grad_norm": 0.5529909133911133,
      "learning_rate": 3.6747679053786147e-06,
      "loss": 0.14,
      "num_input_tokens_seen": 62560384,
      "step": 33010
    },
    {
      "epoch": 5.38583897544661,
      "grad_norm": 0.20159374177455902,
      "learning_rate": 3.669645850389228e-06,
      "loss": 0.1245,
      "num_input_tokens_seen": 62569120,
      "step": 33015
    },
    {
      "epoch": 5.386654702667428,
      "grad_norm": 0.5096997022628784,
      "learning_rate": 3.664527084808514e-06,
      "loss": 0.0414,
      "num_input_tokens_seen": 62578592,
      "step": 33020
    },
    {
      "epoch": 5.387470429888245,
      "grad_norm": 0.0988834872841835,
      "learning_rate": 3.6594116094258337e-06,
      "loss": 0.1418,
      "num_input_tokens_seen": 62587792,
      "step": 33025
    },
    {
      "epoch": 5.388286157109063,
      "grad_norm": 0.3762648105621338,
      "learning_rate": 3.6542994250300665e-06,
      "loss": 0.1219,
      "num_input_tokens_seen": 62597840,
      "step": 33030
    },
    {
      "epoch": 5.38910188432988,
      "grad_norm": 0.47295063734054565,
      "learning_rate": 3.6491905324095825e-06,
      "loss": 0.2122,
      "num_input_tokens_seen": 62608048,
      "step": 33035
    },
    {
      "epoch": 5.389917611550698,
      "grad_norm": 0.17738226056098938,
      "learning_rate": 3.644084932352221e-06,
      "loss": 0.1013,
      "num_input_tokens_seen": 62617376,
      "step": 33040
    },
    {
      "epoch": 5.390733338771515,
      "grad_norm": 0.4282730519771576,
      "learning_rate": 3.6389826256453457e-06,
      "loss": 0.1634,
      "num_input_tokens_seen": 62627776,
      "step": 33045
    },
    {
      "epoch": 5.391549065992332,
      "grad_norm": 0.2815495729446411,
      "learning_rate": 3.633883613075781e-06,
      "loss": 0.2053,
      "num_input_tokens_seen": 62636752,
      "step": 33050
    },
    {
      "epoch": 5.392364793213149,
      "grad_norm": 0.30943402647972107,
      "learning_rate": 3.6287878954298693e-06,
      "loss": 0.089,
      "num_input_tokens_seen": 62646624,
      "step": 33055
    },
    {
      "epoch": 5.393180520433967,
      "grad_norm": 0.799969494342804,
      "learning_rate": 3.6236954734934354e-06,
      "loss": 0.0598,
      "num_input_tokens_seen": 62655952,
      "step": 33060
    },
    {
      "epoch": 5.393996247654784,
      "grad_norm": 0.6099685430526733,
      "learning_rate": 3.618606348051784e-06,
      "loss": 0.0931,
      "num_input_tokens_seen": 62663872,
      "step": 33065
    },
    {
      "epoch": 5.394811974875601,
      "grad_norm": 0.3378593623638153,
      "learning_rate": 3.6135205198897376e-06,
      "loss": 0.0291,
      "num_input_tokens_seen": 62673600,
      "step": 33070
    },
    {
      "epoch": 5.395627702096419,
      "grad_norm": 1.339013695716858,
      "learning_rate": 3.6084379897915854e-06,
      "loss": 0.1217,
      "num_input_tokens_seen": 62683552,
      "step": 33075
    },
    {
      "epoch": 5.3964434293172365,
      "grad_norm": 0.6815502047538757,
      "learning_rate": 3.6033587585411115e-06,
      "loss": 0.037,
      "num_input_tokens_seen": 62691968,
      "step": 33080
    },
    {
      "epoch": 5.397259156538054,
      "grad_norm": 0.18286092579364777,
      "learning_rate": 3.5982828269216117e-06,
      "loss": 0.1047,
      "num_input_tokens_seen": 62701872,
      "step": 33085
    },
    {
      "epoch": 5.398074883758871,
      "grad_norm": 1.3024115562438965,
      "learning_rate": 3.593210195715843e-06,
      "loss": 0.3177,
      "num_input_tokens_seen": 62711776,
      "step": 33090
    },
    {
      "epoch": 5.398890610979688,
      "grad_norm": 0.14395743608474731,
      "learning_rate": 3.5881408657060773e-06,
      "loss": 0.1705,
      "num_input_tokens_seen": 62721888,
      "step": 33095
    },
    {
      "epoch": 5.399706338200506,
      "grad_norm": 0.2729150950908661,
      "learning_rate": 3.583074837674075e-06,
      "loss": 0.0392,
      "num_input_tokens_seen": 62730464,
      "step": 33100
    },
    {
      "epoch": 5.400522065421323,
      "grad_norm": 0.6599184274673462,
      "learning_rate": 3.578012112401069e-06,
      "loss": 0.0894,
      "num_input_tokens_seen": 62739536,
      "step": 33105
    },
    {
      "epoch": 5.40133779264214,
      "grad_norm": 0.27862679958343506,
      "learning_rate": 3.5729526906677996e-06,
      "loss": 0.1024,
      "num_input_tokens_seen": 62749312,
      "step": 33110
    },
    {
      "epoch": 5.402153519862958,
      "grad_norm": 0.18005037307739258,
      "learning_rate": 3.5678965732545007e-06,
      "loss": 0.1398,
      "num_input_tokens_seen": 62758496,
      "step": 33115
    },
    {
      "epoch": 5.402969247083775,
      "grad_norm": 0.4043392539024353,
      "learning_rate": 3.562843760940876e-06,
      "loss": 0.0732,
      "num_input_tokens_seen": 62766896,
      "step": 33120
    },
    {
      "epoch": 5.4037849743045925,
      "grad_norm": 0.9807651042938232,
      "learning_rate": 3.5577942545061473e-06,
      "loss": 0.1185,
      "num_input_tokens_seen": 62775792,
      "step": 33125
    },
    {
      "epoch": 5.40460070152541,
      "grad_norm": 2.9854519367218018,
      "learning_rate": 3.5527480547289967e-06,
      "loss": 0.2297,
      "num_input_tokens_seen": 62785424,
      "step": 33130
    },
    {
      "epoch": 5.405416428746228,
      "grad_norm": 0.11323998868465424,
      "learning_rate": 3.547705162387624e-06,
      "loss": 0.1228,
      "num_input_tokens_seen": 62794624,
      "step": 33135
    },
    {
      "epoch": 5.406232155967045,
      "grad_norm": 0.15345382690429688,
      "learning_rate": 3.542665578259699e-06,
      "loss": 0.0899,
      "num_input_tokens_seen": 62804032,
      "step": 33140
    },
    {
      "epoch": 5.407047883187862,
      "grad_norm": 0.0398428738117218,
      "learning_rate": 3.5376293031223945e-06,
      "loss": 0.169,
      "num_input_tokens_seen": 62813920,
      "step": 33145
    },
    {
      "epoch": 5.407863610408679,
      "grad_norm": 0.18377456068992615,
      "learning_rate": 3.5325963377523614e-06,
      "loss": 0.0859,
      "num_input_tokens_seen": 62821280,
      "step": 33150
    },
    {
      "epoch": 5.408679337629497,
      "grad_norm": 0.05116834491491318,
      "learning_rate": 3.5275666829257536e-06,
      "loss": 0.2551,
      "num_input_tokens_seen": 62830368,
      "step": 33155
    },
    {
      "epoch": 5.409495064850314,
      "grad_norm": 0.7018431425094604,
      "learning_rate": 3.5225403394181955e-06,
      "loss": 0.1326,
      "num_input_tokens_seen": 62839456,
      "step": 33160
    },
    {
      "epoch": 5.410310792071131,
      "grad_norm": 0.6493010520935059,
      "learning_rate": 3.517517308004828e-06,
      "loss": 0.0543,
      "num_input_tokens_seen": 62848608,
      "step": 33165
    },
    {
      "epoch": 5.411126519291948,
      "grad_norm": 0.10965611040592194,
      "learning_rate": 3.512497589460251e-06,
      "loss": 0.1854,
      "num_input_tokens_seen": 62858128,
      "step": 33170
    },
    {
      "epoch": 5.4119422465127665,
      "grad_norm": 0.4178744852542877,
      "learning_rate": 3.5074811845585727e-06,
      "loss": 0.0403,
      "num_input_tokens_seen": 62866752,
      "step": 33175
    },
    {
      "epoch": 5.412757973733584,
      "grad_norm": 0.3662075698375702,
      "learning_rate": 3.5024680940733937e-06,
      "loss": 0.0971,
      "num_input_tokens_seen": 62876576,
      "step": 33180
    },
    {
      "epoch": 5.413573700954401,
      "grad_norm": 1.6002675294876099,
      "learning_rate": 3.4974583187777852e-06,
      "loss": 0.1364,
      "num_input_tokens_seen": 62885760,
      "step": 33185
    },
    {
      "epoch": 5.414389428175218,
      "grad_norm": 0.10180404037237167,
      "learning_rate": 3.4924518594443204e-06,
      "loss": 0.0422,
      "num_input_tokens_seen": 62895984,
      "step": 33190
    },
    {
      "epoch": 5.415205155396036,
      "grad_norm": 1.1983014345169067,
      "learning_rate": 3.4874487168450682e-06,
      "loss": 0.1509,
      "num_input_tokens_seen": 62904096,
      "step": 33195
    },
    {
      "epoch": 5.416020882616853,
      "grad_norm": 0.32532912492752075,
      "learning_rate": 3.482448891751558e-06,
      "loss": 0.0645,
      "num_input_tokens_seen": 62912976,
      "step": 33200
    },
    {
      "epoch": 5.416020882616853,
      "eval_loss": 0.14185647666454315,
      "eval_runtime": 133.2332,
      "eval_samples_per_second": 20.453,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 62912976,
      "step": 33200
    },
    {
      "epoch": 5.41683660983767,
      "grad_norm": 0.9101196527481079,
      "learning_rate": 3.477452384934843e-06,
      "loss": 0.074,
      "num_input_tokens_seen": 62922112,
      "step": 33205
    },
    {
      "epoch": 5.417652337058487,
      "grad_norm": 0.6575267910957336,
      "learning_rate": 3.472459197165434e-06,
      "loss": 0.0742,
      "num_input_tokens_seen": 62931136,
      "step": 33210
    },
    {
      "epoch": 5.418468064279305,
      "grad_norm": 1.2451024055480957,
      "learning_rate": 3.4674693292133518e-06,
      "loss": 0.1491,
      "num_input_tokens_seen": 62940384,
      "step": 33215
    },
    {
      "epoch": 5.419283791500122,
      "grad_norm": 0.33829638361930847,
      "learning_rate": 3.4624827818480977e-06,
      "loss": 0.243,
      "num_input_tokens_seen": 62948224,
      "step": 33220
    },
    {
      "epoch": 5.4200995187209395,
      "grad_norm": 0.5968812704086304,
      "learning_rate": 3.4574995558386474e-06,
      "loss": 0.0736,
      "num_input_tokens_seen": 62957904,
      "step": 33225
    },
    {
      "epoch": 5.420915245941757,
      "grad_norm": 0.3859690725803375,
      "learning_rate": 3.452519651953487e-06,
      "loss": 0.1183,
      "num_input_tokens_seen": 62968384,
      "step": 33230
    },
    {
      "epoch": 5.421730973162575,
      "grad_norm": 1.3384408950805664,
      "learning_rate": 3.447543070960585e-06,
      "loss": 0.2293,
      "num_input_tokens_seen": 62977376,
      "step": 33235
    },
    {
      "epoch": 5.422546700383392,
      "grad_norm": 0.3761911690235138,
      "learning_rate": 3.4425698136273778e-06,
      "loss": 0.0584,
      "num_input_tokens_seen": 62986912,
      "step": 33240
    },
    {
      "epoch": 5.423362427604209,
      "grad_norm": 0.37048494815826416,
      "learning_rate": 3.437599880720821e-06,
      "loss": 0.1354,
      "num_input_tokens_seen": 62995968,
      "step": 33245
    },
    {
      "epoch": 5.424178154825027,
      "grad_norm": 0.5155497789382935,
      "learning_rate": 3.4326332730073267e-06,
      "loss": 0.2109,
      "num_input_tokens_seen": 63005696,
      "step": 33250
    },
    {
      "epoch": 5.424993882045844,
      "grad_norm": 0.15770287811756134,
      "learning_rate": 3.427669991252813e-06,
      "loss": 0.1209,
      "num_input_tokens_seen": 63016064,
      "step": 33255
    },
    {
      "epoch": 5.425809609266661,
      "grad_norm": 0.9647607207298279,
      "learning_rate": 3.42271003622269e-06,
      "loss": 0.1324,
      "num_input_tokens_seen": 63025888,
      "step": 33260
    },
    {
      "epoch": 5.426625336487478,
      "grad_norm": 1.3944706916809082,
      "learning_rate": 3.4177534086818286e-06,
      "loss": 0.1482,
      "num_input_tokens_seen": 63034912,
      "step": 33265
    },
    {
      "epoch": 5.4274410637082955,
      "grad_norm": 0.9225355386734009,
      "learning_rate": 3.412800109394612e-06,
      "loss": 0.1683,
      "num_input_tokens_seen": 63044624,
      "step": 33270
    },
    {
      "epoch": 5.4282567909291135,
      "grad_norm": 1.2687292098999023,
      "learning_rate": 3.4078501391249044e-06,
      "loss": 0.1555,
      "num_input_tokens_seen": 63054304,
      "step": 33275
    },
    {
      "epoch": 5.429072518149931,
      "grad_norm": 0.40451183915138245,
      "learning_rate": 3.4029034986360453e-06,
      "loss": 0.0391,
      "num_input_tokens_seen": 63063536,
      "step": 33280
    },
    {
      "epoch": 5.429888245370748,
      "grad_norm": 0.17614704370498657,
      "learning_rate": 3.397960188690877e-06,
      "loss": 0.0535,
      "num_input_tokens_seen": 63072688,
      "step": 33285
    },
    {
      "epoch": 5.430703972591566,
      "grad_norm": 0.06321921199560165,
      "learning_rate": 3.393020210051717e-06,
      "loss": 0.1028,
      "num_input_tokens_seen": 63081136,
      "step": 33290
    },
    {
      "epoch": 5.431519699812383,
      "grad_norm": 0.7228341102600098,
      "learning_rate": 3.3880835634803655e-06,
      "loss": 0.1164,
      "num_input_tokens_seen": 63090512,
      "step": 33295
    },
    {
      "epoch": 5.4323354270332,
      "grad_norm": 0.18225044012069702,
      "learning_rate": 3.383150249738126e-06,
      "loss": 0.091,
      "num_input_tokens_seen": 63099472,
      "step": 33300
    },
    {
      "epoch": 5.433151154254017,
      "grad_norm": 0.05272641032934189,
      "learning_rate": 3.3782202695857663e-06,
      "loss": 0.1303,
      "num_input_tokens_seen": 63108528,
      "step": 33305
    },
    {
      "epoch": 5.433966881474835,
      "grad_norm": 0.899839460849762,
      "learning_rate": 3.373293623783558e-06,
      "loss": 0.081,
      "num_input_tokens_seen": 63119248,
      "step": 33310
    },
    {
      "epoch": 5.434782608695652,
      "grad_norm": 0.33731380105018616,
      "learning_rate": 3.368370313091257e-06,
      "loss": 0.062,
      "num_input_tokens_seen": 63128256,
      "step": 33315
    },
    {
      "epoch": 5.4355983359164695,
      "grad_norm": 0.29081210494041443,
      "learning_rate": 3.363450338268087e-06,
      "loss": 0.1338,
      "num_input_tokens_seen": 63137744,
      "step": 33320
    },
    {
      "epoch": 5.436414063137287,
      "grad_norm": 0.12169547379016876,
      "learning_rate": 3.358533700072783e-06,
      "loss": 0.1675,
      "num_input_tokens_seen": 63146688,
      "step": 33325
    },
    {
      "epoch": 5.437229790358105,
      "grad_norm": 0.20344814658164978,
      "learning_rate": 3.3536203992635377e-06,
      "loss": 0.0685,
      "num_input_tokens_seen": 63156528,
      "step": 33330
    },
    {
      "epoch": 5.438045517578922,
      "grad_norm": 0.2764643430709839,
      "learning_rate": 3.348710436598057e-06,
      "loss": 0.1017,
      "num_input_tokens_seen": 63166032,
      "step": 33335
    },
    {
      "epoch": 5.438861244799739,
      "grad_norm": 0.29783865809440613,
      "learning_rate": 3.3438038128335155e-06,
      "loss": 0.0436,
      "num_input_tokens_seen": 63175392,
      "step": 33340
    },
    {
      "epoch": 5.439676972020556,
      "grad_norm": 1.6824679374694824,
      "learning_rate": 3.338900528726571e-06,
      "loss": 0.3147,
      "num_input_tokens_seen": 63185312,
      "step": 33345
    },
    {
      "epoch": 5.440492699241374,
      "grad_norm": 1.1783572435379028,
      "learning_rate": 3.3340005850333812e-06,
      "loss": 0.2511,
      "num_input_tokens_seen": 63195584,
      "step": 33350
    },
    {
      "epoch": 5.441308426462191,
      "grad_norm": 0.08065398782491684,
      "learning_rate": 3.329103982509568e-06,
      "loss": 0.0375,
      "num_input_tokens_seen": 63205184,
      "step": 33355
    },
    {
      "epoch": 5.442124153683008,
      "grad_norm": 0.2624196708202362,
      "learning_rate": 3.324210721910259e-06,
      "loss": 0.0635,
      "num_input_tokens_seen": 63214432,
      "step": 33360
    },
    {
      "epoch": 5.442939880903825,
      "grad_norm": 0.7812062501907349,
      "learning_rate": 3.319320803990053e-06,
      "loss": 0.1941,
      "num_input_tokens_seen": 63224208,
      "step": 33365
    },
    {
      "epoch": 5.443755608124643,
      "grad_norm": 0.09038634598255157,
      "learning_rate": 3.3144342295030274e-06,
      "loss": 0.0473,
      "num_input_tokens_seen": 63233952,
      "step": 33370
    },
    {
      "epoch": 5.444571335345461,
      "grad_norm": 0.6063143610954285,
      "learning_rate": 3.309550999202765e-06,
      "loss": 0.0907,
      "num_input_tokens_seen": 63244000,
      "step": 33375
    },
    {
      "epoch": 5.445387062566278,
      "grad_norm": 0.10301809012889862,
      "learning_rate": 3.3046711138423197e-06,
      "loss": 0.1602,
      "num_input_tokens_seen": 63252064,
      "step": 33380
    },
    {
      "epoch": 5.446202789787095,
      "grad_norm": 0.19574685394763947,
      "learning_rate": 3.2997945741742255e-06,
      "loss": 0.121,
      "num_input_tokens_seen": 63262416,
      "step": 33385
    },
    {
      "epoch": 5.447018517007913,
      "grad_norm": 0.41348040103912354,
      "learning_rate": 3.2949213809505082e-06,
      "loss": 0.174,
      "num_input_tokens_seen": 63271440,
      "step": 33390
    },
    {
      "epoch": 5.44783424422873,
      "grad_norm": 2.1192259788513184,
      "learning_rate": 3.2900515349226834e-06,
      "loss": 0.1511,
      "num_input_tokens_seen": 63280976,
      "step": 33395
    },
    {
      "epoch": 5.448649971449547,
      "grad_norm": 0.5224822163581848,
      "learning_rate": 3.285185036841731e-06,
      "loss": 0.181,
      "num_input_tokens_seen": 63289520,
      "step": 33400
    },
    {
      "epoch": 5.448649971449547,
      "eval_loss": 0.14167185127735138,
      "eval_runtime": 133.3165,
      "eval_samples_per_second": 20.44,
      "eval_steps_per_second": 10.224,
      "num_input_tokens_seen": 63289520,
      "step": 33400
    },
    {
      "epoch": 5.449465698670364,
      "grad_norm": 0.12995991110801697,
      "learning_rate": 3.2803218874581377e-06,
      "loss": 0.0518,
      "num_input_tokens_seen": 63298400,
      "step": 33405
    },
    {
      "epoch": 5.450281425891182,
      "grad_norm": 0.6340125203132629,
      "learning_rate": 3.2754620875218494e-06,
      "loss": 0.064,
      "num_input_tokens_seen": 63307808,
      "step": 33410
    },
    {
      "epoch": 5.451097153111999,
      "grad_norm": 0.7217298150062561,
      "learning_rate": 3.2706056377823146e-06,
      "loss": 0.0823,
      "num_input_tokens_seen": 63316560,
      "step": 33415
    },
    {
      "epoch": 5.4519128803328165,
      "grad_norm": 0.20778027176856995,
      "learning_rate": 3.2657525389884647e-06,
      "loss": 0.0699,
      "num_input_tokens_seen": 63326288,
      "step": 33420
    },
    {
      "epoch": 5.4527286075536345,
      "grad_norm": 0.19461703300476074,
      "learning_rate": 3.260902791888698e-06,
      "loss": 0.0686,
      "num_input_tokens_seen": 63335728,
      "step": 33425
    },
    {
      "epoch": 5.453544334774452,
      "grad_norm": 0.6649834513664246,
      "learning_rate": 3.2560563972309166e-06,
      "loss": 0.1864,
      "num_input_tokens_seen": 63345168,
      "step": 33430
    },
    {
      "epoch": 5.454360061995269,
      "grad_norm": 0.3506159782409668,
      "learning_rate": 3.251213355762489e-06,
      "loss": 0.1082,
      "num_input_tokens_seen": 63355440,
      "step": 33435
    },
    {
      "epoch": 5.455175789216086,
      "grad_norm": 1.129103422164917,
      "learning_rate": 3.2463736682302707e-06,
      "loss": 0.1935,
      "num_input_tokens_seen": 63365088,
      "step": 33440
    },
    {
      "epoch": 5.455991516436903,
      "grad_norm": 0.5016814470291138,
      "learning_rate": 3.2415373353806124e-06,
      "loss": 0.0905,
      "num_input_tokens_seen": 63374240,
      "step": 33445
    },
    {
      "epoch": 5.456807243657721,
      "grad_norm": 0.9339399337768555,
      "learning_rate": 3.236704357959322e-06,
      "loss": 0.091,
      "num_input_tokens_seen": 63383120,
      "step": 33450
    },
    {
      "epoch": 5.457622970878538,
      "grad_norm": 1.3504002094268799,
      "learning_rate": 3.2318747367117154e-06,
      "loss": 0.2091,
      "num_input_tokens_seen": 63392128,
      "step": 33455
    },
    {
      "epoch": 5.458438698099355,
      "grad_norm": 0.0910894125699997,
      "learning_rate": 3.227048472382585e-06,
      "loss": 0.1082,
      "num_input_tokens_seen": 63400864,
      "step": 33460
    },
    {
      "epoch": 5.459254425320173,
      "grad_norm": 0.12993043661117554,
      "learning_rate": 3.2222255657161915e-06,
      "loss": 0.0957,
      "num_input_tokens_seen": 63409824,
      "step": 33465
    },
    {
      "epoch": 5.4600701525409905,
      "grad_norm": 1.2944865226745605,
      "learning_rate": 3.2174060174562924e-06,
      "loss": 0.1365,
      "num_input_tokens_seen": 63419824,
      "step": 33470
    },
    {
      "epoch": 5.460885879761808,
      "grad_norm": 1.158398985862732,
      "learning_rate": 3.2125898283461298e-06,
      "loss": 0.0764,
      "num_input_tokens_seen": 63429712,
      "step": 33475
    },
    {
      "epoch": 5.461701606982625,
      "grad_norm": 0.9607349038124084,
      "learning_rate": 3.207776999128406e-06,
      "loss": 0.1266,
      "num_input_tokens_seen": 63439360,
      "step": 33480
    },
    {
      "epoch": 5.462517334203443,
      "grad_norm": 0.9867411255836487,
      "learning_rate": 3.202967530545331e-06,
      "loss": 0.222,
      "num_input_tokens_seen": 63449696,
      "step": 33485
    },
    {
      "epoch": 5.46333306142426,
      "grad_norm": 1.1485878229141235,
      "learning_rate": 3.1981614233385778e-06,
      "loss": 0.2936,
      "num_input_tokens_seen": 63459856,
      "step": 33490
    },
    {
      "epoch": 5.464148788645077,
      "grad_norm": 0.06748447567224503,
      "learning_rate": 3.1933586782493115e-06,
      "loss": 0.0966,
      "num_input_tokens_seen": 63467824,
      "step": 33495
    },
    {
      "epoch": 5.464964515865894,
      "grad_norm": 1.2444918155670166,
      "learning_rate": 3.188559296018184e-06,
      "loss": 0.1567,
      "num_input_tokens_seen": 63477664,
      "step": 33500
    },
    {
      "epoch": 5.465780243086712,
      "grad_norm": 0.6945288181304932,
      "learning_rate": 3.1837632773853098e-06,
      "loss": 0.2035,
      "num_input_tokens_seen": 63486544,
      "step": 33505
    },
    {
      "epoch": 5.466595970307529,
      "grad_norm": 0.3503521680831909,
      "learning_rate": 3.178970623090294e-06,
      "loss": 0.172,
      "num_input_tokens_seen": 63495616,
      "step": 33510
    },
    {
      "epoch": 5.467411697528346,
      "grad_norm": 0.16461335122585297,
      "learning_rate": 3.174181333872234e-06,
      "loss": 0.1245,
      "num_input_tokens_seen": 63504608,
      "step": 33515
    },
    {
      "epoch": 5.468227424749164,
      "grad_norm": 1.1470146179199219,
      "learning_rate": 3.169395410469686e-06,
      "loss": 0.0774,
      "num_input_tokens_seen": 63514160,
      "step": 33520
    },
    {
      "epoch": 5.469043151969982,
      "grad_norm": 1.1267060041427612,
      "learning_rate": 3.164612853620713e-06,
      "loss": 0.1323,
      "num_input_tokens_seen": 63524272,
      "step": 33525
    },
    {
      "epoch": 5.469858879190799,
      "grad_norm": 0.6797387003898621,
      "learning_rate": 3.1598336640628333e-06,
      "loss": 0.0615,
      "num_input_tokens_seen": 63534048,
      "step": 33530
    },
    {
      "epoch": 5.470674606411616,
      "grad_norm": 0.1695612519979477,
      "learning_rate": 3.155057842533063e-06,
      "loss": 0.2078,
      "num_input_tokens_seen": 63543920,
      "step": 33535
    },
    {
      "epoch": 5.471490333632433,
      "grad_norm": 0.21972782909870148,
      "learning_rate": 3.1502853897678984e-06,
      "loss": 0.0511,
      "num_input_tokens_seen": 63553072,
      "step": 33540
    },
    {
      "epoch": 5.472306060853251,
      "grad_norm": 0.22530464828014374,
      "learning_rate": 3.1455163065033017e-06,
      "loss": 0.0226,
      "num_input_tokens_seen": 63561648,
      "step": 33545
    },
    {
      "epoch": 5.473121788074068,
      "grad_norm": 0.16763219237327576,
      "learning_rate": 3.140750593474734e-06,
      "loss": 0.0505,
      "num_input_tokens_seen": 63570320,
      "step": 33550
    },
    {
      "epoch": 5.473937515294885,
      "grad_norm": 1.2897942066192627,
      "learning_rate": 3.1359882514171294e-06,
      "loss": 0.1655,
      "num_input_tokens_seen": 63580576,
      "step": 33555
    },
    {
      "epoch": 5.474753242515702,
      "grad_norm": 1.2986773252487183,
      "learning_rate": 3.1312292810648903e-06,
      "loss": 0.208,
      "num_input_tokens_seen": 63591232,
      "step": 33560
    },
    {
      "epoch": 5.47556896973652,
      "grad_norm": 0.03159867599606514,
      "learning_rate": 3.1264736831519204e-06,
      "loss": 0.1026,
      "num_input_tokens_seen": 63601232,
      "step": 33565
    },
    {
      "epoch": 5.4763846969573375,
      "grad_norm": 0.6812056303024292,
      "learning_rate": 3.1217214584115863e-06,
      "loss": 0.0907,
      "num_input_tokens_seen": 63611264,
      "step": 33570
    },
    {
      "epoch": 5.477200424178155,
      "grad_norm": 0.6480364203453064,
      "learning_rate": 3.116972607576746e-06,
      "loss": 0.0712,
      "num_input_tokens_seen": 63620976,
      "step": 33575
    },
    {
      "epoch": 5.478016151398972,
      "grad_norm": 2.3244879245758057,
      "learning_rate": 3.1122271313797303e-06,
      "loss": 0.1858,
      "num_input_tokens_seen": 63630880,
      "step": 33580
    },
    {
      "epoch": 5.47883187861979,
      "grad_norm": 1.3678282499313354,
      "learning_rate": 3.107485030552343e-06,
      "loss": 0.3391,
      "num_input_tokens_seen": 63640704,
      "step": 33585
    },
    {
      "epoch": 5.479647605840607,
      "grad_norm": 1.45662260055542,
      "learning_rate": 3.1027463058258848e-06,
      "loss": 0.2018,
      "num_input_tokens_seen": 63649504,
      "step": 33590
    },
    {
      "epoch": 5.480463333061424,
      "grad_norm": 0.043817583471536636,
      "learning_rate": 3.0980109579311273e-06,
      "loss": 0.0731,
      "num_input_tokens_seen": 63658448,
      "step": 33595
    },
    {
      "epoch": 5.481279060282241,
      "grad_norm": 1.2482022047042847,
      "learning_rate": 3.093278987598314e-06,
      "loss": 0.3059,
      "num_input_tokens_seen": 63668416,
      "step": 33600
    },
    {
      "epoch": 5.481279060282241,
      "eval_loss": 0.14177285134792328,
      "eval_runtime": 133.2694,
      "eval_samples_per_second": 20.447,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 63668416,
      "step": 33600
    },
    {
      "epoch": 5.482094787503059,
      "grad_norm": 0.774105966091156,
      "learning_rate": 3.0885503955571826e-06,
      "loss": 0.1333,
      "num_input_tokens_seen": 63678304,
      "step": 33605
    },
    {
      "epoch": 5.482910514723876,
      "grad_norm": 0.09038911759853363,
      "learning_rate": 3.0838251825369313e-06,
      "loss": 0.2127,
      "num_input_tokens_seen": 63687888,
      "step": 33610
    },
    {
      "epoch": 5.4837262419446935,
      "grad_norm": 0.27099427580833435,
      "learning_rate": 3.0791033492662517e-06,
      "loss": 0.1163,
      "num_input_tokens_seen": 63697872,
      "step": 33615
    },
    {
      "epoch": 5.484541969165511,
      "grad_norm": 0.14205190539360046,
      "learning_rate": 3.0743848964733203e-06,
      "loss": 0.1448,
      "num_input_tokens_seen": 63708304,
      "step": 33620
    },
    {
      "epoch": 5.485357696386329,
      "grad_norm": 0.5964170098304749,
      "learning_rate": 3.0696698248857625e-06,
      "loss": 0.099,
      "num_input_tokens_seen": 63718048,
      "step": 33625
    },
    {
      "epoch": 5.486173423607146,
      "grad_norm": 0.9423452019691467,
      "learning_rate": 3.0649581352307192e-06,
      "loss": 0.1325,
      "num_input_tokens_seen": 63727872,
      "step": 33630
    },
    {
      "epoch": 5.486989150827963,
      "grad_norm": 0.0787886455655098,
      "learning_rate": 3.060249828234776e-06,
      "loss": 0.1267,
      "num_input_tokens_seen": 63737664,
      "step": 33635
    },
    {
      "epoch": 5.487804878048781,
      "grad_norm": 0.11749038100242615,
      "learning_rate": 3.055544904624025e-06,
      "loss": 0.0443,
      "num_input_tokens_seen": 63747504,
      "step": 33640
    },
    {
      "epoch": 5.488620605269598,
      "grad_norm": 0.68630450963974,
      "learning_rate": 3.050843365124026e-06,
      "loss": 0.0682,
      "num_input_tokens_seen": 63757408,
      "step": 33645
    },
    {
      "epoch": 5.489436332490415,
      "grad_norm": 1.2757368087768555,
      "learning_rate": 3.0461452104598083e-06,
      "loss": 0.2472,
      "num_input_tokens_seen": 63766720,
      "step": 33650
    },
    {
      "epoch": 5.490252059711232,
      "grad_norm": 0.32887834310531616,
      "learning_rate": 3.0414504413558836e-06,
      "loss": 0.295,
      "num_input_tokens_seen": 63775824,
      "step": 33655
    },
    {
      "epoch": 5.49106778693205,
      "grad_norm": 1.6007722616195679,
      "learning_rate": 3.0367590585362564e-06,
      "loss": 0.3221,
      "num_input_tokens_seen": 63784464,
      "step": 33660
    },
    {
      "epoch": 5.4918835141528675,
      "grad_norm": 0.15191997587680817,
      "learning_rate": 3.0320710627243813e-06,
      "loss": 0.1556,
      "num_input_tokens_seen": 63794496,
      "step": 33665
    },
    {
      "epoch": 5.492699241373685,
      "grad_norm": 1.2680110931396484,
      "learning_rate": 3.027386454643222e-06,
      "loss": 0.1606,
      "num_input_tokens_seen": 63804480,
      "step": 33670
    },
    {
      "epoch": 5.493514968594502,
      "grad_norm": 0.05587727576494217,
      "learning_rate": 3.0227052350151914e-06,
      "loss": 0.0433,
      "num_input_tokens_seen": 63813792,
      "step": 33675
    },
    {
      "epoch": 5.49433069581532,
      "grad_norm": 0.7212698459625244,
      "learning_rate": 3.0180274045621957e-06,
      "loss": 0.2392,
      "num_input_tokens_seen": 63823440,
      "step": 33680
    },
    {
      "epoch": 5.495146423036137,
      "grad_norm": 1.2610994577407837,
      "learning_rate": 3.013352964005625e-06,
      "loss": 0.1426,
      "num_input_tokens_seen": 63832160,
      "step": 33685
    },
    {
      "epoch": 5.495962150256954,
      "grad_norm": 0.11243008822202682,
      "learning_rate": 3.0086819140663218e-06,
      "loss": 0.0179,
      "num_input_tokens_seen": 63841104,
      "step": 33690
    },
    {
      "epoch": 5.496777877477771,
      "grad_norm": 1.3000251054763794,
      "learning_rate": 3.0040142554646265e-06,
      "loss": 0.1532,
      "num_input_tokens_seen": 63850192,
      "step": 33695
    },
    {
      "epoch": 5.497593604698589,
      "grad_norm": 1.0856751203536987,
      "learning_rate": 2.999349988920361e-06,
      "loss": 0.1032,
      "num_input_tokens_seen": 63859968,
      "step": 33700
    },
    {
      "epoch": 5.498409331919406,
      "grad_norm": 1.2355096340179443,
      "learning_rate": 2.994689115152796e-06,
      "loss": 0.1746,
      "num_input_tokens_seen": 63869088,
      "step": 33705
    },
    {
      "epoch": 5.499225059140223,
      "grad_norm": 0.6530975699424744,
      "learning_rate": 2.9900316348807105e-06,
      "loss": 0.0958,
      "num_input_tokens_seen": 63878432,
      "step": 33710
    },
    {
      "epoch": 5.5000407863610405,
      "grad_norm": 2.1003646850585938,
      "learning_rate": 2.985377548822338e-06,
      "loss": 0.2105,
      "num_input_tokens_seen": 63887792,
      "step": 33715
    },
    {
      "epoch": 5.500856513581859,
      "grad_norm": 0.5184425115585327,
      "learning_rate": 2.980726857695404e-06,
      "loss": 0.075,
      "num_input_tokens_seen": 63897456,
      "step": 33720
    },
    {
      "epoch": 5.501672240802676,
      "grad_norm": 0.5160555839538574,
      "learning_rate": 2.9760795622171017e-06,
      "loss": 0.0368,
      "num_input_tokens_seen": 63905184,
      "step": 33725
    },
    {
      "epoch": 5.502487968023493,
      "grad_norm": 0.025533705949783325,
      "learning_rate": 2.971435663104094e-06,
      "loss": 0.0832,
      "num_input_tokens_seen": 63913040,
      "step": 33730
    },
    {
      "epoch": 5.50330369524431,
      "grad_norm": 0.4490862190723419,
      "learning_rate": 2.9667951610725385e-06,
      "loss": 0.0652,
      "num_input_tokens_seen": 63921664,
      "step": 33735
    },
    {
      "epoch": 5.504119422465128,
      "grad_norm": 1.1446119546890259,
      "learning_rate": 2.9621580568380575e-06,
      "loss": 0.1432,
      "num_input_tokens_seen": 63930848,
      "step": 33740
    },
    {
      "epoch": 5.504935149685945,
      "grad_norm": 0.7966005802154541,
      "learning_rate": 2.9575243511157453e-06,
      "loss": 0.1189,
      "num_input_tokens_seen": 63940672,
      "step": 33745
    },
    {
      "epoch": 5.505750876906762,
      "grad_norm": 0.11235613375902176,
      "learning_rate": 2.952894044620186e-06,
      "loss": 0.0542,
      "num_input_tokens_seen": 63950704,
      "step": 33750
    },
    {
      "epoch": 5.506566604127579,
      "grad_norm": 0.29032662510871887,
      "learning_rate": 2.948267138065419e-06,
      "loss": 0.0919,
      "num_input_tokens_seen": 63960032,
      "step": 33755
    },
    {
      "epoch": 5.507382331348397,
      "grad_norm": 0.26604998111724854,
      "learning_rate": 2.943643632164983e-06,
      "loss": 0.1607,
      "num_input_tokens_seen": 63968560,
      "step": 33760
    },
    {
      "epoch": 5.5081980585692145,
      "grad_norm": 1.6346601247787476,
      "learning_rate": 2.939023527631879e-06,
      "loss": 0.3648,
      "num_input_tokens_seen": 63978416,
      "step": 33765
    },
    {
      "epoch": 5.509013785790032,
      "grad_norm": 0.7350406646728516,
      "learning_rate": 2.934406825178576e-06,
      "loss": 0.0654,
      "num_input_tokens_seen": 63988528,
      "step": 33770
    },
    {
      "epoch": 5.50982951301085,
      "grad_norm": 0.9773479700088501,
      "learning_rate": 2.9297935255170357e-06,
      "loss": 0.0735,
      "num_input_tokens_seen": 63997248,
      "step": 33775
    },
    {
      "epoch": 5.510645240231667,
      "grad_norm": 0.057598453015089035,
      "learning_rate": 2.925183629358691e-06,
      "loss": 0.1158,
      "num_input_tokens_seen": 64007008,
      "step": 33780
    },
    {
      "epoch": 5.511460967452484,
      "grad_norm": 0.1058405190706253,
      "learning_rate": 2.9205771374144346e-06,
      "loss": 0.1523,
      "num_input_tokens_seen": 64015632,
      "step": 33785
    },
    {
      "epoch": 5.512276694673301,
      "grad_norm": 0.20223161578178406,
      "learning_rate": 2.915974050394657e-06,
      "loss": 0.0761,
      "num_input_tokens_seen": 64024848,
      "step": 33790
    },
    {
      "epoch": 5.513092421894118,
      "grad_norm": 0.5895995497703552,
      "learning_rate": 2.9113743690092067e-06,
      "loss": 0.051,
      "num_input_tokens_seen": 64034112,
      "step": 33795
    },
    {
      "epoch": 5.513908149114936,
      "grad_norm": 0.07325702905654907,
      "learning_rate": 2.906778093967402e-06,
      "loss": 0.0325,
      "num_input_tokens_seen": 64043792,
      "step": 33800
    },
    {
      "epoch": 5.513908149114936,
      "eval_loss": 0.14099855720996857,
      "eval_runtime": 133.3009,
      "eval_samples_per_second": 20.442,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 64043792,
      "step": 33800
    },
    {
      "epoch": 5.514723876335753,
      "grad_norm": 0.08776328712701797,
      "learning_rate": 2.9021852259780656e-06,
      "loss": 0.0252,
      "num_input_tokens_seen": 64053968,
      "step": 33805
    },
    {
      "epoch": 5.5155396035565705,
      "grad_norm": 0.9111771583557129,
      "learning_rate": 2.8975957657494583e-06,
      "loss": 0.068,
      "num_input_tokens_seen": 64063456,
      "step": 33810
    },
    {
      "epoch": 5.5163553307773885,
      "grad_norm": 0.16579169034957886,
      "learning_rate": 2.8930097139893417e-06,
      "loss": 0.2501,
      "num_input_tokens_seen": 64074016,
      "step": 33815
    },
    {
      "epoch": 5.517171057998206,
      "grad_norm": 0.7944153547286987,
      "learning_rate": 2.888427071404945e-06,
      "loss": 0.0518,
      "num_input_tokens_seen": 64083952,
      "step": 33820
    },
    {
      "epoch": 5.517986785219023,
      "grad_norm": 0.4468885660171509,
      "learning_rate": 2.8838478387029606e-06,
      "loss": 0.1732,
      "num_input_tokens_seen": 64094656,
      "step": 33825
    },
    {
      "epoch": 5.51880251243984,
      "grad_norm": 0.44511452317237854,
      "learning_rate": 2.8792720165895737e-06,
      "loss": 0.2296,
      "num_input_tokens_seen": 64105200,
      "step": 33830
    },
    {
      "epoch": 5.519618239660657,
      "grad_norm": 0.8633498549461365,
      "learning_rate": 2.874699605770423e-06,
      "loss": 0.1689,
      "num_input_tokens_seen": 64114624,
      "step": 33835
    },
    {
      "epoch": 5.520433966881475,
      "grad_norm": 0.15051770210266113,
      "learning_rate": 2.8701306069506383e-06,
      "loss": 0.316,
      "num_input_tokens_seen": 64125120,
      "step": 33840
    },
    {
      "epoch": 5.521249694102292,
      "grad_norm": 1.1038507223129272,
      "learning_rate": 2.8655650208348178e-06,
      "loss": 0.1301,
      "num_input_tokens_seen": 64135168,
      "step": 33845
    },
    {
      "epoch": 5.522065421323109,
      "grad_norm": 0.11596563458442688,
      "learning_rate": 2.8610028481270257e-06,
      "loss": 0.1729,
      "num_input_tokens_seen": 64145312,
      "step": 33850
    },
    {
      "epoch": 5.522881148543927,
      "grad_norm": 0.07197333127260208,
      "learning_rate": 2.856444089530813e-06,
      "loss": 0.101,
      "num_input_tokens_seen": 64155376,
      "step": 33855
    },
    {
      "epoch": 5.523696875764744,
      "grad_norm": 1.2209227085113525,
      "learning_rate": 2.8518887457491955e-06,
      "loss": 0.1703,
      "num_input_tokens_seen": 64165232,
      "step": 33860
    },
    {
      "epoch": 5.524512602985562,
      "grad_norm": 0.383592814207077,
      "learning_rate": 2.8473368174846666e-06,
      "loss": 0.1764,
      "num_input_tokens_seen": 64174096,
      "step": 33865
    },
    {
      "epoch": 5.525328330206379,
      "grad_norm": 0.6579084396362305,
      "learning_rate": 2.842788305439184e-06,
      "loss": 0.1569,
      "num_input_tokens_seen": 64184528,
      "step": 33870
    },
    {
      "epoch": 5.526144057427197,
      "grad_norm": 0.1619454175233841,
      "learning_rate": 2.8382432103141925e-06,
      "loss": 0.1566,
      "num_input_tokens_seen": 64193776,
      "step": 33875
    },
    {
      "epoch": 5.526959784648014,
      "grad_norm": 0.6737205386161804,
      "learning_rate": 2.833701532810598e-06,
      "loss": 0.0947,
      "num_input_tokens_seen": 64203456,
      "step": 33880
    },
    {
      "epoch": 5.527775511868831,
      "grad_norm": 0.6712958216667175,
      "learning_rate": 2.8291632736287877e-06,
      "loss": 0.0552,
      "num_input_tokens_seen": 64214240,
      "step": 33885
    },
    {
      "epoch": 5.528591239089648,
      "grad_norm": 0.1826777160167694,
      "learning_rate": 2.824628433468615e-06,
      "loss": 0.0547,
      "num_input_tokens_seen": 64224080,
      "step": 33890
    },
    {
      "epoch": 5.529406966310466,
      "grad_norm": 0.21332557499408722,
      "learning_rate": 2.8200970130294073e-06,
      "loss": 0.1938,
      "num_input_tokens_seen": 64232832,
      "step": 33895
    },
    {
      "epoch": 5.530222693531283,
      "grad_norm": 0.2406429648399353,
      "learning_rate": 2.8155690130099775e-06,
      "loss": 0.172,
      "num_input_tokens_seen": 64241216,
      "step": 33900
    },
    {
      "epoch": 5.5310384207521,
      "grad_norm": 1.0798900127410889,
      "learning_rate": 2.8110444341085895e-06,
      "loss": 0.2571,
      "num_input_tokens_seen": 64251808,
      "step": 33905
    },
    {
      "epoch": 5.5318541479729175,
      "grad_norm": 0.05667674541473389,
      "learning_rate": 2.806523277022996e-06,
      "loss": 0.0735,
      "num_input_tokens_seen": 64260880,
      "step": 33910
    },
    {
      "epoch": 5.5326698751937355,
      "grad_norm": 1.3197145462036133,
      "learning_rate": 2.802005542450409e-06,
      "loss": 0.134,
      "num_input_tokens_seen": 64269248,
      "step": 33915
    },
    {
      "epoch": 5.533485602414553,
      "grad_norm": 0.894358217716217,
      "learning_rate": 2.797491231087526e-06,
      "loss": 0.1218,
      "num_input_tokens_seen": 64277952,
      "step": 33920
    },
    {
      "epoch": 5.53430132963537,
      "grad_norm": 1.76656973361969,
      "learning_rate": 2.7929803436305137e-06,
      "loss": 0.1544,
      "num_input_tokens_seen": 64287824,
      "step": 33925
    },
    {
      "epoch": 5.535117056856187,
      "grad_norm": 1.5858256816864014,
      "learning_rate": 2.788472880774998e-06,
      "loss": 0.1832,
      "num_input_tokens_seen": 64296624,
      "step": 33930
    },
    {
      "epoch": 5.535932784077005,
      "grad_norm": 0.7380752563476562,
      "learning_rate": 2.7839688432160977e-06,
      "loss": 0.0543,
      "num_input_tokens_seen": 64306432,
      "step": 33935
    },
    {
      "epoch": 5.536748511297822,
      "grad_norm": 0.9062879681587219,
      "learning_rate": 2.779468231648383e-06,
      "loss": 0.235,
      "num_input_tokens_seen": 64316832,
      "step": 33940
    },
    {
      "epoch": 5.537564238518639,
      "grad_norm": 0.4403841197490692,
      "learning_rate": 2.774971046765906e-06,
      "loss": 0.2364,
      "num_input_tokens_seen": 64326400,
      "step": 33945
    },
    {
      "epoch": 5.538379965739456,
      "grad_norm": 0.5061506628990173,
      "learning_rate": 2.770477289262194e-06,
      "loss": 0.2342,
      "num_input_tokens_seen": 64337280,
      "step": 33950
    },
    {
      "epoch": 5.539195692960274,
      "grad_norm": 1.3609423637390137,
      "learning_rate": 2.765986959830233e-06,
      "loss": 0.1493,
      "num_input_tokens_seen": 64345312,
      "step": 33955
    },
    {
      "epoch": 5.5400114201810915,
      "grad_norm": 0.8210045695304871,
      "learning_rate": 2.761500059162492e-06,
      "loss": 0.268,
      "num_input_tokens_seen": 64353376,
      "step": 33960
    },
    {
      "epoch": 5.540827147401909,
      "grad_norm": 0.3201478123664856,
      "learning_rate": 2.757016587950914e-06,
      "loss": 0.0982,
      "num_input_tokens_seen": 64363760,
      "step": 33965
    },
    {
      "epoch": 5.541642874622726,
      "grad_norm": 0.17385652661323547,
      "learning_rate": 2.752536546886897e-06,
      "loss": 0.0983,
      "num_input_tokens_seen": 64373376,
      "step": 33970
    },
    {
      "epoch": 5.542458601843544,
      "grad_norm": 0.07841465622186661,
      "learning_rate": 2.7480599366613234e-06,
      "loss": 0.2253,
      "num_input_tokens_seen": 64383248,
      "step": 33975
    },
    {
      "epoch": 5.543274329064361,
      "grad_norm": 0.49234944581985474,
      "learning_rate": 2.7435867579645473e-06,
      "loss": 0.08,
      "num_input_tokens_seen": 64394144,
      "step": 33980
    },
    {
      "epoch": 5.544090056285178,
      "grad_norm": 0.20509013533592224,
      "learning_rate": 2.739117011486378e-06,
      "loss": 0.0719,
      "num_input_tokens_seen": 64403568,
      "step": 33985
    },
    {
      "epoch": 5.544905783505996,
      "grad_norm": 1.4945276975631714,
      "learning_rate": 2.7346506979161216e-06,
      "loss": 0.2951,
      "num_input_tokens_seen": 64414384,
      "step": 33990
    },
    {
      "epoch": 5.545721510726813,
      "grad_norm": 0.13959655165672302,
      "learning_rate": 2.7301878179425227e-06,
      "loss": 0.0966,
      "num_input_tokens_seen": 64424544,
      "step": 33995
    },
    {
      "epoch": 5.54653723794763,
      "grad_norm": 0.22822581231594086,
      "learning_rate": 2.7257283722538244e-06,
      "loss": 0.0339,
      "num_input_tokens_seen": 64433840,
      "step": 34000
    },
    {
      "epoch": 5.54653723794763,
      "eval_loss": 0.14089542627334595,
      "eval_runtime": 133.2663,
      "eval_samples_per_second": 20.448,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 64433840,
      "step": 34000
    },
    {
      "epoch": 5.5473529651684474,
      "grad_norm": 0.9108707308769226,
      "learning_rate": 2.7212723615377326e-06,
      "loss": 0.1502,
      "num_input_tokens_seen": 64443344,
      "step": 34005
    },
    {
      "epoch": 5.548168692389265,
      "grad_norm": 1.892119288444519,
      "learning_rate": 2.7168197864814145e-06,
      "loss": 0.1913,
      "num_input_tokens_seen": 64453168,
      "step": 34010
    },
    {
      "epoch": 5.548984419610083,
      "grad_norm": 1.572105050086975,
      "learning_rate": 2.712370647771509e-06,
      "loss": 0.0854,
      "num_input_tokens_seen": 64462592,
      "step": 34015
    },
    {
      "epoch": 5.5498001468309,
      "grad_norm": 1.1200248003005981,
      "learning_rate": 2.707924946094137e-06,
      "loss": 0.1091,
      "num_input_tokens_seen": 64471584,
      "step": 34020
    },
    {
      "epoch": 5.550615874051717,
      "grad_norm": 1.2846753597259521,
      "learning_rate": 2.7034826821348723e-06,
      "loss": 0.1899,
      "num_input_tokens_seen": 64480016,
      "step": 34025
    },
    {
      "epoch": 5.551431601272535,
      "grad_norm": 0.15526260435581207,
      "learning_rate": 2.6990438565787786e-06,
      "loss": 0.0444,
      "num_input_tokens_seen": 64488224,
      "step": 34030
    },
    {
      "epoch": 5.552247328493352,
      "grad_norm": 1.5086886882781982,
      "learning_rate": 2.6946084701103714e-06,
      "loss": 0.1481,
      "num_input_tokens_seen": 64497792,
      "step": 34035
    },
    {
      "epoch": 5.553063055714169,
      "grad_norm": 0.5753306746482849,
      "learning_rate": 2.6901765234136428e-06,
      "loss": 0.1235,
      "num_input_tokens_seen": 64507136,
      "step": 34040
    },
    {
      "epoch": 5.553878782934986,
      "grad_norm": 0.15248313546180725,
      "learning_rate": 2.685748017172063e-06,
      "loss": 0.0317,
      "num_input_tokens_seen": 64516496,
      "step": 34045
    },
    {
      "epoch": 5.554694510155803,
      "grad_norm": 0.9758445620536804,
      "learning_rate": 2.681322952068549e-06,
      "loss": 0.056,
      "num_input_tokens_seen": 64525504,
      "step": 34050
    },
    {
      "epoch": 5.555510237376621,
      "grad_norm": 0.34197917580604553,
      "learning_rate": 2.6769013287855137e-06,
      "loss": 0.1542,
      "num_input_tokens_seen": 64534320,
      "step": 34055
    },
    {
      "epoch": 5.5563259645974385,
      "grad_norm": 0.8384179472923279,
      "learning_rate": 2.6724831480048286e-06,
      "loss": 0.1258,
      "num_input_tokens_seen": 64544384,
      "step": 34060
    },
    {
      "epoch": 5.557141691818256,
      "grad_norm": 0.16579635441303253,
      "learning_rate": 2.66806841040782e-06,
      "loss": 0.1539,
      "num_input_tokens_seen": 64555168,
      "step": 34065
    },
    {
      "epoch": 5.557957419039074,
      "grad_norm": 0.7188186049461365,
      "learning_rate": 2.6636571166753083e-06,
      "loss": 0.1723,
      "num_input_tokens_seen": 64565600,
      "step": 34070
    },
    {
      "epoch": 5.558773146259891,
      "grad_norm": 0.475697785615921,
      "learning_rate": 2.6592492674875598e-06,
      "loss": 0.1335,
      "num_input_tokens_seen": 64574416,
      "step": 34075
    },
    {
      "epoch": 5.559588873480708,
      "grad_norm": 0.3449759781360626,
      "learning_rate": 2.6548448635243305e-06,
      "loss": 0.0881,
      "num_input_tokens_seen": 64583920,
      "step": 34080
    },
    {
      "epoch": 5.560404600701525,
      "grad_norm": 0.5977740287780762,
      "learning_rate": 2.650443905464828e-06,
      "loss": 0.0661,
      "num_input_tokens_seen": 64593616,
      "step": 34085
    },
    {
      "epoch": 5.561220327922343,
      "grad_norm": 0.9470469951629639,
      "learning_rate": 2.646046393987739e-06,
      "loss": 0.2292,
      "num_input_tokens_seen": 64602768,
      "step": 34090
    },
    {
      "epoch": 5.56203605514316,
      "grad_norm": 0.30954039096832275,
      "learning_rate": 2.64165232977121e-06,
      "loss": 0.0599,
      "num_input_tokens_seen": 64612384,
      "step": 34095
    },
    {
      "epoch": 5.562851782363977,
      "grad_norm": 0.10661760717630386,
      "learning_rate": 2.6372617134928695e-06,
      "loss": 0.048,
      "num_input_tokens_seen": 64622384,
      "step": 34100
    },
    {
      "epoch": 5.5636675095847945,
      "grad_norm": 0.901894748210907,
      "learning_rate": 2.6328745458297943e-06,
      "loss": 0.1143,
      "num_input_tokens_seen": 64630352,
      "step": 34105
    },
    {
      "epoch": 5.5644832368056125,
      "grad_norm": 1.4930132627487183,
      "learning_rate": 2.6284908274585546e-06,
      "loss": 0.1226,
      "num_input_tokens_seen": 64638928,
      "step": 34110
    },
    {
      "epoch": 5.56529896402643,
      "grad_norm": 0.17689348757266998,
      "learning_rate": 2.6241105590551595e-06,
      "loss": 0.2609,
      "num_input_tokens_seen": 64647104,
      "step": 34115
    },
    {
      "epoch": 5.566114691247247,
      "grad_norm": 2.5921807289123535,
      "learning_rate": 2.6197337412951105e-06,
      "loss": 0.1587,
      "num_input_tokens_seen": 64657344,
      "step": 34120
    },
    {
      "epoch": 5.566930418468064,
      "grad_norm": 0.15272220969200134,
      "learning_rate": 2.6153603748533705e-06,
      "loss": 0.0938,
      "num_input_tokens_seen": 64666768,
      "step": 34125
    },
    {
      "epoch": 5.567746145688882,
      "grad_norm": 1.2559372186660767,
      "learning_rate": 2.6109904604043585e-06,
      "loss": 0.104,
      "num_input_tokens_seen": 64675728,
      "step": 34130
    },
    {
      "epoch": 5.568561872909699,
      "grad_norm": 0.3989292085170746,
      "learning_rate": 2.6066239986219765e-06,
      "loss": 0.0706,
      "num_input_tokens_seen": 64685024,
      "step": 34135
    },
    {
      "epoch": 5.569377600130516,
      "grad_norm": 0.36044174432754517,
      "learning_rate": 2.602260990179592e-06,
      "loss": 0.2185,
      "num_input_tokens_seen": 64694720,
      "step": 34140
    },
    {
      "epoch": 5.570193327351333,
      "grad_norm": 0.5410049557685852,
      "learning_rate": 2.5979014357500248e-06,
      "loss": 0.1344,
      "num_input_tokens_seen": 64704496,
      "step": 34145
    },
    {
      "epoch": 5.571009054572151,
      "grad_norm": 0.20151551067829132,
      "learning_rate": 2.5935453360055844e-06,
      "loss": 0.0703,
      "num_input_tokens_seen": 64714336,
      "step": 34150
    },
    {
      "epoch": 5.5718247817929685,
      "grad_norm": 0.34006941318511963,
      "learning_rate": 2.5891926916180283e-06,
      "loss": 0.2551,
      "num_input_tokens_seen": 64724144,
      "step": 34155
    },
    {
      "epoch": 5.572640509013786,
      "grad_norm": 0.6038566827774048,
      "learning_rate": 2.5848435032585883e-06,
      "loss": 0.1406,
      "num_input_tokens_seen": 64733360,
      "step": 34160
    },
    {
      "epoch": 5.573456236234604,
      "grad_norm": 1.3068808317184448,
      "learning_rate": 2.58049777159797e-06,
      "loss": 0.1874,
      "num_input_tokens_seen": 64742432,
      "step": 34165
    },
    {
      "epoch": 5.574271963455421,
      "grad_norm": 0.1023147702217102,
      "learning_rate": 2.576155497306332e-06,
      "loss": 0.0803,
      "num_input_tokens_seen": 64751696,
      "step": 34170
    },
    {
      "epoch": 5.575087690676238,
      "grad_norm": 1.5051466226577759,
      "learning_rate": 2.57181668105331e-06,
      "loss": 0.1746,
      "num_input_tokens_seen": 64761728,
      "step": 34175
    },
    {
      "epoch": 5.575903417897055,
      "grad_norm": 0.31447267532348633,
      "learning_rate": 2.567481323508014e-06,
      "loss": 0.1161,
      "num_input_tokens_seen": 64770368,
      "step": 34180
    },
    {
      "epoch": 5.576719145117872,
      "grad_norm": 1.4335052967071533,
      "learning_rate": 2.5631494253389954e-06,
      "loss": 0.1224,
      "num_input_tokens_seen": 64780144,
      "step": 34185
    },
    {
      "epoch": 5.57753487233869,
      "grad_norm": 0.7131007313728333,
      "learning_rate": 2.5588209872142997e-06,
      "loss": 0.1504,
      "num_input_tokens_seen": 64790288,
      "step": 34190
    },
    {
      "epoch": 5.578350599559507,
      "grad_norm": 0.7570000290870667,
      "learning_rate": 2.5544960098014186e-06,
      "loss": 0.0532,
      "num_input_tokens_seen": 64799584,
      "step": 34195
    },
    {
      "epoch": 5.579166326780324,
      "grad_norm": 0.3717283010482788,
      "learning_rate": 2.550174493767318e-06,
      "loss": 0.1333,
      "num_input_tokens_seen": 64808624,
      "step": 34200
    },
    {
      "epoch": 5.579166326780324,
      "eval_loss": 0.14131809771060944,
      "eval_runtime": 133.2403,
      "eval_samples_per_second": 20.452,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 64808624,
      "step": 34200
    },
    {
      "epoch": 5.5799820540011424,
      "grad_norm": 0.8342035412788391,
      "learning_rate": 2.545856439778438e-06,
      "loss": 0.0601,
      "num_input_tokens_seen": 64817808,
      "step": 34205
    },
    {
      "epoch": 5.58079778122196,
      "grad_norm": 0.3151121735572815,
      "learning_rate": 2.541541848500667e-06,
      "loss": 0.0507,
      "num_input_tokens_seen": 64827728,
      "step": 34210
    },
    {
      "epoch": 5.581613508442777,
      "grad_norm": 0.8415140509605408,
      "learning_rate": 2.5372307205993733e-06,
      "loss": 0.0876,
      "num_input_tokens_seen": 64836496,
      "step": 34215
    },
    {
      "epoch": 5.582429235663594,
      "grad_norm": 0.2745361328125,
      "learning_rate": 2.5329230567393917e-06,
      "loss": 0.0948,
      "num_input_tokens_seen": 64845728,
      "step": 34220
    },
    {
      "epoch": 5.583244962884411,
      "grad_norm": 0.18501824140548706,
      "learning_rate": 2.5286188575850164e-06,
      "loss": 0.0194,
      "num_input_tokens_seen": 64855936,
      "step": 34225
    },
    {
      "epoch": 5.584060690105229,
      "grad_norm": 0.529229462146759,
      "learning_rate": 2.5243181237999984e-06,
      "loss": 0.1469,
      "num_input_tokens_seen": 64865424,
      "step": 34230
    },
    {
      "epoch": 5.584876417326046,
      "grad_norm": 0.2662545144557953,
      "learning_rate": 2.520020856047578e-06,
      "loss": 0.1699,
      "num_input_tokens_seen": 64874304,
      "step": 34235
    },
    {
      "epoch": 5.585692144546863,
      "grad_norm": 1.4224861860275269,
      "learning_rate": 2.515727054990438e-06,
      "loss": 0.2393,
      "num_input_tokens_seen": 64885232,
      "step": 34240
    },
    {
      "epoch": 5.586507871767681,
      "grad_norm": 1.2026920318603516,
      "learning_rate": 2.511436721290747e-06,
      "loss": 0.1379,
      "num_input_tokens_seen": 64895264,
      "step": 34245
    },
    {
      "epoch": 5.587323598988498,
      "grad_norm": 0.3279459476470947,
      "learning_rate": 2.5071498556101164e-06,
      "loss": 0.1781,
      "num_input_tokens_seen": 64904096,
      "step": 34250
    },
    {
      "epoch": 5.5881393262093155,
      "grad_norm": 0.2635965049266815,
      "learning_rate": 2.5028664586096485e-06,
      "loss": 0.0988,
      "num_input_tokens_seen": 64913216,
      "step": 34255
    },
    {
      "epoch": 5.588955053430133,
      "grad_norm": 0.10534791648387909,
      "learning_rate": 2.498586530949881e-06,
      "loss": 0.0176,
      "num_input_tokens_seen": 64923184,
      "step": 34260
    },
    {
      "epoch": 5.589770780650951,
      "grad_norm": 0.4364071786403656,
      "learning_rate": 2.4943100732908427e-06,
      "loss": 0.0455,
      "num_input_tokens_seen": 64930448,
      "step": 34265
    },
    {
      "epoch": 5.590586507871768,
      "grad_norm": 1.1117019653320312,
      "learning_rate": 2.4900370862920188e-06,
      "loss": 0.0675,
      "num_input_tokens_seen": 64940368,
      "step": 34270
    },
    {
      "epoch": 5.591402235092585,
      "grad_norm": 0.22458522021770477,
      "learning_rate": 2.4857675706123518e-06,
      "loss": 0.0585,
      "num_input_tokens_seen": 64949728,
      "step": 34275
    },
    {
      "epoch": 5.592217962313402,
      "grad_norm": 0.14336206018924713,
      "learning_rate": 2.4815015269102543e-06,
      "loss": 0.0591,
      "num_input_tokens_seen": 64958704,
      "step": 34280
    },
    {
      "epoch": 5.59303368953422,
      "grad_norm": 1.5904936790466309,
      "learning_rate": 2.477238955843611e-06,
      "loss": 0.1859,
      "num_input_tokens_seen": 64967568,
      "step": 34285
    },
    {
      "epoch": 5.593849416755037,
      "grad_norm": 0.3168010115623474,
      "learning_rate": 2.4729798580697573e-06,
      "loss": 0.2222,
      "num_input_tokens_seen": 64977760,
      "step": 34290
    },
    {
      "epoch": 5.594665143975854,
      "grad_norm": 0.3385723829269409,
      "learning_rate": 2.4687242342455034e-06,
      "loss": 0.0376,
      "num_input_tokens_seen": 64986800,
      "step": 34295
    },
    {
      "epoch": 5.5954808711966715,
      "grad_norm": 0.3086030185222626,
      "learning_rate": 2.4644720850271196e-06,
      "loss": 0.0966,
      "num_input_tokens_seen": 64996208,
      "step": 34300
    },
    {
      "epoch": 5.5962965984174895,
      "grad_norm": 0.6233262419700623,
      "learning_rate": 2.4602234110703364e-06,
      "loss": 0.1632,
      "num_input_tokens_seen": 65005600,
      "step": 34305
    },
    {
      "epoch": 5.597112325638307,
      "grad_norm": 0.16564229130744934,
      "learning_rate": 2.4559782130303576e-06,
      "loss": 0.0465,
      "num_input_tokens_seen": 65014688,
      "step": 34310
    },
    {
      "epoch": 5.597928052859124,
      "grad_norm": 1.122965931892395,
      "learning_rate": 2.451736491561843e-06,
      "loss": 0.1878,
      "num_input_tokens_seen": 65023504,
      "step": 34315
    },
    {
      "epoch": 5.598743780079941,
      "grad_norm": 1.3188679218292236,
      "learning_rate": 2.4474982473189163e-06,
      "loss": 0.1147,
      "num_input_tokens_seen": 65033936,
      "step": 34320
    },
    {
      "epoch": 5.599559507300759,
      "grad_norm": 0.808472752571106,
      "learning_rate": 2.4432634809551796e-06,
      "loss": 0.1182,
      "num_input_tokens_seen": 65042224,
      "step": 34325
    },
    {
      "epoch": 5.600375234521576,
      "grad_norm": 0.5742913484573364,
      "learning_rate": 2.439032193123675e-06,
      "loss": 0.1325,
      "num_input_tokens_seen": 65052368,
      "step": 34330
    },
    {
      "epoch": 5.601190961742393,
      "grad_norm": 0.38623863458633423,
      "learning_rate": 2.4348043844769297e-06,
      "loss": 0.2042,
      "num_input_tokens_seen": 65061952,
      "step": 34335
    },
    {
      "epoch": 5.602006688963211,
      "grad_norm": 0.22221703827381134,
      "learning_rate": 2.4305800556669146e-06,
      "loss": 0.084,
      "num_input_tokens_seen": 65070176,
      "step": 34340
    },
    {
      "epoch": 5.602822416184028,
      "grad_norm": 0.9906127452850342,
      "learning_rate": 2.426359207345083e-06,
      "loss": 0.0869,
      "num_input_tokens_seen": 65080048,
      "step": 34345
    },
    {
      "epoch": 5.6036381434048455,
      "grad_norm": 1.246800422668457,
      "learning_rate": 2.4221418401623396e-06,
      "loss": 0.0918,
      "num_input_tokens_seen": 65089056,
      "step": 34350
    },
    {
      "epoch": 5.604453870625663,
      "grad_norm": 0.37507396936416626,
      "learning_rate": 2.4179279547690557e-06,
      "loss": 0.0805,
      "num_input_tokens_seen": 65098176,
      "step": 34355
    },
    {
      "epoch": 5.60526959784648,
      "grad_norm": 0.28250932693481445,
      "learning_rate": 2.413717551815062e-06,
      "loss": 0.0524,
      "num_input_tokens_seen": 65107312,
      "step": 34360
    },
    {
      "epoch": 5.606085325067298,
      "grad_norm": 1.5150099992752075,
      "learning_rate": 2.409510631949666e-06,
      "loss": 0.1996,
      "num_input_tokens_seen": 65117088,
      "step": 34365
    },
    {
      "epoch": 5.606901052288115,
      "grad_norm": 0.9857009649276733,
      "learning_rate": 2.405307195821618e-06,
      "loss": 0.1649,
      "num_input_tokens_seen": 65126896,
      "step": 34370
    },
    {
      "epoch": 5.607716779508932,
      "grad_norm": 0.5957353711128235,
      "learning_rate": 2.4011072440791372e-06,
      "loss": 0.0626,
      "num_input_tokens_seen": 65135504,
      "step": 34375
    },
    {
      "epoch": 5.60853250672975,
      "grad_norm": 1.3779296875,
      "learning_rate": 2.3969107773699233e-06,
      "loss": 0.1913,
      "num_input_tokens_seen": 65144864,
      "step": 34380
    },
    {
      "epoch": 5.609348233950567,
      "grad_norm": 1.3829323053359985,
      "learning_rate": 2.3927177963411096e-06,
      "loss": 0.2591,
      "num_input_tokens_seen": 65154736,
      "step": 34385
    },
    {
      "epoch": 5.610163961171384,
      "grad_norm": 0.054936718195676804,
      "learning_rate": 2.3885283016393144e-06,
      "loss": 0.1607,
      "num_input_tokens_seen": 65164704,
      "step": 34390
    },
    {
      "epoch": 5.610979688392201,
      "grad_norm": 1.0113534927368164,
      "learning_rate": 2.3843422939106076e-06,
      "loss": 0.0846,
      "num_input_tokens_seen": 65173920,
      "step": 34395
    },
    {
      "epoch": 5.6117954156130185,
      "grad_norm": 0.061335016041994095,
      "learning_rate": 2.380159773800525e-06,
      "loss": 0.0653,
      "num_input_tokens_seen": 65182704,
      "step": 34400
    },
    {
      "epoch": 5.6117954156130185,
      "eval_loss": 0.1409878134727478,
      "eval_runtime": 133.3378,
      "eval_samples_per_second": 20.437,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 65182704,
      "step": 34400
    },
    {
      "epoch": 5.6126111428338366,
      "grad_norm": 0.7077380418777466,
      "learning_rate": 2.3759807419540675e-06,
      "loss": 0.1644,
      "num_input_tokens_seen": 65191888,
      "step": 34405
    },
    {
      "epoch": 5.613426870054654,
      "grad_norm": 0.2791897654533386,
      "learning_rate": 2.3718051990156835e-06,
      "loss": 0.0482,
      "num_input_tokens_seen": 65200176,
      "step": 34410
    },
    {
      "epoch": 5.614242597275471,
      "grad_norm": 3.225511074066162,
      "learning_rate": 2.367633145629311e-06,
      "loss": 0.2934,
      "num_input_tokens_seen": 65210464,
      "step": 34415
    },
    {
      "epoch": 5.615058324496289,
      "grad_norm": 0.441853404045105,
      "learning_rate": 2.363464582438316e-06,
      "loss": 0.075,
      "num_input_tokens_seen": 65219696,
      "step": 34420
    },
    {
      "epoch": 5.615874051717106,
      "grad_norm": 0.7394428849220276,
      "learning_rate": 2.3592995100855526e-06,
      "loss": 0.0475,
      "num_input_tokens_seen": 65229024,
      "step": 34425
    },
    {
      "epoch": 5.616689778937923,
      "grad_norm": 0.49946972727775574,
      "learning_rate": 2.3551379292133273e-06,
      "loss": 0.1361,
      "num_input_tokens_seen": 65238352,
      "step": 34430
    },
    {
      "epoch": 5.61750550615874,
      "grad_norm": 1.788073182106018,
      "learning_rate": 2.3509798404634047e-06,
      "loss": 0.134,
      "num_input_tokens_seen": 65248208,
      "step": 34435
    },
    {
      "epoch": 5.618321233379558,
      "grad_norm": 0.07617887109518051,
      "learning_rate": 2.346825244477019e-06,
      "loss": 0.0659,
      "num_input_tokens_seen": 65256944,
      "step": 34440
    },
    {
      "epoch": 5.619136960600375,
      "grad_norm": 0.38335439562797546,
      "learning_rate": 2.3426741418948545e-06,
      "loss": 0.0741,
      "num_input_tokens_seen": 65267456,
      "step": 34445
    },
    {
      "epoch": 5.6199526878211925,
      "grad_norm": 1.487789273262024,
      "learning_rate": 2.3385265333570715e-06,
      "loss": 0.1263,
      "num_input_tokens_seen": 65277232,
      "step": 34450
    },
    {
      "epoch": 5.62076841504201,
      "grad_norm": 0.44251149892807007,
      "learning_rate": 2.334382419503278e-06,
      "loss": 0.1062,
      "num_input_tokens_seen": 65286544,
      "step": 34455
    },
    {
      "epoch": 5.621584142262828,
      "grad_norm": 1.1392176151275635,
      "learning_rate": 2.3302418009725465e-06,
      "loss": 0.2382,
      "num_input_tokens_seen": 65295584,
      "step": 34460
    },
    {
      "epoch": 5.622399869483645,
      "grad_norm": 0.15797759592533112,
      "learning_rate": 2.326104678403415e-06,
      "loss": 0.3042,
      "num_input_tokens_seen": 65305968,
      "step": 34465
    },
    {
      "epoch": 5.623215596704462,
      "grad_norm": 0.5413333773612976,
      "learning_rate": 2.321971052433883e-06,
      "loss": 0.0781,
      "num_input_tokens_seen": 65315728,
      "step": 34470
    },
    {
      "epoch": 5.624031323925279,
      "grad_norm": 0.08829787373542786,
      "learning_rate": 2.3178409237014004e-06,
      "loss": 0.0338,
      "num_input_tokens_seen": 65325136,
      "step": 34475
    },
    {
      "epoch": 5.624847051146097,
      "grad_norm": 0.7490702867507935,
      "learning_rate": 2.313714292842889e-06,
      "loss": 0.2318,
      "num_input_tokens_seen": 65335968,
      "step": 34480
    },
    {
      "epoch": 5.625662778366914,
      "grad_norm": 0.5833306312561035,
      "learning_rate": 2.309591160494734e-06,
      "loss": 0.0892,
      "num_input_tokens_seen": 65345344,
      "step": 34485
    },
    {
      "epoch": 5.626478505587731,
      "grad_norm": 1.2115256786346436,
      "learning_rate": 2.305471527292763e-06,
      "loss": 0.1548,
      "num_input_tokens_seen": 65352832,
      "step": 34490
    },
    {
      "epoch": 5.6272942328085485,
      "grad_norm": 0.06578151881694794,
      "learning_rate": 2.3013553938722817e-06,
      "loss": 0.1179,
      "num_input_tokens_seen": 65362592,
      "step": 34495
    },
    {
      "epoch": 5.6281099600293665,
      "grad_norm": 0.5088229775428772,
      "learning_rate": 2.297242760868043e-06,
      "loss": 0.146,
      "num_input_tokens_seen": 65371952,
      "step": 34500
    },
    {
      "epoch": 5.628925687250184,
      "grad_norm": 1.6297647953033447,
      "learning_rate": 2.2931336289142735e-06,
      "loss": 0.0857,
      "num_input_tokens_seen": 65381584,
      "step": 34505
    },
    {
      "epoch": 5.629741414471001,
      "grad_norm": 0.17479455471038818,
      "learning_rate": 2.289027998644655e-06,
      "loss": 0.0678,
      "num_input_tokens_seen": 65389504,
      "step": 34510
    },
    {
      "epoch": 5.630557141691818,
      "grad_norm": 0.40569302439689636,
      "learning_rate": 2.2849258706923228e-06,
      "loss": 0.1673,
      "num_input_tokens_seen": 65398384,
      "step": 34515
    },
    {
      "epoch": 5.631372868912636,
      "grad_norm": 0.9858798384666443,
      "learning_rate": 2.2808272456898705e-06,
      "loss": 0.1249,
      "num_input_tokens_seen": 65407872,
      "step": 34520
    },
    {
      "epoch": 5.632188596133453,
      "grad_norm": 1.293871521949768,
      "learning_rate": 2.2767321242693707e-06,
      "loss": 0.1706,
      "num_input_tokens_seen": 65418096,
      "step": 34525
    },
    {
      "epoch": 5.63300432335427,
      "grad_norm": 0.10313685983419418,
      "learning_rate": 2.272640507062329e-06,
      "loss": 0.0637,
      "num_input_tokens_seen": 65426560,
      "step": 34530
    },
    {
      "epoch": 5.633820050575087,
      "grad_norm": 0.7249055504798889,
      "learning_rate": 2.2685523946997382e-06,
      "loss": 0.0906,
      "num_input_tokens_seen": 65437408,
      "step": 34535
    },
    {
      "epoch": 5.634635777795905,
      "grad_norm": 0.6803945899009705,
      "learning_rate": 2.2644677878120245e-06,
      "loss": 0.1661,
      "num_input_tokens_seen": 65447648,
      "step": 34540
    },
    {
      "epoch": 5.635451505016722,
      "grad_norm": 0.8067421317100525,
      "learning_rate": 2.2603866870290897e-06,
      "loss": 0.063,
      "num_input_tokens_seen": 65457584,
      "step": 34545
    },
    {
      "epoch": 5.63626723223754,
      "grad_norm": 0.10816114395856857,
      "learning_rate": 2.256309092980294e-06,
      "loss": 0.1105,
      "num_input_tokens_seen": 65468272,
      "step": 34550
    },
    {
      "epoch": 5.637082959458358,
      "grad_norm": 0.06541014462709427,
      "learning_rate": 2.252235006294448e-06,
      "loss": 0.3156,
      "num_input_tokens_seen": 65478000,
      "step": 34555
    },
    {
      "epoch": 5.637898686679175,
      "grad_norm": 0.1548587828874588,
      "learning_rate": 2.2481644275998333e-06,
      "loss": 0.1584,
      "num_input_tokens_seen": 65487216,
      "step": 34560
    },
    {
      "epoch": 5.638714413899992,
      "grad_norm": 1.4991934299468994,
      "learning_rate": 2.2440973575241832e-06,
      "loss": 0.0894,
      "num_input_tokens_seen": 65497088,
      "step": 34565
    },
    {
      "epoch": 5.639530141120809,
      "grad_norm": 1.8296678066253662,
      "learning_rate": 2.240033796694685e-06,
      "loss": 0.2852,
      "num_input_tokens_seen": 65507280,
      "step": 34570
    },
    {
      "epoch": 5.640345868341626,
      "grad_norm": 0.03767570108175278,
      "learning_rate": 2.235973745737999e-06,
      "loss": 0.1611,
      "num_input_tokens_seen": 65517008,
      "step": 34575
    },
    {
      "epoch": 5.641161595562444,
      "grad_norm": 1.2662397623062134,
      "learning_rate": 2.2319172052802263e-06,
      "loss": 0.0867,
      "num_input_tokens_seen": 65526816,
      "step": 34580
    },
    {
      "epoch": 5.641977322783261,
      "grad_norm": 0.10098744183778763,
      "learning_rate": 2.2278641759469477e-06,
      "loss": 0.0725,
      "num_input_tokens_seen": 65535520,
      "step": 34585
    },
    {
      "epoch": 5.642793050004078,
      "grad_norm": 0.8645139336585999,
      "learning_rate": 2.2238146583631825e-06,
      "loss": 0.1109,
      "num_input_tokens_seen": 65544096,
      "step": 34590
    },
    {
      "epoch": 5.643608777224896,
      "grad_norm": 0.4090847373008728,
      "learning_rate": 2.2197686531534256e-06,
      "loss": 0.0903,
      "num_input_tokens_seen": 65552736,
      "step": 34595
    },
    {
      "epoch": 5.6444245044457135,
      "grad_norm": 0.1682523787021637,
      "learning_rate": 2.2157261609416087e-06,
      "loss": 0.0782,
      "num_input_tokens_seen": 65562192,
      "step": 34600
    },
    {
      "epoch": 5.6444245044457135,
      "eval_loss": 0.14120827615261078,
      "eval_runtime": 133.3971,
      "eval_samples_per_second": 20.428,
      "eval_steps_per_second": 10.218,
      "num_input_tokens_seen": 65562192,
      "step": 34600
    },
    {
      "epoch": 5.645240231666531,
      "grad_norm": 0.6563504338264465,
      "learning_rate": 2.211687182351149e-06,
      "loss": 0.0685,
      "num_input_tokens_seen": 65572096,
      "step": 34605
    },
    {
      "epoch": 5.646055958887348,
      "grad_norm": 1.102212905883789,
      "learning_rate": 2.2076517180048993e-06,
      "loss": 0.1415,
      "num_input_tokens_seen": 65581888,
      "step": 34610
    },
    {
      "epoch": 5.646871686108166,
      "grad_norm": 0.9403263926506042,
      "learning_rate": 2.2036197685251834e-06,
      "loss": 0.3894,
      "num_input_tokens_seen": 65591232,
      "step": 34615
    },
    {
      "epoch": 5.647687413328983,
      "grad_norm": 0.14041569828987122,
      "learning_rate": 2.199591334533771e-06,
      "loss": 0.1529,
      "num_input_tokens_seen": 65600224,
      "step": 34620
    },
    {
      "epoch": 5.6485031405498,
      "grad_norm": 0.3731609582901001,
      "learning_rate": 2.1955664166519036e-06,
      "loss": 0.2125,
      "num_input_tokens_seen": 65610272,
      "step": 34625
    },
    {
      "epoch": 5.649318867770617,
      "grad_norm": 0.5596976280212402,
      "learning_rate": 2.1915450155002793e-06,
      "loss": 0.0702,
      "num_input_tokens_seen": 65620128,
      "step": 34630
    },
    {
      "epoch": 5.650134594991435,
      "grad_norm": 1.5569827556610107,
      "learning_rate": 2.187527131699038e-06,
      "loss": 0.2731,
      "num_input_tokens_seen": 65630416,
      "step": 34635
    },
    {
      "epoch": 5.650950322212252,
      "grad_norm": 0.10803090035915375,
      "learning_rate": 2.18351276586779e-06,
      "loss": 0.0906,
      "num_input_tokens_seen": 65639968,
      "step": 34640
    },
    {
      "epoch": 5.6517660494330695,
      "grad_norm": 0.6836075782775879,
      "learning_rate": 2.1795019186256092e-06,
      "loss": 0.0426,
      "num_input_tokens_seen": 65649440,
      "step": 34645
    },
    {
      "epoch": 5.652581776653887,
      "grad_norm": 0.1271536648273468,
      "learning_rate": 2.1754945905910094e-06,
      "loss": 0.0276,
      "num_input_tokens_seen": 65658944,
      "step": 34650
    },
    {
      "epoch": 5.653397503874705,
      "grad_norm": 0.03937457129359245,
      "learning_rate": 2.171490782381977e-06,
      "loss": 0.0906,
      "num_input_tokens_seen": 65668048,
      "step": 34655
    },
    {
      "epoch": 5.654213231095522,
      "grad_norm": 0.4509480893611908,
      "learning_rate": 2.1674904946159425e-06,
      "loss": 0.0549,
      "num_input_tokens_seen": 65676784,
      "step": 34660
    },
    {
      "epoch": 5.655028958316339,
      "grad_norm": 0.2760498523712158,
      "learning_rate": 2.16349372790981e-06,
      "loss": 0.0838,
      "num_input_tokens_seen": 65686480,
      "step": 34665
    },
    {
      "epoch": 5.655844685537156,
      "grad_norm": 0.23847739398479462,
      "learning_rate": 2.159500482879928e-06,
      "loss": 0.0591,
      "num_input_tokens_seen": 65695264,
      "step": 34670
    },
    {
      "epoch": 5.656660412757974,
      "grad_norm": 0.11453074961900711,
      "learning_rate": 2.155510760142096e-06,
      "loss": 0.1053,
      "num_input_tokens_seen": 65704400,
      "step": 34675
    },
    {
      "epoch": 5.657476139978791,
      "grad_norm": 0.07598204910755157,
      "learning_rate": 2.151524560311588e-06,
      "loss": 0.0988,
      "num_input_tokens_seen": 65713616,
      "step": 34680
    },
    {
      "epoch": 5.658291867199608,
      "grad_norm": 0.04091746732592583,
      "learning_rate": 2.147541884003129e-06,
      "loss": 0.0312,
      "num_input_tokens_seen": 65721776,
      "step": 34685
    },
    {
      "epoch": 5.659107594420425,
      "grad_norm": 0.3703174293041229,
      "learning_rate": 2.1435627318308895e-06,
      "loss": 0.0677,
      "num_input_tokens_seen": 65732160,
      "step": 34690
    },
    {
      "epoch": 5.6599233216412435,
      "grad_norm": 0.5565459728240967,
      "learning_rate": 2.139587104408511e-06,
      "loss": 0.0731,
      "num_input_tokens_seen": 65741776,
      "step": 34695
    },
    {
      "epoch": 5.660739048862061,
      "grad_norm": 0.09012982994318008,
      "learning_rate": 2.1356150023490783e-06,
      "loss": 0.1307,
      "num_input_tokens_seen": 65750640,
      "step": 34700
    },
    {
      "epoch": 5.661554776082878,
      "grad_norm": 1.946156620979309,
      "learning_rate": 2.1316464262651464e-06,
      "loss": 0.0819,
      "num_input_tokens_seen": 65761632,
      "step": 34705
    },
    {
      "epoch": 5.662370503303695,
      "grad_norm": 0.9633821845054626,
      "learning_rate": 2.1276813767687224e-06,
      "loss": 0.0604,
      "num_input_tokens_seen": 65771312,
      "step": 34710
    },
    {
      "epoch": 5.663186230524513,
      "grad_norm": 1.2501294612884521,
      "learning_rate": 2.123719854471254e-06,
      "loss": 0.1474,
      "num_input_tokens_seen": 65780880,
      "step": 34715
    },
    {
      "epoch": 5.66400195774533,
      "grad_norm": 0.12349727004766464,
      "learning_rate": 2.119761859983668e-06,
      "loss": 0.0557,
      "num_input_tokens_seen": 65790256,
      "step": 34720
    },
    {
      "epoch": 5.664817684966147,
      "grad_norm": 0.5071682929992676,
      "learning_rate": 2.1158073939163386e-06,
      "loss": 0.0724,
      "num_input_tokens_seen": 65799664,
      "step": 34725
    },
    {
      "epoch": 5.665633412186965,
      "grad_norm": 0.6038040518760681,
      "learning_rate": 2.111856456879088e-06,
      "loss": 0.1072,
      "num_input_tokens_seen": 65808624,
      "step": 34730
    },
    {
      "epoch": 5.666449139407782,
      "grad_norm": 0.515776515007019,
      "learning_rate": 2.1079090494811993e-06,
      "loss": 0.028,
      "num_input_tokens_seen": 65817760,
      "step": 34735
    },
    {
      "epoch": 5.667264866628599,
      "grad_norm": 0.27942559123039246,
      "learning_rate": 2.103965172331418e-06,
      "loss": 0.0515,
      "num_input_tokens_seen": 65826848,
      "step": 34740
    },
    {
      "epoch": 5.6680805938494165,
      "grad_norm": 0.5423218011856079,
      "learning_rate": 2.100024826037933e-06,
      "loss": 0.3514,
      "num_input_tokens_seen": 65835808,
      "step": 34745
    },
    {
      "epoch": 5.668896321070234,
      "grad_norm": 0.17866243422031403,
      "learning_rate": 2.0960880112084027e-06,
      "loss": 0.1518,
      "num_input_tokens_seen": 65845520,
      "step": 34750
    },
    {
      "epoch": 5.669712048291052,
      "grad_norm": 0.4500112533569336,
      "learning_rate": 2.092154728449927e-06,
      "loss": 0.0982,
      "num_input_tokens_seen": 65854624,
      "step": 34755
    },
    {
      "epoch": 5.670527775511869,
      "grad_norm": 1.6508378982543945,
      "learning_rate": 2.0882249783690687e-06,
      "loss": 0.124,
      "num_input_tokens_seen": 65863808,
      "step": 34760
    },
    {
      "epoch": 5.671343502732686,
      "grad_norm": 0.26559823751449585,
      "learning_rate": 2.084298761571851e-06,
      "loss": 0.3237,
      "num_input_tokens_seen": 65873680,
      "step": 34765
    },
    {
      "epoch": 5.672159229953504,
      "grad_norm": 0.10509172827005386,
      "learning_rate": 2.080376078663737e-06,
      "loss": 0.0992,
      "num_input_tokens_seen": 65883376,
      "step": 34770
    },
    {
      "epoch": 5.672974957174321,
      "grad_norm": 0.07189783453941345,
      "learning_rate": 2.0764569302496593e-06,
      "loss": 0.1027,
      "num_input_tokens_seen": 65894048,
      "step": 34775
    },
    {
      "epoch": 5.673790684395138,
      "grad_norm": 0.09661045670509338,
      "learning_rate": 2.0725413169339957e-06,
      "loss": 0.1249,
      "num_input_tokens_seen": 65904320,
      "step": 34780
    },
    {
      "epoch": 5.674606411615955,
      "grad_norm": 2.0057220458984375,
      "learning_rate": 2.068629239320588e-06,
      "loss": 0.264,
      "num_input_tokens_seen": 65913024,
      "step": 34785
    },
    {
      "epoch": 5.6754221388367725,
      "grad_norm": 1.3064799308776855,
      "learning_rate": 2.064720698012726e-06,
      "loss": 0.2699,
      "num_input_tokens_seen": 65922432,
      "step": 34790
    },
    {
      "epoch": 5.6762378660575905,
      "grad_norm": 0.7363913655281067,
      "learning_rate": 2.0608156936131522e-06,
      "loss": 0.1214,
      "num_input_tokens_seen": 65931744,
      "step": 34795
    },
    {
      "epoch": 5.677053593278408,
      "grad_norm": 0.03312453255057335,
      "learning_rate": 2.056914226724074e-06,
      "loss": 0.1815,
      "num_input_tokens_seen": 65940816,
      "step": 34800
    },
    {
      "epoch": 5.677053593278408,
      "eval_loss": 0.14196398854255676,
      "eval_runtime": 133.3435,
      "eval_samples_per_second": 20.436,
      "eval_steps_per_second": 10.222,
      "num_input_tokens_seen": 65940816,
      "step": 34800
    },
    {
      "epoch": 5.677869320499225,
      "grad_norm": 0.16627436876296997,
      "learning_rate": 2.0530162979471385e-06,
      "loss": 0.0851,
      "num_input_tokens_seen": 65950176,
      "step": 34805
    },
    {
      "epoch": 5.678685047720043,
      "grad_norm": 0.20599789917469025,
      "learning_rate": 2.0491219078834667e-06,
      "loss": 0.1022,
      "num_input_tokens_seen": 65962032,
      "step": 34810
    },
    {
      "epoch": 5.67950077494086,
      "grad_norm": 0.14348997175693512,
      "learning_rate": 2.045231057133612e-06,
      "loss": 0.0954,
      "num_input_tokens_seen": 65971648,
      "step": 34815
    },
    {
      "epoch": 5.680316502161677,
      "grad_norm": 1.0973645448684692,
      "learning_rate": 2.0413437462975944e-06,
      "loss": 0.0817,
      "num_input_tokens_seen": 65980400,
      "step": 34820
    },
    {
      "epoch": 5.681132229382494,
      "grad_norm": 0.4576854705810547,
      "learning_rate": 2.0374599759748843e-06,
      "loss": 0.0832,
      "num_input_tokens_seen": 65989872,
      "step": 34825
    },
    {
      "epoch": 5.681947956603312,
      "grad_norm": 1.5280942916870117,
      "learning_rate": 2.033579746764419e-06,
      "loss": 0.0932,
      "num_input_tokens_seen": 65997760,
      "step": 34830
    },
    {
      "epoch": 5.682763683824129,
      "grad_norm": 0.20915623009204865,
      "learning_rate": 2.029703059264565e-06,
      "loss": 0.2214,
      "num_input_tokens_seen": 66008032,
      "step": 34835
    },
    {
      "epoch": 5.6835794110449465,
      "grad_norm": 0.8369594812393188,
      "learning_rate": 2.02582991407316e-06,
      "loss": 0.1443,
      "num_input_tokens_seen": 66017552,
      "step": 34840
    },
    {
      "epoch": 5.684395138265764,
      "grad_norm": 0.5460664629936218,
      "learning_rate": 2.0219603117874992e-06,
      "loss": 0.1568,
      "num_input_tokens_seen": 66026560,
      "step": 34845
    },
    {
      "epoch": 5.685210865486582,
      "grad_norm": 0.03360798954963684,
      "learning_rate": 2.0180942530043156e-06,
      "loss": 0.0507,
      "num_input_tokens_seen": 66035664,
      "step": 34850
    },
    {
      "epoch": 5.686026592707399,
      "grad_norm": 0.13771222531795502,
      "learning_rate": 2.0142317383198107e-06,
      "loss": 0.102,
      "num_input_tokens_seen": 66045504,
      "step": 34855
    },
    {
      "epoch": 5.686842319928216,
      "grad_norm": 0.3433722257614136,
      "learning_rate": 2.0103727683296243e-06,
      "loss": 0.0838,
      "num_input_tokens_seen": 66055664,
      "step": 34860
    },
    {
      "epoch": 5.687658047149033,
      "grad_norm": 0.10415921360254288,
      "learning_rate": 2.0065173436288636e-06,
      "loss": 0.0631,
      "num_input_tokens_seen": 66064336,
      "step": 34865
    },
    {
      "epoch": 5.688473774369851,
      "grad_norm": 0.021367579698562622,
      "learning_rate": 2.002665464812087e-06,
      "loss": 0.0708,
      "num_input_tokens_seen": 66074848,
      "step": 34870
    },
    {
      "epoch": 5.689289501590668,
      "grad_norm": 0.32161399722099304,
      "learning_rate": 1.998817132473291e-06,
      "loss": 0.089,
      "num_input_tokens_seen": 66085248,
      "step": 34875
    },
    {
      "epoch": 5.690105228811485,
      "grad_norm": 0.7345934510231018,
      "learning_rate": 1.9949723472059507e-06,
      "loss": 0.1646,
      "num_input_tokens_seen": 66094736,
      "step": 34880
    },
    {
      "epoch": 5.690920956032302,
      "grad_norm": 0.19519633054733276,
      "learning_rate": 1.9911311096029726e-06,
      "loss": 0.0589,
      "num_input_tokens_seen": 66105472,
      "step": 34885
    },
    {
      "epoch": 5.69173668325312,
      "grad_norm": 1.409328579902649,
      "learning_rate": 1.9872934202567224e-06,
      "loss": 0.1004,
      "num_input_tokens_seen": 66114256,
      "step": 34890
    },
    {
      "epoch": 5.692552410473938,
      "grad_norm": 0.6510486006736755,
      "learning_rate": 1.9834592797590257e-06,
      "loss": 0.2471,
      "num_input_tokens_seen": 66123648,
      "step": 34895
    },
    {
      "epoch": 5.693368137694755,
      "grad_norm": 0.24402382969856262,
      "learning_rate": 1.979628688701149e-06,
      "loss": 0.2242,
      "num_input_tokens_seen": 66133424,
      "step": 34900
    },
    {
      "epoch": 5.694183864915573,
      "grad_norm": 0.3752649426460266,
      "learning_rate": 1.9758016476738193e-06,
      "loss": 0.0484,
      "num_input_tokens_seen": 66144944,
      "step": 34905
    },
    {
      "epoch": 5.69499959213639,
      "grad_norm": 0.6054390668869019,
      "learning_rate": 1.971978157267221e-06,
      "loss": 0.1126,
      "num_input_tokens_seen": 66155552,
      "step": 34910
    },
    {
      "epoch": 5.695815319357207,
      "grad_norm": 2.170743465423584,
      "learning_rate": 1.968158218070973e-06,
      "loss": 0.1888,
      "num_input_tokens_seen": 66165040,
      "step": 34915
    },
    {
      "epoch": 5.696631046578024,
      "grad_norm": 0.11566080152988434,
      "learning_rate": 1.9643418306741682e-06,
      "loss": 0.0707,
      "num_input_tokens_seen": 66175120,
      "step": 34920
    },
    {
      "epoch": 5.697446773798841,
      "grad_norm": 0.2846405804157257,
      "learning_rate": 1.9605289956653337e-06,
      "loss": 0.0766,
      "num_input_tokens_seen": 66184496,
      "step": 34925
    },
    {
      "epoch": 5.698262501019659,
      "grad_norm": 1.2085753679275513,
      "learning_rate": 1.9567197136324626e-06,
      "loss": 0.1701,
      "num_input_tokens_seen": 66194384,
      "step": 34930
    },
    {
      "epoch": 5.699078228240476,
      "grad_norm": 0.06541457772254944,
      "learning_rate": 1.9529139851629935e-06,
      "loss": 0.1134,
      "num_input_tokens_seen": 66205072,
      "step": 34935
    },
    {
      "epoch": 5.6998939554612935,
      "grad_norm": 0.9698434472084045,
      "learning_rate": 1.949111810843812e-06,
      "loss": 0.1593,
      "num_input_tokens_seen": 66214560,
      "step": 34940
    },
    {
      "epoch": 5.7007096826821115,
      "grad_norm": 1.0761685371398926,
      "learning_rate": 1.9453131912612694e-06,
      "loss": 0.0568,
      "num_input_tokens_seen": 66224224,
      "step": 34945
    },
    {
      "epoch": 5.701525409902929,
      "grad_norm": 1.283669114112854,
      "learning_rate": 1.941518127001149e-06,
      "loss": 0.286,
      "num_input_tokens_seen": 66234880,
      "step": 34950
    },
    {
      "epoch": 5.702341137123746,
      "grad_norm": 0.6765309572219849,
      "learning_rate": 1.9377266186487107e-06,
      "loss": 0.0566,
      "num_input_tokens_seen": 66244096,
      "step": 34955
    },
    {
      "epoch": 5.703156864344563,
      "grad_norm": 0.28838619589805603,
      "learning_rate": 1.9339386667886483e-06,
      "loss": 0.023,
      "num_input_tokens_seen": 66254304,
      "step": 34960
    },
    {
      "epoch": 5.70397259156538,
      "grad_norm": 0.7186369895935059,
      "learning_rate": 1.9301542720051024e-06,
      "loss": 0.0969,
      "num_input_tokens_seen": 66263328,
      "step": 34965
    },
    {
      "epoch": 5.704788318786198,
      "grad_norm": 0.9494758248329163,
      "learning_rate": 1.926373434881684e-06,
      "loss": 0.118,
      "num_input_tokens_seen": 66271456,
      "step": 34970
    },
    {
      "epoch": 5.705604046007015,
      "grad_norm": 1.1168559789657593,
      "learning_rate": 1.9225961560014468e-06,
      "loss": 0.053,
      "num_input_tokens_seen": 66279584,
      "step": 34975
    },
    {
      "epoch": 5.706419773227832,
      "grad_norm": 0.28282925486564636,
      "learning_rate": 1.918822435946885e-06,
      "loss": 0.1018,
      "num_input_tokens_seen": 66290240,
      "step": 34980
    },
    {
      "epoch": 5.70723550044865,
      "grad_norm": 0.15202951431274414,
      "learning_rate": 1.915052275299961e-06,
      "loss": 0.0992,
      "num_input_tokens_seen": 66299056,
      "step": 34985
    },
    {
      "epoch": 5.7080512276694675,
      "grad_norm": 0.4789859354496002,
      "learning_rate": 1.9112856746420854e-06,
      "loss": 0.0415,
      "num_input_tokens_seen": 66308976,
      "step": 34990
    },
    {
      "epoch": 5.708866954890285,
      "grad_norm": 0.6059602499008179,
      "learning_rate": 1.907522634554104e-06,
      "loss": 0.1883,
      "num_input_tokens_seen": 66318336,
      "step": 34995
    },
    {
      "epoch": 5.709682682111102,
      "grad_norm": 0.03985058143734932,
      "learning_rate": 1.9037631556163337e-06,
      "loss": 0.1105,
      "num_input_tokens_seen": 66326768,
      "step": 35000
    },
    {
      "epoch": 5.709682682111102,
      "eval_loss": 0.1420750617980957,
      "eval_runtime": 133.2096,
      "eval_samples_per_second": 20.456,
      "eval_steps_per_second": 10.232,
      "num_input_tokens_seen": 66326768,
      "step": 35000
    },
    {
      "epoch": 5.71049840933192,
      "grad_norm": 0.1551789939403534,
      "learning_rate": 1.9000072384085272e-06,
      "loss": 0.0535,
      "num_input_tokens_seen": 66335968,
      "step": 35005
    },
    {
      "epoch": 5.711314136552737,
      "grad_norm": 0.0541737899184227,
      "learning_rate": 1.8962548835098987e-06,
      "loss": 0.0208,
      "num_input_tokens_seen": 66343728,
      "step": 35010
    },
    {
      "epoch": 5.712129863773554,
      "grad_norm": 0.2733137905597687,
      "learning_rate": 1.8925060914991077e-06,
      "loss": 0.1846,
      "num_input_tokens_seen": 66352640,
      "step": 35015
    },
    {
      "epoch": 5.712945590994371,
      "grad_norm": 0.809597909450531,
      "learning_rate": 1.888760862954264e-06,
      "loss": 0.1005,
      "num_input_tokens_seen": 66361680,
      "step": 35020
    },
    {
      "epoch": 5.713761318215189,
      "grad_norm": 0.129475936293602,
      "learning_rate": 1.8850191984529309e-06,
      "loss": 0.0779,
      "num_input_tokens_seen": 66370368,
      "step": 35025
    },
    {
      "epoch": 5.714577045436006,
      "grad_norm": 0.7254397869110107,
      "learning_rate": 1.8812810985721186e-06,
      "loss": 0.0824,
      "num_input_tokens_seen": 66381120,
      "step": 35030
    },
    {
      "epoch": 5.715392772656823,
      "grad_norm": 0.2062244862318039,
      "learning_rate": 1.8775465638882856e-06,
      "loss": 0.098,
      "num_input_tokens_seen": 66390256,
      "step": 35035
    },
    {
      "epoch": 5.716208499877641,
      "grad_norm": 0.37693050503730774,
      "learning_rate": 1.8738155949773517e-06,
      "loss": 0.0269,
      "num_input_tokens_seen": 66399152,
      "step": 35040
    },
    {
      "epoch": 5.717024227098459,
      "grad_norm": 1.2484321594238281,
      "learning_rate": 1.8700881924146707e-06,
      "loss": 0.2106,
      "num_input_tokens_seen": 66407488,
      "step": 35045
    },
    {
      "epoch": 5.717839954319276,
      "grad_norm": 1.2993433475494385,
      "learning_rate": 1.8663643567750577e-06,
      "loss": 0.2405,
      "num_input_tokens_seen": 66418144,
      "step": 35050
    },
    {
      "epoch": 5.718655681540093,
      "grad_norm": 0.19333074986934662,
      "learning_rate": 1.8626440886327813e-06,
      "loss": 0.0558,
      "num_input_tokens_seen": 66427568,
      "step": 35055
    },
    {
      "epoch": 5.71947140876091,
      "grad_norm": 0.13372503221035004,
      "learning_rate": 1.8589273885615432e-06,
      "loss": 0.1679,
      "num_input_tokens_seen": 66438160,
      "step": 35060
    },
    {
      "epoch": 5.720287135981728,
      "grad_norm": 0.0885121077299118,
      "learning_rate": 1.8552142571345133e-06,
      "loss": 0.0782,
      "num_input_tokens_seen": 66447568,
      "step": 35065
    },
    {
      "epoch": 5.721102863202545,
      "grad_norm": 0.24180126190185547,
      "learning_rate": 1.8515046949243025e-06,
      "loss": 0.0288,
      "num_input_tokens_seen": 66457280,
      "step": 35070
    },
    {
      "epoch": 5.721918590423362,
      "grad_norm": 0.10371411591768265,
      "learning_rate": 1.8477987025029674e-06,
      "loss": 0.0709,
      "num_input_tokens_seen": 66467600,
      "step": 35075
    },
    {
      "epoch": 5.72273431764418,
      "grad_norm": 0.4982903301715851,
      "learning_rate": 1.8440962804420232e-06,
      "loss": 0.0983,
      "num_input_tokens_seen": 66476992,
      "step": 35080
    },
    {
      "epoch": 5.723550044864997,
      "grad_norm": 1.3811203241348267,
      "learning_rate": 1.8403974293124265e-06,
      "loss": 0.1255,
      "num_input_tokens_seen": 66485520,
      "step": 35085
    },
    {
      "epoch": 5.7243657720858145,
      "grad_norm": 0.06459513306617737,
      "learning_rate": 1.8367021496845854e-06,
      "loss": 0.1252,
      "num_input_tokens_seen": 66495600,
      "step": 35090
    },
    {
      "epoch": 5.725181499306632,
      "grad_norm": 0.0338037945330143,
      "learning_rate": 1.8330104421283662e-06,
      "loss": 0.0248,
      "num_input_tokens_seen": 66504832,
      "step": 35095
    },
    {
      "epoch": 5.725997226527449,
      "grad_norm": 0.8511983156204224,
      "learning_rate": 1.8293223072130717e-06,
      "loss": 0.1531,
      "num_input_tokens_seen": 66514784,
      "step": 35100
    },
    {
      "epoch": 5.726812953748267,
      "grad_norm": 0.0644291490316391,
      "learning_rate": 1.8256377455074525e-06,
      "loss": 0.0844,
      "num_input_tokens_seen": 66524464,
      "step": 35105
    },
    {
      "epoch": 5.727628680969084,
      "grad_norm": 0.18543298542499542,
      "learning_rate": 1.8219567575797263e-06,
      "loss": 0.0637,
      "num_input_tokens_seen": 66534336,
      "step": 35110
    },
    {
      "epoch": 5.728444408189901,
      "grad_norm": 0.7770499587059021,
      "learning_rate": 1.8182793439975365e-06,
      "loss": 0.0512,
      "num_input_tokens_seen": 66543952,
      "step": 35115
    },
    {
      "epoch": 5.729260135410719,
      "grad_norm": 0.3912498354911804,
      "learning_rate": 1.8146055053279958e-06,
      "loss": 0.1884,
      "num_input_tokens_seen": 66553760,
      "step": 35120
    },
    {
      "epoch": 5.730075862631536,
      "grad_norm": 0.21439802646636963,
      "learning_rate": 1.8109352421376486e-06,
      "loss": 0.1634,
      "num_input_tokens_seen": 66562192,
      "step": 35125
    },
    {
      "epoch": 5.730891589852353,
      "grad_norm": 0.4280273914337158,
      "learning_rate": 1.8072685549924972e-06,
      "loss": 0.1241,
      "num_input_tokens_seen": 66572256,
      "step": 35130
    },
    {
      "epoch": 5.7317073170731705,
      "grad_norm": 0.25395527482032776,
      "learning_rate": 1.8036054444579982e-06,
      "loss": 0.0949,
      "num_input_tokens_seen": 66581376,
      "step": 35135
    },
    {
      "epoch": 5.732523044293988,
      "grad_norm": 0.07760205119848251,
      "learning_rate": 1.7999459110990407e-06,
      "loss": 0.0898,
      "num_input_tokens_seen": 66592256,
      "step": 35140
    },
    {
      "epoch": 5.733338771514806,
      "grad_norm": 0.366260826587677,
      "learning_rate": 1.7962899554799712e-06,
      "loss": 0.1919,
      "num_input_tokens_seen": 66602272,
      "step": 35145
    },
    {
      "epoch": 5.734154498735623,
      "grad_norm": 0.7660773396492004,
      "learning_rate": 1.7926375781645937e-06,
      "loss": 0.0611,
      "num_input_tokens_seen": 66612400,
      "step": 35150
    },
    {
      "epoch": 5.73497022595644,
      "grad_norm": 0.11699582636356354,
      "learning_rate": 1.7889887797161359e-06,
      "loss": 0.1082,
      "num_input_tokens_seen": 66623248,
      "step": 35155
    },
    {
      "epoch": 5.735785953177258,
      "grad_norm": 2.626810312271118,
      "learning_rate": 1.7853435606973028e-06,
      "loss": 0.3595,
      "num_input_tokens_seen": 66632880,
      "step": 35160
    },
    {
      "epoch": 5.736601680398075,
      "grad_norm": 0.19238348305225372,
      "learning_rate": 1.781701921670223e-06,
      "loss": 0.2741,
      "num_input_tokens_seen": 66642688,
      "step": 35165
    },
    {
      "epoch": 5.737417407618892,
      "grad_norm": 0.6691216826438904,
      "learning_rate": 1.7780638631964886e-06,
      "loss": 0.1604,
      "num_input_tokens_seen": 66652288,
      "step": 35170
    },
    {
      "epoch": 5.738233134839709,
      "grad_norm": 0.19618400931358337,
      "learning_rate": 1.7744293858371314e-06,
      "loss": 0.1502,
      "num_input_tokens_seen": 66660928,
      "step": 35175
    },
    {
      "epoch": 5.739048862060527,
      "grad_norm": 1.4192416667938232,
      "learning_rate": 1.770798490152631e-06,
      "loss": 0.1782,
      "num_input_tokens_seen": 66670240,
      "step": 35180
    },
    {
      "epoch": 5.7398645892813445,
      "grad_norm": 0.18777605891227722,
      "learning_rate": 1.767171176702917e-06,
      "loss": 0.1571,
      "num_input_tokens_seen": 66679360,
      "step": 35185
    },
    {
      "epoch": 5.740680316502162,
      "grad_norm": 1.1597036123275757,
      "learning_rate": 1.7635474460473755e-06,
      "loss": 0.146,
      "num_input_tokens_seen": 66688032,
      "step": 35190
    },
    {
      "epoch": 5.741496043722979,
      "grad_norm": 0.17707909643650055,
      "learning_rate": 1.7599272987448206e-06,
      "loss": 0.1129,
      "num_input_tokens_seen": 66696512,
      "step": 35195
    },
    {
      "epoch": 5.742311770943797,
      "grad_norm": 1.0368843078613281,
      "learning_rate": 1.7563107353535362e-06,
      "loss": 0.1614,
      "num_input_tokens_seen": 66705744,
      "step": 35200
    },
    {
      "epoch": 5.742311770943797,
      "eval_loss": 0.14190725982189178,
      "eval_runtime": 133.3764,
      "eval_samples_per_second": 20.431,
      "eval_steps_per_second": 10.219,
      "num_input_tokens_seen": 66705744,
      "step": 35200
    },
    {
      "epoch": 5.743127498164614,
      "grad_norm": 0.3416934907436371,
      "learning_rate": 1.7526977564312263e-06,
      "loss": 0.0713,
      "num_input_tokens_seen": 66715520,
      "step": 35205
    },
    {
      "epoch": 5.743943225385431,
      "grad_norm": 0.20047102868556976,
      "learning_rate": 1.7490883625350701e-06,
      "loss": 0.1276,
      "num_input_tokens_seen": 66726208,
      "step": 35210
    },
    {
      "epoch": 5.744758952606248,
      "grad_norm": 1.842671513557434,
      "learning_rate": 1.7454825542216807e-06,
      "loss": 0.1774,
      "num_input_tokens_seen": 66736128,
      "step": 35215
    },
    {
      "epoch": 5.745574679827066,
      "grad_norm": 1.0995780229568481,
      "learning_rate": 1.7418803320471105e-06,
      "loss": 0.3083,
      "num_input_tokens_seen": 66745952,
      "step": 35220
    },
    {
      "epoch": 5.746390407047883,
      "grad_norm": 0.2572075426578522,
      "learning_rate": 1.7382816965668737e-06,
      "loss": 0.2705,
      "num_input_tokens_seen": 66755792,
      "step": 35225
    },
    {
      "epoch": 5.7472061342687,
      "grad_norm": 1.3731019496917725,
      "learning_rate": 1.7346866483359285e-06,
      "loss": 0.1731,
      "num_input_tokens_seen": 66765360,
      "step": 35230
    },
    {
      "epoch": 5.7480218614895175,
      "grad_norm": 0.09883128106594086,
      "learning_rate": 1.7310951879086657e-06,
      "loss": 0.0899,
      "num_input_tokens_seen": 66775776,
      "step": 35235
    },
    {
      "epoch": 5.748837588710336,
      "grad_norm": 0.2272624522447586,
      "learning_rate": 1.7275073158389471e-06,
      "loss": 0.1142,
      "num_input_tokens_seen": 66785008,
      "step": 35240
    },
    {
      "epoch": 5.749653315931153,
      "grad_norm": 1.1834039688110352,
      "learning_rate": 1.723923032680061e-06,
      "loss": 0.2283,
      "num_input_tokens_seen": 66795040,
      "step": 35245
    },
    {
      "epoch": 5.75046904315197,
      "grad_norm": 0.2552722990512848,
      "learning_rate": 1.7203423389847428e-06,
      "loss": 0.1586,
      "num_input_tokens_seen": 66803504,
      "step": 35250
    },
    {
      "epoch": 5.751284770372787,
      "grad_norm": 0.7274467945098877,
      "learning_rate": 1.7167652353051928e-06,
      "loss": 0.0884,
      "num_input_tokens_seen": 66814064,
      "step": 35255
    },
    {
      "epoch": 5.752100497593605,
      "grad_norm": 0.5149352550506592,
      "learning_rate": 1.7131917221930333e-06,
      "loss": 0.0583,
      "num_input_tokens_seen": 66823280,
      "step": 35260
    },
    {
      "epoch": 5.752916224814422,
      "grad_norm": 1.0344712734222412,
      "learning_rate": 1.7096218001993513e-06,
      "loss": 0.1596,
      "num_input_tokens_seen": 66832384,
      "step": 35265
    },
    {
      "epoch": 5.753731952035239,
      "grad_norm": 0.04484211653470993,
      "learning_rate": 1.706055469874676e-06,
      "loss": 0.0454,
      "num_input_tokens_seen": 66840592,
      "step": 35270
    },
    {
      "epoch": 5.754547679256056,
      "grad_norm": 0.19975945353507996,
      "learning_rate": 1.702492731768976e-06,
      "loss": 0.1252,
      "num_input_tokens_seen": 66850448,
      "step": 35275
    },
    {
      "epoch": 5.755363406476874,
      "grad_norm": 1.9445421695709229,
      "learning_rate": 1.6989335864316724e-06,
      "loss": 0.1288,
      "num_input_tokens_seen": 66859536,
      "step": 35280
    },
    {
      "epoch": 5.7561791336976915,
      "grad_norm": 1.80393648147583,
      "learning_rate": 1.6953780344116265e-06,
      "loss": 0.1695,
      "num_input_tokens_seen": 66869408,
      "step": 35285
    },
    {
      "epoch": 5.756994860918509,
      "grad_norm": 0.5034258961677551,
      "learning_rate": 1.6918260762571497e-06,
      "loss": 0.112,
      "num_input_tokens_seen": 66879728,
      "step": 35290
    },
    {
      "epoch": 5.757810588139327,
      "grad_norm": 0.34835535287857056,
      "learning_rate": 1.6882777125160093e-06,
      "loss": 0.0728,
      "num_input_tokens_seen": 66888704,
      "step": 35295
    },
    {
      "epoch": 5.758626315360144,
      "grad_norm": 1.6458505392074585,
      "learning_rate": 1.6847329437353899e-06,
      "loss": 0.1879,
      "num_input_tokens_seen": 66897872,
      "step": 35300
    },
    {
      "epoch": 5.759442042580961,
      "grad_norm": 1.252160668373108,
      "learning_rate": 1.6811917704619511e-06,
      "loss": 0.0978,
      "num_input_tokens_seen": 66906832,
      "step": 35305
    },
    {
      "epoch": 5.760257769801778,
      "grad_norm": 0.43828484416007996,
      "learning_rate": 1.67765419324179e-06,
      "loss": 0.1795,
      "num_input_tokens_seen": 66915024,
      "step": 35310
    },
    {
      "epoch": 5.761073497022595,
      "grad_norm": 0.8347132802009583,
      "learning_rate": 1.6741202126204364e-06,
      "loss": 0.2208,
      "num_input_tokens_seen": 66924624,
      "step": 35315
    },
    {
      "epoch": 5.761889224243413,
      "grad_norm": 1.1849592924118042,
      "learning_rate": 1.6705898291428767e-06,
      "loss": 0.1029,
      "num_input_tokens_seen": 66933728,
      "step": 35320
    },
    {
      "epoch": 5.76270495146423,
      "grad_norm": 0.47846636176109314,
      "learning_rate": 1.6670630433535395e-06,
      "loss": 0.2343,
      "num_input_tokens_seen": 66943648,
      "step": 35325
    },
    {
      "epoch": 5.7635206786850475,
      "grad_norm": 1.8237205743789673,
      "learning_rate": 1.6635398557962979e-06,
      "loss": 0.1564,
      "num_input_tokens_seen": 66951760,
      "step": 35330
    },
    {
      "epoch": 5.7643364059058655,
      "grad_norm": 0.16204817593097687,
      "learning_rate": 1.660020267014481e-06,
      "loss": 0.1188,
      "num_input_tokens_seen": 66961200,
      "step": 35335
    },
    {
      "epoch": 5.765152133126683,
      "grad_norm": 0.7855775356292725,
      "learning_rate": 1.6565042775508438e-06,
      "loss": 0.2355,
      "num_input_tokens_seen": 66970208,
      "step": 35340
    },
    {
      "epoch": 5.7659678603475,
      "grad_norm": 1.6745775938034058,
      "learning_rate": 1.6529918879475997e-06,
      "loss": 0.1728,
      "num_input_tokens_seen": 66980272,
      "step": 35345
    },
    {
      "epoch": 5.766783587568317,
      "grad_norm": 1.9562773704528809,
      "learning_rate": 1.6494830987464043e-06,
      "loss": 0.2003,
      "num_input_tokens_seen": 66989584,
      "step": 35350
    },
    {
      "epoch": 5.767599314789135,
      "grad_norm": 0.5598201751708984,
      "learning_rate": 1.6459779104883555e-06,
      "loss": 0.2,
      "num_input_tokens_seen": 66997840,
      "step": 35355
    },
    {
      "epoch": 5.768415042009952,
      "grad_norm": 0.798901379108429,
      "learning_rate": 1.6424763237140013e-06,
      "loss": 0.1263,
      "num_input_tokens_seen": 67006432,
      "step": 35360
    },
    {
      "epoch": 5.769230769230769,
      "grad_norm": 0.855695903301239,
      "learning_rate": 1.6389783389633207e-06,
      "loss": 0.0932,
      "num_input_tokens_seen": 67016640,
      "step": 35365
    },
    {
      "epoch": 5.770046496451586,
      "grad_norm": 0.23979870975017548,
      "learning_rate": 1.6354839567757546e-06,
      "loss": 0.1834,
      "num_input_tokens_seen": 67026288,
      "step": 35370
    },
    {
      "epoch": 5.770862223672404,
      "grad_norm": 0.49368980526924133,
      "learning_rate": 1.6319931776901831e-06,
      "loss": 0.0648,
      "num_input_tokens_seen": 67036560,
      "step": 35375
    },
    {
      "epoch": 5.771677950893221,
      "grad_norm": 0.12220870703458786,
      "learning_rate": 1.6285060022449229e-06,
      "loss": 0.1071,
      "num_input_tokens_seen": 67046112,
      "step": 35380
    },
    {
      "epoch": 5.772493678114039,
      "grad_norm": 0.6839032173156738,
      "learning_rate": 1.6250224309777434e-06,
      "loss": 0.0921,
      "num_input_tokens_seen": 67056832,
      "step": 35385
    },
    {
      "epoch": 5.773309405334856,
      "grad_norm": 0.19646252691745758,
      "learning_rate": 1.6215424644258515e-06,
      "loss": 0.1126,
      "num_input_tokens_seen": 67064784,
      "step": 35390
    },
    {
      "epoch": 5.774125132555674,
      "grad_norm": 1.2140848636627197,
      "learning_rate": 1.6180661031259036e-06,
      "loss": 0.0952,
      "num_input_tokens_seen": 67074720,
      "step": 35395
    },
    {
      "epoch": 5.774940859776491,
      "grad_norm": 0.5636078119277954,
      "learning_rate": 1.614593347613999e-06,
      "loss": 0.2069,
      "num_input_tokens_seen": 67084928,
      "step": 35400
    },
    {
      "epoch": 5.774940859776491,
      "eval_loss": 0.14219442009925842,
      "eval_runtime": 133.2709,
      "eval_samples_per_second": 20.447,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 67084928,
      "step": 35400
    },
    {
      "epoch": 5.775756586997308,
      "grad_norm": 1.0003244876861572,
      "learning_rate": 1.6111241984256758e-06,
      "loss": 0.1147,
      "num_input_tokens_seen": 67094208,
      "step": 35405
    },
    {
      "epoch": 5.776572314218125,
      "grad_norm": 0.07430919259786606,
      "learning_rate": 1.6076586560959257e-06,
      "loss": 0.0366,
      "num_input_tokens_seen": 67103872,
      "step": 35410
    },
    {
      "epoch": 5.777388041438943,
      "grad_norm": 1.684292197227478,
      "learning_rate": 1.604196721159182e-06,
      "loss": 0.1668,
      "num_input_tokens_seen": 67112816,
      "step": 35415
    },
    {
      "epoch": 5.77820376865976,
      "grad_norm": 0.22564175724983215,
      "learning_rate": 1.6007383941493092e-06,
      "loss": 0.2775,
      "num_input_tokens_seen": 67122480,
      "step": 35420
    },
    {
      "epoch": 5.779019495880577,
      "grad_norm": 0.06615103781223297,
      "learning_rate": 1.5972836755996285e-06,
      "loss": 0.1016,
      "num_input_tokens_seen": 67130736,
      "step": 35425
    },
    {
      "epoch": 5.7798352231013945,
      "grad_norm": 0.9675799012184143,
      "learning_rate": 1.5938325660429076e-06,
      "loss": 0.1529,
      "num_input_tokens_seen": 67140480,
      "step": 35430
    },
    {
      "epoch": 5.7806509503222125,
      "grad_norm": 0.40957802534103394,
      "learning_rate": 1.5903850660113378e-06,
      "loss": 0.2263,
      "num_input_tokens_seen": 67149792,
      "step": 35435
    },
    {
      "epoch": 5.78146667754303,
      "grad_norm": 1.3402334451675415,
      "learning_rate": 1.5869411760365826e-06,
      "loss": 0.1426,
      "num_input_tokens_seen": 67158688,
      "step": 35440
    },
    {
      "epoch": 5.782282404763847,
      "grad_norm": 0.48301389813423157,
      "learning_rate": 1.58350089664972e-06,
      "loss": 0.0622,
      "num_input_tokens_seen": 67168224,
      "step": 35445
    },
    {
      "epoch": 5.783098131984664,
      "grad_norm": 0.05570094659924507,
      "learning_rate": 1.5800642283812865e-06,
      "loss": 0.0277,
      "num_input_tokens_seen": 67178208,
      "step": 35450
    },
    {
      "epoch": 5.783913859205482,
      "grad_norm": 0.4242340326309204,
      "learning_rate": 1.5766311717612698e-06,
      "loss": 0.0603,
      "num_input_tokens_seen": 67188064,
      "step": 35455
    },
    {
      "epoch": 5.784729586426299,
      "grad_norm": 1.0694303512573242,
      "learning_rate": 1.5732017273190818e-06,
      "loss": 0.1464,
      "num_input_tokens_seen": 67198080,
      "step": 35460
    },
    {
      "epoch": 5.785545313647116,
      "grad_norm": 0.5790186524391174,
      "learning_rate": 1.5697758955835806e-06,
      "loss": 0.1219,
      "num_input_tokens_seen": 67207840,
      "step": 35465
    },
    {
      "epoch": 5.786361040867934,
      "grad_norm": 0.402442991733551,
      "learning_rate": 1.566353677083085e-06,
      "loss": 0.0906,
      "num_input_tokens_seen": 67217216,
      "step": 35470
    },
    {
      "epoch": 5.787176768088751,
      "grad_norm": 0.35393837094306946,
      "learning_rate": 1.562935072345334e-06,
      "loss": 0.0657,
      "num_input_tokens_seen": 67226448,
      "step": 35475
    },
    {
      "epoch": 5.7879924953095685,
      "grad_norm": 0.1294674426317215,
      "learning_rate": 1.5595200818975281e-06,
      "loss": 0.1407,
      "num_input_tokens_seen": 67235328,
      "step": 35480
    },
    {
      "epoch": 5.788808222530386,
      "grad_norm": 0.2038077861070633,
      "learning_rate": 1.5561087062662905e-06,
      "loss": 0.0616,
      "num_input_tokens_seen": 67244368,
      "step": 35485
    },
    {
      "epoch": 5.789623949751203,
      "grad_norm": 0.6072971224784851,
      "learning_rate": 1.5527009459777087e-06,
      "loss": 0.2199,
      "num_input_tokens_seen": 67253648,
      "step": 35490
    },
    {
      "epoch": 5.790439676972021,
      "grad_norm": 0.323087602853775,
      "learning_rate": 1.5492968015572984e-06,
      "loss": 0.0712,
      "num_input_tokens_seen": 67264624,
      "step": 35495
    },
    {
      "epoch": 5.791255404192838,
      "grad_norm": 0.4031101167201996,
      "learning_rate": 1.5458962735300203e-06,
      "loss": 0.1005,
      "num_input_tokens_seen": 67275216,
      "step": 35500
    },
    {
      "epoch": 5.792071131413655,
      "grad_norm": 0.2168450653553009,
      "learning_rate": 1.54249936242028e-06,
      "loss": 0.1073,
      "num_input_tokens_seen": 67284352,
      "step": 35505
    },
    {
      "epoch": 5.792886858634473,
      "grad_norm": 0.2952955961227417,
      "learning_rate": 1.5391060687519222e-06,
      "loss": 0.1361,
      "num_input_tokens_seen": 67294000,
      "step": 35510
    },
    {
      "epoch": 5.79370258585529,
      "grad_norm": 0.1375323385000229,
      "learning_rate": 1.5357163930482367e-06,
      "loss": 0.0271,
      "num_input_tokens_seen": 67303808,
      "step": 35515
    },
    {
      "epoch": 5.794518313076107,
      "grad_norm": 1.2830324172973633,
      "learning_rate": 1.532330335831955e-06,
      "loss": 0.0985,
      "num_input_tokens_seen": 67313520,
      "step": 35520
    },
    {
      "epoch": 5.7953340402969244,
      "grad_norm": 2.44299054145813,
      "learning_rate": 1.5289478976252491e-06,
      "loss": 0.1631,
      "num_input_tokens_seen": 67321920,
      "step": 35525
    },
    {
      "epoch": 5.796149767517742,
      "grad_norm": 0.2733326256275177,
      "learning_rate": 1.5255690789497345e-06,
      "loss": 0.1393,
      "num_input_tokens_seen": 67331120,
      "step": 35530
    },
    {
      "epoch": 5.79696549473856,
      "grad_norm": 1.6221377849578857,
      "learning_rate": 1.5221938803264641e-06,
      "loss": 0.2047,
      "num_input_tokens_seen": 67341184,
      "step": 35535
    },
    {
      "epoch": 5.797781221959377,
      "grad_norm": 0.09939737617969513,
      "learning_rate": 1.518822302275938e-06,
      "loss": 0.1978,
      "num_input_tokens_seen": 67350864,
      "step": 35540
    },
    {
      "epoch": 5.798596949180194,
      "grad_norm": 0.23873448371887207,
      "learning_rate": 1.5154543453180958e-06,
      "loss": 0.0584,
      "num_input_tokens_seen": 67359664,
      "step": 35545
    },
    {
      "epoch": 5.799412676401012,
      "grad_norm": 0.08457756042480469,
      "learning_rate": 1.5120900099723167e-06,
      "loss": 0.0988,
      "num_input_tokens_seen": 67369520,
      "step": 35550
    },
    {
      "epoch": 5.800228403621829,
      "grad_norm": 1.4831854104995728,
      "learning_rate": 1.5087292967574273e-06,
      "loss": 0.1723,
      "num_input_tokens_seen": 67378160,
      "step": 35555
    },
    {
      "epoch": 5.801044130842646,
      "grad_norm": 0.10286463797092438,
      "learning_rate": 1.5053722061916908e-06,
      "loss": 0.068,
      "num_input_tokens_seen": 67387472,
      "step": 35560
    },
    {
      "epoch": 5.801859858063463,
      "grad_norm": 1.3662123680114746,
      "learning_rate": 1.5020187387928124e-06,
      "loss": 0.2759,
      "num_input_tokens_seen": 67396064,
      "step": 35565
    },
    {
      "epoch": 5.802675585284281,
      "grad_norm": 0.2945880591869354,
      "learning_rate": 1.4986688950779343e-06,
      "loss": 0.0657,
      "num_input_tokens_seen": 67404832,
      "step": 35570
    },
    {
      "epoch": 5.803491312505098,
      "grad_norm": 1.267778992652893,
      "learning_rate": 1.495322675563654e-06,
      "loss": 0.1115,
      "num_input_tokens_seen": 67414688,
      "step": 35575
    },
    {
      "epoch": 5.8043070397259156,
      "grad_norm": 0.8979448676109314,
      "learning_rate": 1.4919800807659922e-06,
      "loss": 0.1712,
      "num_input_tokens_seen": 67424352,
      "step": 35580
    },
    {
      "epoch": 5.805122766946733,
      "grad_norm": 0.9064595103263855,
      "learning_rate": 1.4886411112004255e-06,
      "loss": 0.0954,
      "num_input_tokens_seen": 67434224,
      "step": 35585
    },
    {
      "epoch": 5.805938494167551,
      "grad_norm": 0.2764817178249359,
      "learning_rate": 1.4853057673818588e-06,
      "loss": 0.1152,
      "num_input_tokens_seen": 67443312,
      "step": 35590
    },
    {
      "epoch": 5.806754221388368,
      "grad_norm": 0.35370519757270813,
      "learning_rate": 1.481974049824647e-06,
      "loss": 0.219,
      "num_input_tokens_seen": 67452224,
      "step": 35595
    },
    {
      "epoch": 5.807569948609185,
      "grad_norm": 0.2795799672603607,
      "learning_rate": 1.4786459590425849e-06,
      "loss": 0.0295,
      "num_input_tokens_seen": 67462064,
      "step": 35600
    },
    {
      "epoch": 5.807569948609185,
      "eval_loss": 0.14190401136875153,
      "eval_runtime": 133.2572,
      "eval_samples_per_second": 20.449,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 67462064,
      "step": 35600
    },
    {
      "epoch": 5.808385675830002,
      "grad_norm": 0.46416372060775757,
      "learning_rate": 1.4753214955489036e-06,
      "loss": 0.0631,
      "num_input_tokens_seen": 67471728,
      "step": 35605
    },
    {
      "epoch": 5.80920140305082,
      "grad_norm": 1.203023910522461,
      "learning_rate": 1.4720006598562737e-06,
      "loss": 0.1135,
      "num_input_tokens_seen": 67481968,
      "step": 35610
    },
    {
      "epoch": 5.810017130271637,
      "grad_norm": 0.060998912900686264,
      "learning_rate": 1.4686834524768185e-06,
      "loss": 0.0801,
      "num_input_tokens_seen": 67492304,
      "step": 35615
    },
    {
      "epoch": 5.810832857492454,
      "grad_norm": 1.000280737876892,
      "learning_rate": 1.4653698739220844e-06,
      "loss": 0.1113,
      "num_input_tokens_seen": 67500816,
      "step": 35620
    },
    {
      "epoch": 5.8116485847132715,
      "grad_norm": 2.889709234237671,
      "learning_rate": 1.4620599247030715e-06,
      "loss": 0.2333,
      "num_input_tokens_seen": 67511232,
      "step": 35625
    },
    {
      "epoch": 5.8124643119340895,
      "grad_norm": 0.32291674613952637,
      "learning_rate": 1.4587536053302125e-06,
      "loss": 0.1204,
      "num_input_tokens_seen": 67520048,
      "step": 35630
    },
    {
      "epoch": 5.813280039154907,
      "grad_norm": 1.0392799377441406,
      "learning_rate": 1.4554509163133862e-06,
      "loss": 0.0592,
      "num_input_tokens_seen": 67528480,
      "step": 35635
    },
    {
      "epoch": 5.814095766375724,
      "grad_norm": 1.720810890197754,
      "learning_rate": 1.4521518581619098e-06,
      "loss": 0.213,
      "num_input_tokens_seen": 67537904,
      "step": 35640
    },
    {
      "epoch": 5.814911493596542,
      "grad_norm": 0.29965740442276,
      "learning_rate": 1.4488564313845348e-06,
      "loss": 0.0692,
      "num_input_tokens_seen": 67546800,
      "step": 35645
    },
    {
      "epoch": 5.815727220817359,
      "grad_norm": 1.2954857349395752,
      "learning_rate": 1.4455646364894603e-06,
      "loss": 0.1219,
      "num_input_tokens_seen": 67555904,
      "step": 35650
    },
    {
      "epoch": 5.816542948038176,
      "grad_norm": 2.185537815093994,
      "learning_rate": 1.4422764739843247e-06,
      "loss": 0.1821,
      "num_input_tokens_seen": 67564560,
      "step": 35655
    },
    {
      "epoch": 5.817358675258993,
      "grad_norm": 0.7651224136352539,
      "learning_rate": 1.4389919443762e-06,
      "loss": 0.0596,
      "num_input_tokens_seen": 67573232,
      "step": 35660
    },
    {
      "epoch": 5.81817440247981,
      "grad_norm": 1.1012009382247925,
      "learning_rate": 1.4357110481716063e-06,
      "loss": 0.1234,
      "num_input_tokens_seen": 67584112,
      "step": 35665
    },
    {
      "epoch": 5.818990129700628,
      "grad_norm": 0.03100796416401863,
      "learning_rate": 1.4324337858764941e-06,
      "loss": 0.1378,
      "num_input_tokens_seen": 67592592,
      "step": 35670
    },
    {
      "epoch": 5.8198058569214455,
      "grad_norm": 1.8732504844665527,
      "learning_rate": 1.4291601579962622e-06,
      "loss": 0.2669,
      "num_input_tokens_seen": 67602208,
      "step": 35675
    },
    {
      "epoch": 5.820621584142263,
      "grad_norm": 0.9620072245597839,
      "learning_rate": 1.42589016503574e-06,
      "loss": 0.0833,
      "num_input_tokens_seen": 67611760,
      "step": 35680
    },
    {
      "epoch": 5.821437311363081,
      "grad_norm": 0.055021390318870544,
      "learning_rate": 1.4226238074992099e-06,
      "loss": 0.1356,
      "num_input_tokens_seen": 67622384,
      "step": 35685
    },
    {
      "epoch": 5.822253038583898,
      "grad_norm": 0.7670300602912903,
      "learning_rate": 1.4193610858903778e-06,
      "loss": 0.0917,
      "num_input_tokens_seen": 67631456,
      "step": 35690
    },
    {
      "epoch": 5.823068765804715,
      "grad_norm": 1.2150070667266846,
      "learning_rate": 1.416102000712402e-06,
      "loss": 0.0777,
      "num_input_tokens_seen": 67641536,
      "step": 35695
    },
    {
      "epoch": 5.823884493025532,
      "grad_norm": 1.8045958280563354,
      "learning_rate": 1.4128465524678668e-06,
      "loss": 0.1111,
      "num_input_tokens_seen": 67648416,
      "step": 35700
    },
    {
      "epoch": 5.824700220246349,
      "grad_norm": 1.4041857719421387,
      "learning_rate": 1.4095947416588124e-06,
      "loss": 0.1305,
      "num_input_tokens_seen": 67658000,
      "step": 35705
    },
    {
      "epoch": 5.825515947467167,
      "grad_norm": 0.1734107881784439,
      "learning_rate": 1.4063465687866983e-06,
      "loss": 0.2217,
      "num_input_tokens_seen": 67669104,
      "step": 35710
    },
    {
      "epoch": 5.826331674687984,
      "grad_norm": 0.24812185764312744,
      "learning_rate": 1.4031020343524438e-06,
      "loss": 0.2244,
      "num_input_tokens_seen": 67679040,
      "step": 35715
    },
    {
      "epoch": 5.827147401908801,
      "grad_norm": 0.3980655372142792,
      "learning_rate": 1.3998611388563926e-06,
      "loss": 0.0331,
      "num_input_tokens_seen": 67689712,
      "step": 35720
    },
    {
      "epoch": 5.8279631291296194,
      "grad_norm": 0.19322821497917175,
      "learning_rate": 1.3966238827983314e-06,
      "loss": 0.0506,
      "num_input_tokens_seen": 67700352,
      "step": 35725
    },
    {
      "epoch": 5.828778856350437,
      "grad_norm": 1.3726533651351929,
      "learning_rate": 1.393390266677483e-06,
      "loss": 0.1735,
      "num_input_tokens_seen": 67708880,
      "step": 35730
    },
    {
      "epoch": 5.829594583571254,
      "grad_norm": 1.3879045248031616,
      "learning_rate": 1.3901602909925204e-06,
      "loss": 0.0964,
      "num_input_tokens_seen": 67719744,
      "step": 35735
    },
    {
      "epoch": 5.830410310792071,
      "grad_norm": 0.6938957571983337,
      "learning_rate": 1.3869339562415373e-06,
      "loss": 0.1102,
      "num_input_tokens_seen": 67729696,
      "step": 35740
    },
    {
      "epoch": 5.831226038012889,
      "grad_norm": 0.8566824197769165,
      "learning_rate": 1.38371126292208e-06,
      "loss": 0.1272,
      "num_input_tokens_seen": 67739952,
      "step": 35745
    },
    {
      "epoch": 5.832041765233706,
      "grad_norm": 1.5757042169570923,
      "learning_rate": 1.3804922115311286e-06,
      "loss": 0.2891,
      "num_input_tokens_seen": 67749152,
      "step": 35750
    },
    {
      "epoch": 5.832857492454523,
      "grad_norm": 1.5089027881622314,
      "learning_rate": 1.3772768025650945e-06,
      "loss": 0.1306,
      "num_input_tokens_seen": 67757968,
      "step": 35755
    },
    {
      "epoch": 5.83367321967534,
      "grad_norm": 0.13951492309570312,
      "learning_rate": 1.3740650365198448e-06,
      "loss": 0.1142,
      "num_input_tokens_seen": 67768144,
      "step": 35760
    },
    {
      "epoch": 5.834488946896158,
      "grad_norm": 1.6247555017471313,
      "learning_rate": 1.3708569138906612e-06,
      "loss": 0.1456,
      "num_input_tokens_seen": 67777184,
      "step": 35765
    },
    {
      "epoch": 5.835304674116975,
      "grad_norm": 0.6736761927604675,
      "learning_rate": 1.367652435172287e-06,
      "loss": 0.122,
      "num_input_tokens_seen": 67787504,
      "step": 35770
    },
    {
      "epoch": 5.8361204013377925,
      "grad_norm": 1.1254348754882812,
      "learning_rate": 1.364451600858893e-06,
      "loss": 0.0814,
      "num_input_tokens_seen": 67797664,
      "step": 35775
    },
    {
      "epoch": 5.83693612855861,
      "grad_norm": 0.1253909319639206,
      "learning_rate": 1.3612544114440823e-06,
      "loss": 0.1088,
      "num_input_tokens_seen": 67808160,
      "step": 35780
    },
    {
      "epoch": 5.837751855779428,
      "grad_norm": 0.3003474175930023,
      "learning_rate": 1.3580608674209072e-06,
      "loss": 0.1299,
      "num_input_tokens_seen": 67817216,
      "step": 35785
    },
    {
      "epoch": 5.838567583000245,
      "grad_norm": 0.6545428037643433,
      "learning_rate": 1.3548709692818434e-06,
      "loss": 0.1221,
      "num_input_tokens_seen": 67824848,
      "step": 35790
    },
    {
      "epoch": 5.839383310221062,
      "grad_norm": 0.22705966234207153,
      "learning_rate": 1.3516847175188223e-06,
      "loss": 0.1345,
      "num_input_tokens_seen": 67835856,
      "step": 35795
    },
    {
      "epoch": 5.840199037441879,
      "grad_norm": 1.9419902563095093,
      "learning_rate": 1.348502112623204e-06,
      "loss": 0.1712,
      "num_input_tokens_seen": 67846112,
      "step": 35800
    },
    {
      "epoch": 5.840199037441879,
      "eval_loss": 0.14211009442806244,
      "eval_runtime": 133.2447,
      "eval_samples_per_second": 20.451,
      "eval_steps_per_second": 10.229,
      "num_input_tokens_seen": 67846112,
      "step": 35800
    },
    {
      "epoch": 5.841014764662697,
      "grad_norm": 1.3677408695220947,
      "learning_rate": 1.3453231550857787e-06,
      "loss": 0.2286,
      "num_input_tokens_seen": 67856032,
      "step": 35805
    },
    {
      "epoch": 5.841830491883514,
      "grad_norm": 0.24362735450267792,
      "learning_rate": 1.3421478453967878e-06,
      "loss": 0.0637,
      "num_input_tokens_seen": 67865568,
      "step": 35810
    },
    {
      "epoch": 5.842646219104331,
      "grad_norm": 1.9277774095535278,
      "learning_rate": 1.3389761840459065e-06,
      "loss": 0.2877,
      "num_input_tokens_seen": 67875200,
      "step": 35815
    },
    {
      "epoch": 5.8434619463251485,
      "grad_norm": 0.4487297236919403,
      "learning_rate": 1.3358081715222376e-06,
      "loss": 0.0657,
      "num_input_tokens_seen": 67884464,
      "step": 35820
    },
    {
      "epoch": 5.8442776735459665,
      "grad_norm": 1.5414799451828003,
      "learning_rate": 1.3326438083143295e-06,
      "loss": 0.1683,
      "num_input_tokens_seen": 67893248,
      "step": 35825
    },
    {
      "epoch": 5.845093400766784,
      "grad_norm": 0.5036855936050415,
      "learning_rate": 1.3294830949101723e-06,
      "loss": 0.0579,
      "num_input_tokens_seen": 67902384,
      "step": 35830
    },
    {
      "epoch": 5.845909127987601,
      "grad_norm": 0.2268694043159485,
      "learning_rate": 1.3263260317971815e-06,
      "loss": 0.0719,
      "num_input_tokens_seen": 67911600,
      "step": 35835
    },
    {
      "epoch": 5.846724855208418,
      "grad_norm": 0.23034372925758362,
      "learning_rate": 1.3231726194622208e-06,
      "loss": 0.1098,
      "num_input_tokens_seen": 67919792,
      "step": 35840
    },
    {
      "epoch": 5.847540582429236,
      "grad_norm": 0.06465382128953934,
      "learning_rate": 1.3200228583915814e-06,
      "loss": 0.0783,
      "num_input_tokens_seen": 67929008,
      "step": 35845
    },
    {
      "epoch": 5.848356309650053,
      "grad_norm": 0.28067708015441895,
      "learning_rate": 1.3168767490709971e-06,
      "loss": 0.103,
      "num_input_tokens_seen": 67937712,
      "step": 35850
    },
    {
      "epoch": 5.84917203687087,
      "grad_norm": 0.8177248239517212,
      "learning_rate": 1.3137342919856437e-06,
      "loss": 0.1171,
      "num_input_tokens_seen": 67947024,
      "step": 35855
    },
    {
      "epoch": 5.849987764091688,
      "grad_norm": 0.4949091076850891,
      "learning_rate": 1.310595487620117e-06,
      "loss": 0.1496,
      "num_input_tokens_seen": 67956480,
      "step": 35860
    },
    {
      "epoch": 5.850803491312505,
      "grad_norm": 0.5636773109436035,
      "learning_rate": 1.3074603364584715e-06,
      "loss": 0.0509,
      "num_input_tokens_seen": 67965232,
      "step": 35865
    },
    {
      "epoch": 5.8516192185333225,
      "grad_norm": 0.29875311255455017,
      "learning_rate": 1.3043288389841758e-06,
      "loss": 0.2162,
      "num_input_tokens_seen": 67975216,
      "step": 35870
    },
    {
      "epoch": 5.85243494575414,
      "grad_norm": 0.2661893665790558,
      "learning_rate": 1.3012009956801546e-06,
      "loss": 0.0648,
      "num_input_tokens_seen": 67983760,
      "step": 35875
    },
    {
      "epoch": 5.853250672974957,
      "grad_norm": 0.5365604162216187,
      "learning_rate": 1.2980768070287586e-06,
      "loss": 0.1507,
      "num_input_tokens_seen": 67994128,
      "step": 35880
    },
    {
      "epoch": 5.854066400195775,
      "grad_norm": 0.29185783863067627,
      "learning_rate": 1.2949562735117716e-06,
      "loss": 0.0676,
      "num_input_tokens_seen": 68002576,
      "step": 35885
    },
    {
      "epoch": 5.854882127416592,
      "grad_norm": 0.16661038994789124,
      "learning_rate": 1.291839395610428e-06,
      "loss": 0.1742,
      "num_input_tokens_seen": 68011536,
      "step": 35890
    },
    {
      "epoch": 5.855697854637409,
      "grad_norm": 1.4572639465332031,
      "learning_rate": 1.2887261738053852e-06,
      "loss": 0.2799,
      "num_input_tokens_seen": 68021264,
      "step": 35895
    },
    {
      "epoch": 5.856513581858227,
      "grad_norm": 0.9778198599815369,
      "learning_rate": 1.2856166085767396e-06,
      "loss": 0.1993,
      "num_input_tokens_seen": 68031360,
      "step": 35900
    },
    {
      "epoch": 5.857329309079044,
      "grad_norm": 1.3991918563842773,
      "learning_rate": 1.2825107004040272e-06,
      "loss": 0.1477,
      "num_input_tokens_seen": 68039408,
      "step": 35905
    },
    {
      "epoch": 5.858145036299861,
      "grad_norm": 2.4845798015594482,
      "learning_rate": 1.2794084497662146e-06,
      "loss": 0.1707,
      "num_input_tokens_seen": 68048640,
      "step": 35910
    },
    {
      "epoch": 5.858960763520678,
      "grad_norm": 0.24461348354816437,
      "learning_rate": 1.276309857141711e-06,
      "loss": 0.0988,
      "num_input_tokens_seen": 68058064,
      "step": 35915
    },
    {
      "epoch": 5.859776490741496,
      "grad_norm": 1.2547284364700317,
      "learning_rate": 1.273214923008359e-06,
      "loss": 0.2008,
      "num_input_tokens_seen": 68068016,
      "step": 35920
    },
    {
      "epoch": 5.8605922179623136,
      "grad_norm": 0.9941510558128357,
      "learning_rate": 1.2701236478434352e-06,
      "loss": 0.1597,
      "num_input_tokens_seen": 68076160,
      "step": 35925
    },
    {
      "epoch": 5.861407945183131,
      "grad_norm": 0.1033206433057785,
      "learning_rate": 1.2670360321236502e-06,
      "loss": 0.1707,
      "num_input_tokens_seen": 68086288,
      "step": 35930
    },
    {
      "epoch": 5.862223672403948,
      "grad_norm": 1.5568873882293701,
      "learning_rate": 1.2639520763251617e-06,
      "loss": 0.241,
      "num_input_tokens_seen": 68096464,
      "step": 35935
    },
    {
      "epoch": 5.863039399624766,
      "grad_norm": 1.4691822528839111,
      "learning_rate": 1.2608717809235448e-06,
      "loss": 0.1821,
      "num_input_tokens_seen": 68106368,
      "step": 35940
    },
    {
      "epoch": 5.863855126845583,
      "grad_norm": 0.14021846652030945,
      "learning_rate": 1.2577951463938282e-06,
      "loss": 0.0437,
      "num_input_tokens_seen": 68117472,
      "step": 35945
    },
    {
      "epoch": 5.8646708540664,
      "grad_norm": 0.20106235146522522,
      "learning_rate": 1.2547221732104569e-06,
      "loss": 0.1309,
      "num_input_tokens_seen": 68127696,
      "step": 35950
    },
    {
      "epoch": 5.865486581287217,
      "grad_norm": 0.3656187951564789,
      "learning_rate": 1.25165286184733e-06,
      "loss": 0.0316,
      "num_input_tokens_seen": 68136688,
      "step": 35955
    },
    {
      "epoch": 5.866302308508035,
      "grad_norm": 0.2184637039899826,
      "learning_rate": 1.248587212777777e-06,
      "loss": 0.0995,
      "num_input_tokens_seen": 68146096,
      "step": 35960
    },
    {
      "epoch": 5.867118035728852,
      "grad_norm": 0.7899336814880371,
      "learning_rate": 1.2455252264745532e-06,
      "loss": 0.1561,
      "num_input_tokens_seen": 68155408,
      "step": 35965
    },
    {
      "epoch": 5.8679337629496695,
      "grad_norm": 0.32278579473495483,
      "learning_rate": 1.2424669034098528e-06,
      "loss": 0.0775,
      "num_input_tokens_seen": 68165200,
      "step": 35970
    },
    {
      "epoch": 5.868749490170487,
      "grad_norm": 1.101022720336914,
      "learning_rate": 1.2394122440553185e-06,
      "loss": 0.0852,
      "num_input_tokens_seen": 68174528,
      "step": 35975
    },
    {
      "epoch": 5.869565217391305,
      "grad_norm": 0.323542982339859,
      "learning_rate": 1.2363612488820037e-06,
      "loss": 0.0285,
      "num_input_tokens_seen": 68183904,
      "step": 35980
    },
    {
      "epoch": 5.870380944612122,
      "grad_norm": 0.1353980004787445,
      "learning_rate": 1.2333139183604208e-06,
      "loss": 0.0768,
      "num_input_tokens_seen": 68193104,
      "step": 35985
    },
    {
      "epoch": 5.871196671832939,
      "grad_norm": 0.6081187725067139,
      "learning_rate": 1.2302702529604998e-06,
      "loss": 0.1147,
      "num_input_tokens_seen": 68202640,
      "step": 35990
    },
    {
      "epoch": 5.872012399053756,
      "grad_norm": 0.03703802078962326,
      "learning_rate": 1.227230253151615e-06,
      "loss": 0.1619,
      "num_input_tokens_seen": 68213024,
      "step": 35995
    },
    {
      "epoch": 5.872828126274574,
      "grad_norm": 1.5225571393966675,
      "learning_rate": 1.2241939194025748e-06,
      "loss": 0.0931,
      "num_input_tokens_seen": 68221552,
      "step": 36000
    },
    {
      "epoch": 5.872828126274574,
      "eval_loss": 0.14221158623695374,
      "eval_runtime": 133.1764,
      "eval_samples_per_second": 20.462,
      "eval_steps_per_second": 10.235,
      "num_input_tokens_seen": 68221552,
      "step": 36000
    },
    {
      "epoch": 5.873643853495391,
      "grad_norm": 0.36942625045776367,
      "learning_rate": 1.2211612521816156e-06,
      "loss": 0.0959,
      "num_input_tokens_seen": 68231312,
      "step": 36005
    },
    {
      "epoch": 5.874459580716208,
      "grad_norm": 0.4614212214946747,
      "learning_rate": 1.2181322519564137e-06,
      "loss": 0.1827,
      "num_input_tokens_seen": 68240944,
      "step": 36010
    },
    {
      "epoch": 5.8752753079370255,
      "grad_norm": 0.10453931987285614,
      "learning_rate": 1.2151069191940839e-06,
      "loss": 0.0613,
      "num_input_tokens_seen": 68251968,
      "step": 36015
    },
    {
      "epoch": 5.8760910351578435,
      "grad_norm": 0.07354512810707092,
      "learning_rate": 1.2120852543611644e-06,
      "loss": 0.055,
      "num_input_tokens_seen": 68262576,
      "step": 36020
    },
    {
      "epoch": 5.876906762378661,
      "grad_norm": 0.15717509388923645,
      "learning_rate": 1.2090672579236379e-06,
      "loss": 0.0776,
      "num_input_tokens_seen": 68272192,
      "step": 36025
    },
    {
      "epoch": 5.877722489599478,
      "grad_norm": 0.24543917179107666,
      "learning_rate": 1.2060529303469126e-06,
      "loss": 0.0683,
      "num_input_tokens_seen": 68280896,
      "step": 36030
    },
    {
      "epoch": 5.878538216820296,
      "grad_norm": 0.04235986992716789,
      "learning_rate": 1.2030422720958445e-06,
      "loss": 0.1324,
      "num_input_tokens_seen": 68289568,
      "step": 36035
    },
    {
      "epoch": 5.879353944041113,
      "grad_norm": 1.7475104331970215,
      "learning_rate": 1.200035283634704e-06,
      "loss": 0.2024,
      "num_input_tokens_seen": 68300544,
      "step": 36040
    },
    {
      "epoch": 5.88016967126193,
      "grad_norm": 0.4648788869380951,
      "learning_rate": 1.1970319654272144e-06,
      "loss": 0.1209,
      "num_input_tokens_seen": 68309584,
      "step": 36045
    },
    {
      "epoch": 5.880985398482747,
      "grad_norm": 0.6196531653404236,
      "learning_rate": 1.1940323179365192e-06,
      "loss": 0.1214,
      "num_input_tokens_seen": 68318240,
      "step": 36050
    },
    {
      "epoch": 5.881801125703564,
      "grad_norm": 1.1037003993988037,
      "learning_rate": 1.1910363416252095e-06,
      "loss": 0.1326,
      "num_input_tokens_seen": 68326912,
      "step": 36055
    },
    {
      "epoch": 5.882616852924382,
      "grad_norm": 0.3508962392807007,
      "learning_rate": 1.1880440369552964e-06,
      "loss": 0.1766,
      "num_input_tokens_seen": 68336528,
      "step": 36060
    },
    {
      "epoch": 5.883432580145199,
      "grad_norm": 0.5561382174491882,
      "learning_rate": 1.1850554043882328e-06,
      "loss": 0.1414,
      "num_input_tokens_seen": 68346464,
      "step": 36065
    },
    {
      "epoch": 5.884248307366017,
      "grad_norm": 0.39719054102897644,
      "learning_rate": 1.1820704443849028e-06,
      "loss": 0.095,
      "num_input_tokens_seen": 68356448,
      "step": 36070
    },
    {
      "epoch": 5.885064034586835,
      "grad_norm": 2.1377274990081787,
      "learning_rate": 1.1790891574056219e-06,
      "loss": 0.1984,
      "num_input_tokens_seen": 68366256,
      "step": 36075
    },
    {
      "epoch": 5.885879761807652,
      "grad_norm": 0.15889696776866913,
      "learning_rate": 1.1761115439101523e-06,
      "loss": 0.095,
      "num_input_tokens_seen": 68375936,
      "step": 36080
    },
    {
      "epoch": 5.886695489028469,
      "grad_norm": 0.22034792602062225,
      "learning_rate": 1.1731376043576659e-06,
      "loss": 0.0447,
      "num_input_tokens_seen": 68385792,
      "step": 36085
    },
    {
      "epoch": 5.887511216249286,
      "grad_norm": 0.43797096610069275,
      "learning_rate": 1.1701673392067875e-06,
      "loss": 0.0944,
      "num_input_tokens_seen": 68394720,
      "step": 36090
    },
    {
      "epoch": 5.888326943470103,
      "grad_norm": 1.1965473890304565,
      "learning_rate": 1.1672007489155757e-06,
      "loss": 0.0891,
      "num_input_tokens_seen": 68404304,
      "step": 36095
    },
    {
      "epoch": 5.889142670690921,
      "grad_norm": 0.8892207145690918,
      "learning_rate": 1.164237833941506e-06,
      "loss": 0.261,
      "num_input_tokens_seen": 68414560,
      "step": 36100
    },
    {
      "epoch": 5.889958397911738,
      "grad_norm": 0.1343775987625122,
      "learning_rate": 1.1612785947415022e-06,
      "loss": 0.1331,
      "num_input_tokens_seen": 68424400,
      "step": 36105
    },
    {
      "epoch": 5.890774125132555,
      "grad_norm": 0.39655929803848267,
      "learning_rate": 1.1583230317719185e-06,
      "loss": 0.1306,
      "num_input_tokens_seen": 68434496,
      "step": 36110
    },
    {
      "epoch": 5.891589852353373,
      "grad_norm": 0.029962297528982162,
      "learning_rate": 1.1553711454885318e-06,
      "loss": 0.0676,
      "num_input_tokens_seen": 68443664,
      "step": 36115
    },
    {
      "epoch": 5.8924055795741905,
      "grad_norm": 0.7458627820014954,
      "learning_rate": 1.152422936346567e-06,
      "loss": 0.0567,
      "num_input_tokens_seen": 68452864,
      "step": 36120
    },
    {
      "epoch": 5.893221306795008,
      "grad_norm": 1.1259838342666626,
      "learning_rate": 1.1494784048006718e-06,
      "loss": 0.1062,
      "num_input_tokens_seen": 68462224,
      "step": 36125
    },
    {
      "epoch": 5.894037034015825,
      "grad_norm": 0.15443481504917145,
      "learning_rate": 1.1465375513049326e-06,
      "loss": 0.0956,
      "num_input_tokens_seen": 68471632,
      "step": 36130
    },
    {
      "epoch": 5.894852761236643,
      "grad_norm": 1.804677963256836,
      "learning_rate": 1.1436003763128616e-06,
      "loss": 0.2334,
      "num_input_tokens_seen": 68481696,
      "step": 36135
    },
    {
      "epoch": 5.89566848845746,
      "grad_norm": 1.1037490367889404,
      "learning_rate": 1.1406668802774106e-06,
      "loss": 0.1773,
      "num_input_tokens_seen": 68491328,
      "step": 36140
    },
    {
      "epoch": 5.896484215678277,
      "grad_norm": 0.05308882147073746,
      "learning_rate": 1.137737063650965e-06,
      "loss": 0.083,
      "num_input_tokens_seen": 68500096,
      "step": 36145
    },
    {
      "epoch": 5.897299942899094,
      "grad_norm": 1.21183443069458,
      "learning_rate": 1.1348109268853323e-06,
      "loss": 0.2512,
      "num_input_tokens_seen": 68510992,
      "step": 36150
    },
    {
      "epoch": 5.898115670119912,
      "grad_norm": 0.0419083908200264,
      "learning_rate": 1.1318884704317634e-06,
      "loss": 0.0667,
      "num_input_tokens_seen": 68521328,
      "step": 36155
    },
    {
      "epoch": 5.898931397340729,
      "grad_norm": 0.1488085240125656,
      "learning_rate": 1.1289696947409417e-06,
      "loss": 0.0849,
      "num_input_tokens_seen": 68531968,
      "step": 36160
    },
    {
      "epoch": 5.8997471245615465,
      "grad_norm": 1.4065437316894531,
      "learning_rate": 1.126054600262974e-06,
      "loss": 0.2207,
      "num_input_tokens_seen": 68540864,
      "step": 36165
    },
    {
      "epoch": 5.900562851782364,
      "grad_norm": 0.12797172367572784,
      "learning_rate": 1.1231431874474064e-06,
      "loss": 0.056,
      "num_input_tokens_seen": 68549952,
      "step": 36170
    },
    {
      "epoch": 5.901378579003182,
      "grad_norm": 0.049173470586538315,
      "learning_rate": 1.12023545674321e-06,
      "loss": 0.1123,
      "num_input_tokens_seen": 68558960,
      "step": 36175
    },
    {
      "epoch": 5.902194306223999,
      "grad_norm": 1.3232624530792236,
      "learning_rate": 1.117331408598804e-06,
      "loss": 0.2507,
      "num_input_tokens_seen": 68567808,
      "step": 36180
    },
    {
      "epoch": 5.903010033444816,
      "grad_norm": 0.8178847432136536,
      "learning_rate": 1.1144310434620191e-06,
      "loss": 0.1328,
      "num_input_tokens_seen": 68578064,
      "step": 36185
    },
    {
      "epoch": 5.903825760665633,
      "grad_norm": 0.07505334913730621,
      "learning_rate": 1.1115343617801365e-06,
      "loss": 0.0594,
      "num_input_tokens_seen": 68587920,
      "step": 36190
    },
    {
      "epoch": 5.904641487886451,
      "grad_norm": 0.2563749849796295,
      "learning_rate": 1.1086413639998515e-06,
      "loss": 0.2369,
      "num_input_tokens_seen": 68597840,
      "step": 36195
    },
    {
      "epoch": 5.905457215107268,
      "grad_norm": 0.8918622732162476,
      "learning_rate": 1.1057520505673103e-06,
      "loss": 0.1396,
      "num_input_tokens_seen": 68606416,
      "step": 36200
    },
    {
      "epoch": 5.905457215107268,
      "eval_loss": 0.14184993505477905,
      "eval_runtime": 133.2328,
      "eval_samples_per_second": 20.453,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 68606416,
      "step": 36200
    },
    {
      "epoch": 5.906272942328085,
      "grad_norm": 0.1539992243051529,
      "learning_rate": 1.1028664219280727e-06,
      "loss": 0.1011,
      "num_input_tokens_seen": 68616224,
      "step": 36205
    },
    {
      "epoch": 5.907088669548903,
      "grad_norm": 1.338557481765747,
      "learning_rate": 1.0999844785271468e-06,
      "loss": 0.1247,
      "num_input_tokens_seen": 68625344,
      "step": 36210
    },
    {
      "epoch": 5.9079043967697205,
      "grad_norm": 0.5922865867614746,
      "learning_rate": 1.097106220808955e-06,
      "loss": 0.3894,
      "num_input_tokens_seen": 68633680,
      "step": 36215
    },
    {
      "epoch": 5.908720123990538,
      "grad_norm": 1.8833953142166138,
      "learning_rate": 1.0942316492173698e-06,
      "loss": 0.1432,
      "num_input_tokens_seen": 68642880,
      "step": 36220
    },
    {
      "epoch": 5.909535851211355,
      "grad_norm": 1.6820405721664429,
      "learning_rate": 1.0913607641956841e-06,
      "loss": 0.246,
      "num_input_tokens_seen": 68652512,
      "step": 36225
    },
    {
      "epoch": 5.910351578432172,
      "grad_norm": 0.39517346024513245,
      "learning_rate": 1.0884935661866213e-06,
      "loss": 0.0997,
      "num_input_tokens_seen": 68661776,
      "step": 36230
    },
    {
      "epoch": 5.91116730565299,
      "grad_norm": 1.0152504444122314,
      "learning_rate": 1.0856300556323418e-06,
      "loss": 0.1052,
      "num_input_tokens_seen": 68671328,
      "step": 36235
    },
    {
      "epoch": 5.911983032873807,
      "grad_norm": 1.1432229280471802,
      "learning_rate": 1.0827702329744365e-06,
      "loss": 0.1191,
      "num_input_tokens_seen": 68679888,
      "step": 36240
    },
    {
      "epoch": 5.912798760094624,
      "grad_norm": 0.6203641891479492,
      "learning_rate": 1.0799140986539197e-06,
      "loss": 0.0479,
      "num_input_tokens_seen": 68690496,
      "step": 36245
    },
    {
      "epoch": 5.913614487315442,
      "grad_norm": 0.7117956876754761,
      "learning_rate": 1.0770616531112526e-06,
      "loss": 0.277,
      "num_input_tokens_seen": 68699728,
      "step": 36250
    },
    {
      "epoch": 5.914430214536259,
      "grad_norm": 0.03397209942340851,
      "learning_rate": 1.0742128967863085e-06,
      "loss": 0.1447,
      "num_input_tokens_seen": 68708160,
      "step": 36255
    },
    {
      "epoch": 5.915245941757076,
      "grad_norm": 0.03579224646091461,
      "learning_rate": 1.071367830118411e-06,
      "loss": 0.1519,
      "num_input_tokens_seen": 68717904,
      "step": 36260
    },
    {
      "epoch": 5.9160616689778935,
      "grad_norm": 0.2522938847541809,
      "learning_rate": 1.068526453546298e-06,
      "loss": 0.0711,
      "num_input_tokens_seen": 68728048,
      "step": 36265
    },
    {
      "epoch": 5.916877396198711,
      "grad_norm": 0.32708805799484253,
      "learning_rate": 1.0656887675081467e-06,
      "loss": 0.17,
      "num_input_tokens_seen": 68737984,
      "step": 36270
    },
    {
      "epoch": 5.917693123419529,
      "grad_norm": 0.05029922351241112,
      "learning_rate": 1.0628547724415628e-06,
      "loss": 0.0614,
      "num_input_tokens_seen": 68746096,
      "step": 36275
    },
    {
      "epoch": 5.918508850640346,
      "grad_norm": 0.5292062163352966,
      "learning_rate": 1.0600244687835881e-06,
      "loss": 0.0967,
      "num_input_tokens_seen": 68755168,
      "step": 36280
    },
    {
      "epoch": 5.919324577861163,
      "grad_norm": 0.7771658897399902,
      "learning_rate": 1.0571978569706876e-06,
      "loss": 0.0343,
      "num_input_tokens_seen": 68764320,
      "step": 36285
    },
    {
      "epoch": 5.920140305081981,
      "grad_norm": 1.7662197351455688,
      "learning_rate": 1.0543749374387652e-06,
      "loss": 0.2174,
      "num_input_tokens_seen": 68773024,
      "step": 36290
    },
    {
      "epoch": 5.920956032302798,
      "grad_norm": 0.06647103279829025,
      "learning_rate": 1.051555710623142e-06,
      "loss": 0.1215,
      "num_input_tokens_seen": 68782912,
      "step": 36295
    },
    {
      "epoch": 5.921771759523615,
      "grad_norm": 0.3900095820426941,
      "learning_rate": 1.0487401769585847e-06,
      "loss": 0.1547,
      "num_input_tokens_seen": 68793440,
      "step": 36300
    },
    {
      "epoch": 5.922587486744432,
      "grad_norm": 1.7279103994369507,
      "learning_rate": 1.0459283368792845e-06,
      "loss": 0.127,
      "num_input_tokens_seen": 68802224,
      "step": 36305
    },
    {
      "epoch": 5.92340321396525,
      "grad_norm": 1.3864160776138306,
      "learning_rate": 1.043120190818858e-06,
      "loss": 0.0852,
      "num_input_tokens_seen": 68812112,
      "step": 36310
    },
    {
      "epoch": 5.9242189411860675,
      "grad_norm": 1.683591604232788,
      "learning_rate": 1.0403157392103596e-06,
      "loss": 0.2706,
      "num_input_tokens_seen": 68822128,
      "step": 36315
    },
    {
      "epoch": 5.925034668406885,
      "grad_norm": 0.34796154499053955,
      "learning_rate": 1.0375149824862735e-06,
      "loss": 0.1578,
      "num_input_tokens_seen": 68831936,
      "step": 36320
    },
    {
      "epoch": 5.925850395627702,
      "grad_norm": 0.4975568652153015,
      "learning_rate": 1.034717921078507e-06,
      "loss": 0.3276,
      "num_input_tokens_seen": 68841872,
      "step": 36325
    },
    {
      "epoch": 5.92666612284852,
      "grad_norm": 0.3390221893787384,
      "learning_rate": 1.0319245554184009e-06,
      "loss": 0.111,
      "num_input_tokens_seen": 68850256,
      "step": 36330
    },
    {
      "epoch": 5.927481850069337,
      "grad_norm": 0.10504735261201859,
      "learning_rate": 1.0291348859367361e-06,
      "loss": 0.1096,
      "num_input_tokens_seen": 68860064,
      "step": 36335
    },
    {
      "epoch": 5.928297577290154,
      "grad_norm": 1.567978858947754,
      "learning_rate": 1.0263489130637016e-06,
      "loss": 0.0655,
      "num_input_tokens_seen": 68870016,
      "step": 36340
    },
    {
      "epoch": 5.929113304510971,
      "grad_norm": 0.8531976938247681,
      "learning_rate": 1.0235666372289427e-06,
      "loss": 0.1294,
      "num_input_tokens_seen": 68879248,
      "step": 36345
    },
    {
      "epoch": 5.929929031731789,
      "grad_norm": 0.24569779634475708,
      "learning_rate": 1.0207880588615076e-06,
      "loss": 0.0807,
      "num_input_tokens_seen": 68887696,
      "step": 36350
    },
    {
      "epoch": 5.930744758952606,
      "grad_norm": 0.33762866258621216,
      "learning_rate": 1.0180131783898984e-06,
      "loss": 0.0505,
      "num_input_tokens_seen": 68897712,
      "step": 36355
    },
    {
      "epoch": 5.9315604861734235,
      "grad_norm": 0.07262604683637619,
      "learning_rate": 1.0152419962420362e-06,
      "loss": 0.1943,
      "num_input_tokens_seen": 68907344,
      "step": 36360
    },
    {
      "epoch": 5.932376213394241,
      "grad_norm": 1.3201273679733276,
      "learning_rate": 1.0124745128452685e-06,
      "loss": 0.1398,
      "num_input_tokens_seen": 68916624,
      "step": 36365
    },
    {
      "epoch": 5.933191940615059,
      "grad_norm": 0.6301760673522949,
      "learning_rate": 1.0097107286263758e-06,
      "loss": 0.1009,
      "num_input_tokens_seen": 68925376,
      "step": 36370
    },
    {
      "epoch": 5.934007667835876,
      "grad_norm": 0.5278180837631226,
      "learning_rate": 1.00695064401157e-06,
      "loss": 0.2616,
      "num_input_tokens_seen": 68933392,
      "step": 36375
    },
    {
      "epoch": 5.934823395056693,
      "grad_norm": 0.9363027215003967,
      "learning_rate": 1.0041942594264886e-06,
      "loss": 0.1935,
      "num_input_tokens_seen": 68942576,
      "step": 36380
    },
    {
      "epoch": 5.935639122277511,
      "grad_norm": 0.15462736785411835,
      "learning_rate": 1.001441575296208e-06,
      "loss": 0.0843,
      "num_input_tokens_seen": 68951008,
      "step": 36385
    },
    {
      "epoch": 5.936454849498328,
      "grad_norm": 0.3395603597164154,
      "learning_rate": 9.986925920452139e-07,
      "loss": 0.126,
      "num_input_tokens_seen": 68961248,
      "step": 36390
    },
    {
      "epoch": 5.937270576719145,
      "grad_norm": 1.2184932231903076,
      "learning_rate": 9.959473100974475e-07,
      "loss": 0.111,
      "num_input_tokens_seen": 68970944,
      "step": 36395
    },
    {
      "epoch": 5.938086303939962,
      "grad_norm": 1.2687824964523315,
      "learning_rate": 9.932057298762564e-07,
      "loss": 0.1184,
      "num_input_tokens_seen": 68980176,
      "step": 36400
    },
    {
      "epoch": 5.938086303939962,
      "eval_loss": 0.14270642399787903,
      "eval_runtime": 133.3069,
      "eval_samples_per_second": 20.442,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 68980176,
      "step": 36400
    },
    {
      "epoch": 5.938902031160779,
      "grad_norm": 0.44920283555984497,
      "learning_rate": 9.90467851804433e-07,
      "loss": 0.0554,
      "num_input_tokens_seen": 68990016,
      "step": 36405
    },
    {
      "epoch": 5.939717758381597,
      "grad_norm": 1.413880705833435,
      "learning_rate": 9.877336763041895e-07,
      "loss": 0.1768,
      "num_input_tokens_seen": 68998784,
      "step": 36410
    },
    {
      "epoch": 5.940533485602415,
      "grad_norm": 0.7748748660087585,
      "learning_rate": 9.850032037971662e-07,
      "loss": 0.0733,
      "num_input_tokens_seen": 69008000,
      "step": 36415
    },
    {
      "epoch": 5.941349212823232,
      "grad_norm": 0.9322550892829895,
      "learning_rate": 9.822764347044406e-07,
      "loss": 0.2078,
      "num_input_tokens_seen": 69017152,
      "step": 36420
    },
    {
      "epoch": 5.94216494004405,
      "grad_norm": 0.1005963459610939,
      "learning_rate": 9.795533694465175e-07,
      "loss": 0.0898,
      "num_input_tokens_seen": 69027408,
      "step": 36425
    },
    {
      "epoch": 5.942980667264867,
      "grad_norm": 0.3174126446247101,
      "learning_rate": 9.768340084433197e-07,
      "loss": 0.1356,
      "num_input_tokens_seen": 69034384,
      "step": 36430
    },
    {
      "epoch": 5.943796394485684,
      "grad_norm": 0.40159156918525696,
      "learning_rate": 9.741183521142143e-07,
      "loss": 0.0537,
      "num_input_tokens_seen": 69042576,
      "step": 36435
    },
    {
      "epoch": 5.944612121706501,
      "grad_norm": 0.22693291306495667,
      "learning_rate": 9.714064008779889e-07,
      "loss": 0.137,
      "num_input_tokens_seen": 69051616,
      "step": 36440
    },
    {
      "epoch": 5.945427848927318,
      "grad_norm": 1.5203348398208618,
      "learning_rate": 9.686981551528584e-07,
      "loss": 0.1434,
      "num_input_tokens_seen": 69061824,
      "step": 36445
    },
    {
      "epoch": 5.946243576148136,
      "grad_norm": 0.823031485080719,
      "learning_rate": 9.65993615356467e-07,
      "loss": 0.1412,
      "num_input_tokens_seen": 69072128,
      "step": 36450
    },
    {
      "epoch": 5.947059303368953,
      "grad_norm": 0.5715826749801636,
      "learning_rate": 9.632927819058917e-07,
      "loss": 0.0657,
      "num_input_tokens_seen": 69080464,
      "step": 36455
    },
    {
      "epoch": 5.9478750305897705,
      "grad_norm": 2.9601588249206543,
      "learning_rate": 9.605956552176305e-07,
      "loss": 0.1525,
      "num_input_tokens_seen": 69089936,
      "step": 36460
    },
    {
      "epoch": 5.9486907578105885,
      "grad_norm": 0.9844962358474731,
      "learning_rate": 9.579022357076223e-07,
      "loss": 0.091,
      "num_input_tokens_seen": 69098896,
      "step": 36465
    },
    {
      "epoch": 5.949506485031406,
      "grad_norm": 2.246267318725586,
      "learning_rate": 9.552125237912158e-07,
      "loss": 0.1814,
      "num_input_tokens_seen": 69107584,
      "step": 36470
    },
    {
      "epoch": 5.950322212252223,
      "grad_norm": 0.13931936025619507,
      "learning_rate": 9.525265198832096e-07,
      "loss": 0.0469,
      "num_input_tokens_seen": 69117136,
      "step": 36475
    },
    {
      "epoch": 5.95113793947304,
      "grad_norm": 0.16871780157089233,
      "learning_rate": 9.498442243978112e-07,
      "loss": 0.0998,
      "num_input_tokens_seen": 69126224,
      "step": 36480
    },
    {
      "epoch": 5.951953666693858,
      "grad_norm": 0.20908576250076294,
      "learning_rate": 9.471656377486649e-07,
      "loss": 0.0955,
      "num_input_tokens_seen": 69135984,
      "step": 36485
    },
    {
      "epoch": 5.952769393914675,
      "grad_norm": 0.18757565319538116,
      "learning_rate": 9.444907603488456e-07,
      "loss": 0.1886,
      "num_input_tokens_seen": 69145712,
      "step": 36490
    },
    {
      "epoch": 5.953585121135492,
      "grad_norm": 1.1987380981445312,
      "learning_rate": 9.418195926108514e-07,
      "loss": 0.1803,
      "num_input_tokens_seen": 69154848,
      "step": 36495
    },
    {
      "epoch": 5.954400848356309,
      "grad_norm": 0.02659996598958969,
      "learning_rate": 9.391521349466053e-07,
      "loss": 0.0744,
      "num_input_tokens_seen": 69163152,
      "step": 36500
    },
    {
      "epoch": 5.955216575577127,
      "grad_norm": 0.4845864772796631,
      "learning_rate": 9.364883877674758e-07,
      "loss": 0.1771,
      "num_input_tokens_seen": 69172592,
      "step": 36505
    },
    {
      "epoch": 5.9560323027979445,
      "grad_norm": 1.35465669631958,
      "learning_rate": 9.33828351484231e-07,
      "loss": 0.0875,
      "num_input_tokens_seen": 69181936,
      "step": 36510
    },
    {
      "epoch": 5.956848030018762,
      "grad_norm": 1.4284831285476685,
      "learning_rate": 9.311720265070906e-07,
      "loss": 0.1423,
      "num_input_tokens_seen": 69190032,
      "step": 36515
    },
    {
      "epoch": 5.957663757239579,
      "grad_norm": 0.6964465379714966,
      "learning_rate": 9.285194132456931e-07,
      "loss": 0.226,
      "num_input_tokens_seen": 69200048,
      "step": 36520
    },
    {
      "epoch": 5.958479484460397,
      "grad_norm": 0.17277951538562775,
      "learning_rate": 9.258705121091032e-07,
      "loss": 0.1297,
      "num_input_tokens_seen": 69208144,
      "step": 36525
    },
    {
      "epoch": 5.959295211681214,
      "grad_norm": 0.06481810659170151,
      "learning_rate": 9.232253235058136e-07,
      "loss": 0.1025,
      "num_input_tokens_seen": 69217840,
      "step": 36530
    },
    {
      "epoch": 5.960110938902031,
      "grad_norm": 0.8867670893669128,
      "learning_rate": 9.205838478437478e-07,
      "loss": 0.102,
      "num_input_tokens_seen": 69228544,
      "step": 36535
    },
    {
      "epoch": 5.960926666122848,
      "grad_norm": 0.49166324734687805,
      "learning_rate": 9.179460855302524e-07,
      "loss": 0.0796,
      "num_input_tokens_seen": 69238016,
      "step": 36540
    },
    {
      "epoch": 5.961742393343666,
      "grad_norm": 0.9727489948272705,
      "learning_rate": 9.153120369721046e-07,
      "loss": 0.1431,
      "num_input_tokens_seen": 69247552,
      "step": 36545
    },
    {
      "epoch": 5.962558120564483,
      "grad_norm": 0.7966281175613403,
      "learning_rate": 9.126817025755103e-07,
      "loss": 0.1516,
      "num_input_tokens_seen": 69255824,
      "step": 36550
    },
    {
      "epoch": 5.9633738477853,
      "grad_norm": 0.03961006551980972,
      "learning_rate": 9.100550827460947e-07,
      "loss": 0.0815,
      "num_input_tokens_seen": 69266032,
      "step": 36555
    },
    {
      "epoch": 5.964189575006118,
      "grad_norm": 1.0935490131378174,
      "learning_rate": 9.0743217788892e-07,
      "loss": 0.1294,
      "num_input_tokens_seen": 69274080,
      "step": 36560
    },
    {
      "epoch": 5.965005302226936,
      "grad_norm": 1.2834237813949585,
      "learning_rate": 9.048129884084683e-07,
      "loss": 0.062,
      "num_input_tokens_seen": 69282880,
      "step": 36565
    },
    {
      "epoch": 5.965821029447753,
      "grad_norm": 0.2499813288450241,
      "learning_rate": 9.021975147086553e-07,
      "loss": 0.135,
      "num_input_tokens_seen": 69292832,
      "step": 36570
    },
    {
      "epoch": 5.96663675666857,
      "grad_norm": 0.537185788154602,
      "learning_rate": 8.995857571928141e-07,
      "loss": 0.1043,
      "num_input_tokens_seen": 69303024,
      "step": 36575
    },
    {
      "epoch": 5.967452483889387,
      "grad_norm": 0.07704123854637146,
      "learning_rate": 8.969777162637139e-07,
      "loss": 0.0408,
      "num_input_tokens_seen": 69312544,
      "step": 36580
    },
    {
      "epoch": 5.968268211110205,
      "grad_norm": 0.7541214823722839,
      "learning_rate": 8.943733923235525e-07,
      "loss": 0.1567,
      "num_input_tokens_seen": 69321664,
      "step": 36585
    },
    {
      "epoch": 5.969083938331022,
      "grad_norm": 0.13128726184368134,
      "learning_rate": 8.917727857739394e-07,
      "loss": 0.2035,
      "num_input_tokens_seen": 69330640,
      "step": 36590
    },
    {
      "epoch": 5.969899665551839,
      "grad_norm": 0.12864691019058228,
      "learning_rate": 8.891758970159258e-07,
      "loss": 0.0898,
      "num_input_tokens_seen": 69340176,
      "step": 36595
    },
    {
      "epoch": 5.970715392772657,
      "grad_norm": 0.1517655998468399,
      "learning_rate": 8.86582726449986e-07,
      "loss": 0.0667,
      "num_input_tokens_seen": 69349984,
      "step": 36600
    },
    {
      "epoch": 5.970715392772657,
      "eval_loss": 0.14254936575889587,
      "eval_runtime": 133.2488,
      "eval_samples_per_second": 20.45,
      "eval_steps_per_second": 10.229,
      "num_input_tokens_seen": 69349984,
      "step": 36600
    },
    {
      "epoch": 5.971531119993474,
      "grad_norm": 0.5351443290710449,
      "learning_rate": 8.839932744760165e-07,
      "loss": 0.0296,
      "num_input_tokens_seen": 69359760,
      "step": 36605
    },
    {
      "epoch": 5.9723468472142915,
      "grad_norm": 0.34032508730888367,
      "learning_rate": 8.814075414933482e-07,
      "loss": 0.0679,
      "num_input_tokens_seen": 69368864,
      "step": 36610
    },
    {
      "epoch": 5.973162574435109,
      "grad_norm": 1.0774825811386108,
      "learning_rate": 8.788255279007257e-07,
      "loss": 0.2241,
      "num_input_tokens_seen": 69378224,
      "step": 36615
    },
    {
      "epoch": 5.973978301655926,
      "grad_norm": 0.21637648344039917,
      "learning_rate": 8.762472340963362e-07,
      "loss": 0.0963,
      "num_input_tokens_seen": 69387984,
      "step": 36620
    },
    {
      "epoch": 5.974794028876744,
      "grad_norm": 0.8285349607467651,
      "learning_rate": 8.736726604777811e-07,
      "loss": 0.1178,
      "num_input_tokens_seen": 69396480,
      "step": 36625
    },
    {
      "epoch": 5.975609756097561,
      "grad_norm": 1.3228259086608887,
      "learning_rate": 8.711018074420901e-07,
      "loss": 0.287,
      "num_input_tokens_seen": 69406688,
      "step": 36630
    },
    {
      "epoch": 5.976425483318378,
      "grad_norm": 1.869666337966919,
      "learning_rate": 8.685346753857209e-07,
      "loss": 0.1312,
      "num_input_tokens_seen": 69416896,
      "step": 36635
    },
    {
      "epoch": 5.977241210539196,
      "grad_norm": 0.4760694205760956,
      "learning_rate": 8.659712647045654e-07,
      "loss": 0.1452,
      "num_input_tokens_seen": 69426400,
      "step": 36640
    },
    {
      "epoch": 5.978056937760013,
      "grad_norm": 2.6901583671569824,
      "learning_rate": 8.634115757939209e-07,
      "loss": 0.2234,
      "num_input_tokens_seen": 69435776,
      "step": 36645
    },
    {
      "epoch": 5.97887266498083,
      "grad_norm": 0.9989168643951416,
      "learning_rate": 8.608556090485387e-07,
      "loss": 0.1757,
      "num_input_tokens_seen": 69445152,
      "step": 36650
    },
    {
      "epoch": 5.9796883922016475,
      "grad_norm": 1.0476006269454956,
      "learning_rate": 8.583033648625671e-07,
      "loss": 0.1131,
      "num_input_tokens_seen": 69455232,
      "step": 36655
    },
    {
      "epoch": 5.9805041194224655,
      "grad_norm": 0.03667992353439331,
      "learning_rate": 8.557548436295998e-07,
      "loss": 0.2998,
      "num_input_tokens_seen": 69465280,
      "step": 36660
    },
    {
      "epoch": 5.981319846643283,
      "grad_norm": 1.741991400718689,
      "learning_rate": 8.532100457426556e-07,
      "loss": 0.0956,
      "num_input_tokens_seen": 69474704,
      "step": 36665
    },
    {
      "epoch": 5.9821355738641,
      "grad_norm": 1.0259528160095215,
      "learning_rate": 8.506689715941679e-07,
      "loss": 0.1251,
      "num_input_tokens_seen": 69482848,
      "step": 36670
    },
    {
      "epoch": 5.982951301084917,
      "grad_norm": 0.09591516852378845,
      "learning_rate": 8.481316215760011e-07,
      "loss": 0.0817,
      "num_input_tokens_seen": 69493360,
      "step": 36675
    },
    {
      "epoch": 5.983767028305735,
      "grad_norm": 0.09207141399383545,
      "learning_rate": 8.455979960794558e-07,
      "loss": 0.1099,
      "num_input_tokens_seen": 69501872,
      "step": 36680
    },
    {
      "epoch": 5.984582755526552,
      "grad_norm": 0.6448224186897278,
      "learning_rate": 8.430680954952364e-07,
      "loss": 0.0803,
      "num_input_tokens_seen": 69511648,
      "step": 36685
    },
    {
      "epoch": 5.985398482747369,
      "grad_norm": 0.0907505676150322,
      "learning_rate": 8.405419202134974e-07,
      "loss": 0.2128,
      "num_input_tokens_seen": 69521456,
      "step": 36690
    },
    {
      "epoch": 5.986214209968186,
      "grad_norm": 2.0447230339050293,
      "learning_rate": 8.380194706237993e-07,
      "loss": 0.2733,
      "num_input_tokens_seen": 69531648,
      "step": 36695
    },
    {
      "epoch": 5.987029937189004,
      "grad_norm": 0.15340028703212738,
      "learning_rate": 8.355007471151366e-07,
      "loss": 0.1171,
      "num_input_tokens_seen": 69541120,
      "step": 36700
    },
    {
      "epoch": 5.9878456644098215,
      "grad_norm": 0.13314442336559296,
      "learning_rate": 8.329857500759292e-07,
      "loss": 0.1271,
      "num_input_tokens_seen": 69551088,
      "step": 36705
    },
    {
      "epoch": 5.988661391630639,
      "grad_norm": 0.029937807470560074,
      "learning_rate": 8.304744798940194e-07,
      "loss": 0.1304,
      "num_input_tokens_seen": 69560784,
      "step": 36710
    },
    {
      "epoch": 5.989477118851456,
      "grad_norm": 1.2932504415512085,
      "learning_rate": 8.279669369566756e-07,
      "loss": 0.3135,
      "num_input_tokens_seen": 69569664,
      "step": 36715
    },
    {
      "epoch": 5.990292846072274,
      "grad_norm": 0.40829670429229736,
      "learning_rate": 8.254631216505993e-07,
      "loss": 0.1474,
      "num_input_tokens_seen": 69578224,
      "step": 36720
    },
    {
      "epoch": 5.991108573293091,
      "grad_norm": 1.2824385166168213,
      "learning_rate": 8.229630343619038e-07,
      "loss": 0.145,
      "num_input_tokens_seen": 69587696,
      "step": 36725
    },
    {
      "epoch": 5.991924300513908,
      "grad_norm": 0.25694558024406433,
      "learning_rate": 8.204666754761392e-07,
      "loss": 0.0679,
      "num_input_tokens_seen": 69597904,
      "step": 36730
    },
    {
      "epoch": 5.992740027734725,
      "grad_norm": 0.6342456340789795,
      "learning_rate": 8.179740453782669e-07,
      "loss": 0.1352,
      "num_input_tokens_seen": 69607072,
      "step": 36735
    },
    {
      "epoch": 5.993555754955543,
      "grad_norm": 0.8151386380195618,
      "learning_rate": 8.154851444526907e-07,
      "loss": 0.2643,
      "num_input_tokens_seen": 69617728,
      "step": 36740
    },
    {
      "epoch": 5.99437148217636,
      "grad_norm": 0.314422070980072,
      "learning_rate": 8.129999730832283e-07,
      "loss": 0.0591,
      "num_input_tokens_seen": 69626864,
      "step": 36745
    },
    {
      "epoch": 5.995187209397177,
      "grad_norm": 0.10943777114152908,
      "learning_rate": 8.105185316531178e-07,
      "loss": 0.0291,
      "num_input_tokens_seen": 69635024,
      "step": 36750
    },
    {
      "epoch": 5.9960029366179945,
      "grad_norm": 0.8410003781318665,
      "learning_rate": 8.08040820545039e-07,
      "loss": 0.1243,
      "num_input_tokens_seen": 69644976,
      "step": 36755
    },
    {
      "epoch": 5.996818663838813,
      "grad_norm": 0.4331877529621124,
      "learning_rate": 8.055668401410782e-07,
      "loss": 0.1453,
      "num_input_tokens_seen": 69654608,
      "step": 36760
    },
    {
      "epoch": 5.99763439105963,
      "grad_norm": 0.26301077008247375,
      "learning_rate": 8.030965908227578e-07,
      "loss": 0.1186,
      "num_input_tokens_seen": 69663904,
      "step": 36765
    },
    {
      "epoch": 5.998450118280447,
      "grad_norm": 0.5826669335365295,
      "learning_rate": 8.006300729710203e-07,
      "loss": 0.1235,
      "num_input_tokens_seen": 69672608,
      "step": 36770
    },
    {
      "epoch": 5.999265845501265,
      "grad_norm": 0.3343041241168976,
      "learning_rate": 7.981672869662337e-07,
      "loss": 0.0457,
      "num_input_tokens_seen": 69681056,
      "step": 36775
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.13337500393390656,
      "learning_rate": 7.957082331881888e-07,
      "loss": 0.1116,
      "num_input_tokens_seen": 69689456,
      "step": 36780
    },
    {
      "epoch": 6.000815727220817,
      "grad_norm": 0.35152602195739746,
      "learning_rate": 7.932529120161069e-07,
      "loss": 0.079,
      "num_input_tokens_seen": 69699216,
      "step": 36785
    },
    {
      "epoch": 6.001631454441635,
      "grad_norm": 1.659356713294983,
      "learning_rate": 7.908013238286243e-07,
      "loss": 0.1312,
      "num_input_tokens_seen": 69709744,
      "step": 36790
    },
    {
      "epoch": 6.002447181662452,
      "grad_norm": 0.25721976161003113,
      "learning_rate": 7.883534690038136e-07,
      "loss": 0.096,
      "num_input_tokens_seen": 69720256,
      "step": 36795
    },
    {
      "epoch": 6.003262908883269,
      "grad_norm": 0.37693220376968384,
      "learning_rate": 7.859093479191559e-07,
      "loss": 0.0768,
      "num_input_tokens_seen": 69729984,
      "step": 36800
    },
    {
      "epoch": 6.003262908883269,
      "eval_loss": 0.1419488489627838,
      "eval_runtime": 133.0981,
      "eval_samples_per_second": 20.474,
      "eval_steps_per_second": 10.241,
      "num_input_tokens_seen": 69729984,
      "step": 36800
    },
    {
      "epoch": 6.0040786361040865,
      "grad_norm": 0.061566147953271866,
      "learning_rate": 7.834689609515722e-07,
      "loss": 0.1609,
      "num_input_tokens_seen": 69739184,
      "step": 36805
    },
    {
      "epoch": 6.004894363324905,
      "grad_norm": 0.3878574073314667,
      "learning_rate": 7.810323084774002e-07,
      "loss": 0.0928,
      "num_input_tokens_seen": 69748496,
      "step": 36810
    },
    {
      "epoch": 6.005710090545722,
      "grad_norm": 0.8454572558403015,
      "learning_rate": 7.785993908723976e-07,
      "loss": 0.1745,
      "num_input_tokens_seen": 69757344,
      "step": 36815
    },
    {
      "epoch": 6.006525817766539,
      "grad_norm": 0.44609153270721436,
      "learning_rate": 7.761702085117534e-07,
      "loss": 0.0399,
      "num_input_tokens_seen": 69765760,
      "step": 36820
    },
    {
      "epoch": 6.007341544987356,
      "grad_norm": 1.5857166051864624,
      "learning_rate": 7.737447617700844e-07,
      "loss": 0.1488,
      "num_input_tokens_seen": 69776000,
      "step": 36825
    },
    {
      "epoch": 6.008157272208174,
      "grad_norm": 0.09622128307819366,
      "learning_rate": 7.713230510214136e-07,
      "loss": 0.1042,
      "num_input_tokens_seen": 69784688,
      "step": 36830
    },
    {
      "epoch": 6.008972999428991,
      "grad_norm": 0.7524012923240662,
      "learning_rate": 7.689050766392092e-07,
      "loss": 0.2724,
      "num_input_tokens_seen": 69792656,
      "step": 36835
    },
    {
      "epoch": 6.009788726649808,
      "grad_norm": 1.2772387266159058,
      "learning_rate": 7.664908389963477e-07,
      "loss": 0.2604,
      "num_input_tokens_seen": 69802544,
      "step": 36840
    },
    {
      "epoch": 6.010604453870625,
      "grad_norm": 0.6349614262580872,
      "learning_rate": 7.64080338465134e-07,
      "loss": 0.3365,
      "num_input_tokens_seen": 69811200,
      "step": 36845
    },
    {
      "epoch": 6.011420181091443,
      "grad_norm": 1.3271859884262085,
      "learning_rate": 7.616735754173043e-07,
      "loss": 0.1305,
      "num_input_tokens_seen": 69821088,
      "step": 36850
    },
    {
      "epoch": 6.0122359083122605,
      "grad_norm": 0.620930552482605,
      "learning_rate": 7.592705502240005e-07,
      "loss": 0.0865,
      "num_input_tokens_seen": 69830768,
      "step": 36855
    },
    {
      "epoch": 6.013051635533078,
      "grad_norm": 0.6665832996368408,
      "learning_rate": 7.568712632558095e-07,
      "loss": 0.2155,
      "num_input_tokens_seen": 69840864,
      "step": 36860
    },
    {
      "epoch": 6.013867362753895,
      "grad_norm": 0.052765339612960815,
      "learning_rate": 7.544757148827297e-07,
      "loss": 0.1332,
      "num_input_tokens_seen": 69851008,
      "step": 36865
    },
    {
      "epoch": 6.014683089974713,
      "grad_norm": 0.5004833340644836,
      "learning_rate": 7.520839054741797e-07,
      "loss": 0.1148,
      "num_input_tokens_seen": 69861984,
      "step": 36870
    },
    {
      "epoch": 6.01549881719553,
      "grad_norm": 0.9515191316604614,
      "learning_rate": 7.496958353990113e-07,
      "loss": 0.0918,
      "num_input_tokens_seen": 69872320,
      "step": 36875
    },
    {
      "epoch": 6.016314544416347,
      "grad_norm": 2.037884473800659,
      "learning_rate": 7.473115050254941e-07,
      "loss": 0.2235,
      "num_input_tokens_seen": 69882624,
      "step": 36880
    },
    {
      "epoch": 6.017130271637164,
      "grad_norm": 0.7830978631973267,
      "learning_rate": 7.449309147213173e-07,
      "loss": 0.0977,
      "num_input_tokens_seen": 69893488,
      "step": 36885
    },
    {
      "epoch": 6.017945998857982,
      "grad_norm": 0.7206605076789856,
      "learning_rate": 7.425540648536067e-07,
      "loss": 0.1876,
      "num_input_tokens_seen": 69901280,
      "step": 36890
    },
    {
      "epoch": 6.018761726078799,
      "grad_norm": 0.3129546642303467,
      "learning_rate": 7.40180955788894e-07,
      "loss": 0.0845,
      "num_input_tokens_seen": 69910544,
      "step": 36895
    },
    {
      "epoch": 6.0195774532996165,
      "grad_norm": 0.5145483016967773,
      "learning_rate": 7.378115878931474e-07,
      "loss": 0.1204,
      "num_input_tokens_seen": 69919664,
      "step": 36900
    },
    {
      "epoch": 6.020393180520434,
      "grad_norm": 0.0860501080751419,
      "learning_rate": 7.354459615317527e-07,
      "loss": 0.1214,
      "num_input_tokens_seen": 69928256,
      "step": 36905
    },
    {
      "epoch": 6.021208907741252,
      "grad_norm": 1.2237589359283447,
      "learning_rate": 7.33084077069518e-07,
      "loss": 0.2072,
      "num_input_tokens_seen": 69937488,
      "step": 36910
    },
    {
      "epoch": 6.022024634962069,
      "grad_norm": 0.0491558201611042,
      "learning_rate": 7.307259348706768e-07,
      "loss": 0.0563,
      "num_input_tokens_seen": 69945200,
      "step": 36915
    },
    {
      "epoch": 6.022840362182886,
      "grad_norm": 1.914555549621582,
      "learning_rate": 7.283715352988801e-07,
      "loss": 0.1819,
      "num_input_tokens_seen": 69953984,
      "step": 36920
    },
    {
      "epoch": 6.023656089403703,
      "grad_norm": 0.061344634741544724,
      "learning_rate": 7.260208787172068e-07,
      "loss": 0.1072,
      "num_input_tokens_seen": 69963984,
      "step": 36925
    },
    {
      "epoch": 6.024471816624521,
      "grad_norm": 0.5127756595611572,
      "learning_rate": 7.23673965488167e-07,
      "loss": 0.0829,
      "num_input_tokens_seen": 69973200,
      "step": 36930
    },
    {
      "epoch": 6.025287543845338,
      "grad_norm": 0.03660168871283531,
      "learning_rate": 7.213307959736709e-07,
      "loss": 0.0475,
      "num_input_tokens_seen": 69983664,
      "step": 36935
    },
    {
      "epoch": 6.026103271066155,
      "grad_norm": 1.6720125675201416,
      "learning_rate": 7.189913705350715e-07,
      "loss": 0.1102,
      "num_input_tokens_seen": 69992560,
      "step": 36940
    },
    {
      "epoch": 6.026918998286972,
      "grad_norm": 0.6505802273750305,
      "learning_rate": 7.166556895331411e-07,
      "loss": 0.0714,
      "num_input_tokens_seen": 70002528,
      "step": 36945
    },
    {
      "epoch": 6.02773472550779,
      "grad_norm": 1.2028762102127075,
      "learning_rate": 7.143237533280639e-07,
      "loss": 0.1053,
      "num_input_tokens_seen": 70011680,
      "step": 36950
    },
    {
      "epoch": 6.028550452728608,
      "grad_norm": 0.4500434398651123,
      "learning_rate": 7.119955622794578e-07,
      "loss": 0.0811,
      "num_input_tokens_seen": 70021104,
      "step": 36955
    },
    {
      "epoch": 6.029366179949425,
      "grad_norm": 0.05553222820162773,
      "learning_rate": 7.096711167463577e-07,
      "loss": 0.1375,
      "num_input_tokens_seen": 70031088,
      "step": 36960
    },
    {
      "epoch": 6.030181907170242,
      "grad_norm": 0.8416180610656738,
      "learning_rate": 7.073504170872213e-07,
      "loss": 0.2439,
      "num_input_tokens_seen": 70041376,
      "step": 36965
    },
    {
      "epoch": 6.03099763439106,
      "grad_norm": 1.6629427671432495,
      "learning_rate": 7.05033463659932e-07,
      "loss": 0.1619,
      "num_input_tokens_seen": 70049808,
      "step": 36970
    },
    {
      "epoch": 6.031813361611877,
      "grad_norm": 0.9805188775062561,
      "learning_rate": 7.027202568217928e-07,
      "loss": 0.1064,
      "num_input_tokens_seen": 70059968,
      "step": 36975
    },
    {
      "epoch": 6.032629088832694,
      "grad_norm": 1.189431071281433,
      "learning_rate": 7.004107969295293e-07,
      "loss": 0.107,
      "num_input_tokens_seen": 70068848,
      "step": 36980
    },
    {
      "epoch": 6.033444816053512,
      "grad_norm": 1.1235977411270142,
      "learning_rate": 6.9810508433929e-07,
      "loss": 0.1513,
      "num_input_tokens_seen": 70079216,
      "step": 36985
    },
    {
      "epoch": 6.034260543274329,
      "grad_norm": 0.14226585626602173,
      "learning_rate": 6.958031194066406e-07,
      "loss": 0.0386,
      "num_input_tokens_seen": 70088208,
      "step": 36990
    },
    {
      "epoch": 6.035076270495146,
      "grad_norm": 0.15749835968017578,
      "learning_rate": 6.935049024865776e-07,
      "loss": 0.1321,
      "num_input_tokens_seen": 70098480,
      "step": 36995
    },
    {
      "epoch": 6.0358919977159635,
      "grad_norm": 0.7142451405525208,
      "learning_rate": 6.912104339335118e-07,
      "loss": 0.1498,
      "num_input_tokens_seen": 70107936,
      "step": 37000
    },
    {
      "epoch": 6.0358919977159635,
      "eval_loss": 0.1417521834373474,
      "eval_runtime": 133.2597,
      "eval_samples_per_second": 20.449,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 70107936,
      "step": 37000
    },
    {
      "epoch": 6.0367077249367815,
      "grad_norm": 1.0592321157455444,
      "learning_rate": 6.889197141012799e-07,
      "loss": 0.2422,
      "num_input_tokens_seen": 70115760,
      "step": 37005
    },
    {
      "epoch": 6.037523452157599,
      "grad_norm": 0.38119253516197205,
      "learning_rate": 6.866327433431435e-07,
      "loss": 0.0363,
      "num_input_tokens_seen": 70124656,
      "step": 37010
    },
    {
      "epoch": 6.038339179378416,
      "grad_norm": 0.16383963823318481,
      "learning_rate": 6.843495220117735e-07,
      "loss": 0.016,
      "num_input_tokens_seen": 70134848,
      "step": 37015
    },
    {
      "epoch": 6.039154906599233,
      "grad_norm": 0.8881050944328308,
      "learning_rate": 6.820700504592798e-07,
      "loss": 0.1289,
      "num_input_tokens_seen": 70144880,
      "step": 37020
    },
    {
      "epoch": 6.039970633820051,
      "grad_norm": 0.8769283294677734,
      "learning_rate": 6.797943290371839e-07,
      "loss": 0.117,
      "num_input_tokens_seen": 70154672,
      "step": 37025
    },
    {
      "epoch": 6.040786361040868,
      "grad_norm": 2.2005856037139893,
      "learning_rate": 6.775223580964274e-07,
      "loss": 0.2286,
      "num_input_tokens_seen": 70166000,
      "step": 37030
    },
    {
      "epoch": 6.041602088261685,
      "grad_norm": 0.5230036973953247,
      "learning_rate": 6.7525413798738e-07,
      "loss": 0.1271,
      "num_input_tokens_seen": 70176096,
      "step": 37035
    },
    {
      "epoch": 6.042417815482502,
      "grad_norm": 0.7284181714057922,
      "learning_rate": 6.729896690598259e-07,
      "loss": 0.079,
      "num_input_tokens_seen": 70184656,
      "step": 37040
    },
    {
      "epoch": 6.04323354270332,
      "grad_norm": 0.1504366546869278,
      "learning_rate": 6.707289516629772e-07,
      "loss": 0.0586,
      "num_input_tokens_seen": 70194112,
      "step": 37045
    },
    {
      "epoch": 6.0440492699241375,
      "grad_norm": 0.05514499172568321,
      "learning_rate": 6.684719861454692e-07,
      "loss": 0.055,
      "num_input_tokens_seen": 70204592,
      "step": 37050
    },
    {
      "epoch": 6.044864997144955,
      "grad_norm": 0.06403986364603043,
      "learning_rate": 6.662187728553481e-07,
      "loss": 0.0265,
      "num_input_tokens_seen": 70213376,
      "step": 37055
    },
    {
      "epoch": 6.045680724365772,
      "grad_norm": 0.8034577965736389,
      "learning_rate": 6.639693121400892e-07,
      "loss": 0.1348,
      "num_input_tokens_seen": 70224240,
      "step": 37060
    },
    {
      "epoch": 6.04649645158659,
      "grad_norm": 0.24104183912277222,
      "learning_rate": 6.617236043465868e-07,
      "loss": 0.0823,
      "num_input_tokens_seen": 70234448,
      "step": 37065
    },
    {
      "epoch": 6.047312178807407,
      "grad_norm": 1.4369597434997559,
      "learning_rate": 6.594816498211587e-07,
      "loss": 0.1815,
      "num_input_tokens_seen": 70243552,
      "step": 37070
    },
    {
      "epoch": 6.048127906028224,
      "grad_norm": 0.4575193226337433,
      "learning_rate": 6.572434489095447e-07,
      "loss": 0.0509,
      "num_input_tokens_seen": 70253392,
      "step": 37075
    },
    {
      "epoch": 6.048943633249041,
      "grad_norm": 1.0956372022628784,
      "learning_rate": 6.550090019568994e-07,
      "loss": 0.1122,
      "num_input_tokens_seen": 70263248,
      "step": 37080
    },
    {
      "epoch": 6.049759360469859,
      "grad_norm": 0.47465384006500244,
      "learning_rate": 6.527783093078027e-07,
      "loss": 0.0754,
      "num_input_tokens_seen": 70273584,
      "step": 37085
    },
    {
      "epoch": 6.050575087690676,
      "grad_norm": 1.7566686868667603,
      "learning_rate": 6.5055137130626e-07,
      "loss": 0.2,
      "num_input_tokens_seen": 70283024,
      "step": 37090
    },
    {
      "epoch": 6.051390814911493,
      "grad_norm": 2.0046534538269043,
      "learning_rate": 6.483281882956854e-07,
      "loss": 0.0869,
      "num_input_tokens_seen": 70293456,
      "step": 37095
    },
    {
      "epoch": 6.052206542132311,
      "grad_norm": 0.43608877062797546,
      "learning_rate": 6.461087606189298e-07,
      "loss": 0.0605,
      "num_input_tokens_seen": 70301968,
      "step": 37100
    },
    {
      "epoch": 6.053022269353129,
      "grad_norm": 0.6095771193504333,
      "learning_rate": 6.438930886182554e-07,
      "loss": 0.273,
      "num_input_tokens_seen": 70310560,
      "step": 37105
    },
    {
      "epoch": 6.053837996573946,
      "grad_norm": 0.4844958484172821,
      "learning_rate": 6.416811726353417e-07,
      "loss": 0.0772,
      "num_input_tokens_seen": 70319696,
      "step": 37110
    },
    {
      "epoch": 6.054653723794763,
      "grad_norm": 0.03892867639660835,
      "learning_rate": 6.394730130112991e-07,
      "loss": 0.1563,
      "num_input_tokens_seen": 70328592,
      "step": 37115
    },
    {
      "epoch": 6.05546945101558,
      "grad_norm": 0.9429265856742859,
      "learning_rate": 6.372686100866471e-07,
      "loss": 0.1837,
      "num_input_tokens_seen": 70337712,
      "step": 37120
    },
    {
      "epoch": 6.056285178236398,
      "grad_norm": 0.14810894429683685,
      "learning_rate": 6.350679642013413e-07,
      "loss": 0.1277,
      "num_input_tokens_seen": 70347200,
      "step": 37125
    },
    {
      "epoch": 6.057100905457215,
      "grad_norm": 1.1102443933486938,
      "learning_rate": 6.328710756947437e-07,
      "loss": 0.1121,
      "num_input_tokens_seen": 70356672,
      "step": 37130
    },
    {
      "epoch": 6.057916632678032,
      "grad_norm": 0.5827271342277527,
      "learning_rate": 6.306779449056416e-07,
      "loss": 0.1868,
      "num_input_tokens_seen": 70366192,
      "step": 37135
    },
    {
      "epoch": 6.058732359898849,
      "grad_norm": 1.6811280250549316,
      "learning_rate": 6.284885721722422e-07,
      "loss": 0.1577,
      "num_input_tokens_seen": 70376832,
      "step": 37140
    },
    {
      "epoch": 6.059548087119667,
      "grad_norm": 0.12052960693836212,
      "learning_rate": 6.26302957832181e-07,
      "loss": 0.0912,
      "num_input_tokens_seen": 70387344,
      "step": 37145
    },
    {
      "epoch": 6.0603638143404845,
      "grad_norm": 0.20666000247001648,
      "learning_rate": 6.241211022224997e-07,
      "loss": 0.085,
      "num_input_tokens_seen": 70397280,
      "step": 37150
    },
    {
      "epoch": 6.061179541561302,
      "grad_norm": 0.41661524772644043,
      "learning_rate": 6.219430056796732e-07,
      "loss": 0.0741,
      "num_input_tokens_seen": 70406352,
      "step": 37155
    },
    {
      "epoch": 6.06199526878212,
      "grad_norm": 0.032809868454933167,
      "learning_rate": 6.19768668539586e-07,
      "loss": 0.0511,
      "num_input_tokens_seen": 70416368,
      "step": 37160
    },
    {
      "epoch": 6.062810996002937,
      "grad_norm": 0.7543526291847229,
      "learning_rate": 6.175980911375528e-07,
      "loss": 0.1106,
      "num_input_tokens_seen": 70425984,
      "step": 37165
    },
    {
      "epoch": 6.063626723223754,
      "grad_norm": 1.7331457138061523,
      "learning_rate": 6.154312738083034e-07,
      "loss": 0.2122,
      "num_input_tokens_seen": 70435632,
      "step": 37170
    },
    {
      "epoch": 6.064442450444571,
      "grad_norm": 1.514958143234253,
      "learning_rate": 6.132682168859843e-07,
      "loss": 0.1057,
      "num_input_tokens_seen": 70443952,
      "step": 37175
    },
    {
      "epoch": 6.065258177665389,
      "grad_norm": 0.08550339937210083,
      "learning_rate": 6.111089207041704e-07,
      "loss": 0.1328,
      "num_input_tokens_seen": 70452640,
      "step": 37180
    },
    {
      "epoch": 6.066073904886206,
      "grad_norm": 0.17640568315982819,
      "learning_rate": 6.089533855958507e-07,
      "loss": 0.064,
      "num_input_tokens_seen": 70461648,
      "step": 37185
    },
    {
      "epoch": 6.066889632107023,
      "grad_norm": 0.27155017852783203,
      "learning_rate": 6.068016118934372e-07,
      "loss": 0.0498,
      "num_input_tokens_seen": 70469696,
      "step": 37190
    },
    {
      "epoch": 6.0677053593278405,
      "grad_norm": 0.3262200951576233,
      "learning_rate": 6.04653599928759e-07,
      "loss": 0.071,
      "num_input_tokens_seen": 70478736,
      "step": 37195
    },
    {
      "epoch": 6.0685210865486585,
      "grad_norm": 0.08724749088287354,
      "learning_rate": 6.025093500330675e-07,
      "loss": 0.1819,
      "num_input_tokens_seen": 70487856,
      "step": 37200
    },
    {
      "epoch": 6.0685210865486585,
      "eval_loss": 0.14197219908237457,
      "eval_runtime": 133.2292,
      "eval_samples_per_second": 20.453,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 70487856,
      "step": 37200
    },
    {
      "epoch": 6.069336813769476,
      "grad_norm": 0.11832735687494278,
      "learning_rate": 6.003688625370291e-07,
      "loss": 0.0266,
      "num_input_tokens_seen": 70496848,
      "step": 37205
    },
    {
      "epoch": 6.070152540990293,
      "grad_norm": 1.244607925415039,
      "learning_rate": 5.982321377707406e-07,
      "loss": 0.0661,
      "num_input_tokens_seen": 70506176,
      "step": 37210
    },
    {
      "epoch": 6.07096826821111,
      "grad_norm": 0.5658973455429077,
      "learning_rate": 5.96099176063708e-07,
      "loss": 0.2168,
      "num_input_tokens_seen": 70515120,
      "step": 37215
    },
    {
      "epoch": 6.071783995431928,
      "grad_norm": 0.3149878680706024,
      "learning_rate": 5.93969977744857e-07,
      "loss": 0.0495,
      "num_input_tokens_seen": 70525376,
      "step": 37220
    },
    {
      "epoch": 6.072599722652745,
      "grad_norm": 0.1614566594362259,
      "learning_rate": 5.918445431425445e-07,
      "loss": 0.0356,
      "num_input_tokens_seen": 70533904,
      "step": 37225
    },
    {
      "epoch": 6.073415449873562,
      "grad_norm": 0.5123199224472046,
      "learning_rate": 5.897228725845333e-07,
      "loss": 0.0329,
      "num_input_tokens_seen": 70543472,
      "step": 37230
    },
    {
      "epoch": 6.074231177094379,
      "grad_norm": 0.2494301199913025,
      "learning_rate": 5.876049663980171e-07,
      "loss": 0.0645,
      "num_input_tokens_seen": 70551856,
      "step": 37235
    },
    {
      "epoch": 6.075046904315197,
      "grad_norm": 0.32135412096977234,
      "learning_rate": 5.854908249095959e-07,
      "loss": 0.0834,
      "num_input_tokens_seen": 70560384,
      "step": 37240
    },
    {
      "epoch": 6.0758626315360145,
      "grad_norm": 0.5310221910476685,
      "learning_rate": 5.833804484453031e-07,
      "loss": 0.0916,
      "num_input_tokens_seen": 70569488,
      "step": 37245
    },
    {
      "epoch": 6.076678358756832,
      "grad_norm": 0.928122341632843,
      "learning_rate": 5.81273837330587e-07,
      "loss": 0.1225,
      "num_input_tokens_seen": 70579328,
      "step": 37250
    },
    {
      "epoch": 6.077494085977649,
      "grad_norm": 0.1670227348804474,
      "learning_rate": 5.791709918903071e-07,
      "loss": 0.1431,
      "num_input_tokens_seen": 70588576,
      "step": 37255
    },
    {
      "epoch": 6.078309813198467,
      "grad_norm": 0.8141670227050781,
      "learning_rate": 5.770719124487483e-07,
      "loss": 0.1401,
      "num_input_tokens_seen": 70598496,
      "step": 37260
    },
    {
      "epoch": 6.079125540419284,
      "grad_norm": 0.07864611595869064,
      "learning_rate": 5.749765993296241e-07,
      "loss": 0.1146,
      "num_input_tokens_seen": 70607840,
      "step": 37265
    },
    {
      "epoch": 6.079941267640101,
      "grad_norm": 0.40520861744880676,
      "learning_rate": 5.728850528560509e-07,
      "loss": 0.4456,
      "num_input_tokens_seen": 70618016,
      "step": 37270
    },
    {
      "epoch": 6.080756994860918,
      "grad_norm": 0.9351861476898193,
      "learning_rate": 5.707972733505707e-07,
      "loss": 0.0833,
      "num_input_tokens_seen": 70627600,
      "step": 37275
    },
    {
      "epoch": 6.081572722081736,
      "grad_norm": 0.8882789015769958,
      "learning_rate": 5.687132611351509e-07,
      "loss": 0.1614,
      "num_input_tokens_seen": 70637072,
      "step": 37280
    },
    {
      "epoch": 6.082388449302553,
      "grad_norm": 0.5337907671928406,
      "learning_rate": 5.666330165311651e-07,
      "loss": 0.1179,
      "num_input_tokens_seen": 70646640,
      "step": 37285
    },
    {
      "epoch": 6.08320417652337,
      "grad_norm": 0.8534272313117981,
      "learning_rate": 5.645565398594204e-07,
      "loss": 0.0481,
      "num_input_tokens_seen": 70656336,
      "step": 37290
    },
    {
      "epoch": 6.0840199037441876,
      "grad_norm": 0.2017996907234192,
      "learning_rate": 5.624838314401304e-07,
      "loss": 0.1619,
      "num_input_tokens_seen": 70664864,
      "step": 37295
    },
    {
      "epoch": 6.084835630965006,
      "grad_norm": 0.28792017698287964,
      "learning_rate": 5.604148915929336e-07,
      "loss": 0.1569,
      "num_input_tokens_seen": 70674768,
      "step": 37300
    },
    {
      "epoch": 6.085651358185823,
      "grad_norm": 0.3299855887889862,
      "learning_rate": 5.583497206368887e-07,
      "loss": 0.2903,
      "num_input_tokens_seen": 70684032,
      "step": 37305
    },
    {
      "epoch": 6.08646708540664,
      "grad_norm": 0.10085837543010712,
      "learning_rate": 5.562883188904688e-07,
      "loss": 0.1023,
      "num_input_tokens_seen": 70694352,
      "step": 37310
    },
    {
      "epoch": 6.087282812627457,
      "grad_norm": 0.11527833342552185,
      "learning_rate": 5.542306866715724e-07,
      "loss": 0.1579,
      "num_input_tokens_seen": 70703936,
      "step": 37315
    },
    {
      "epoch": 6.088098539848275,
      "grad_norm": 1.2219048738479614,
      "learning_rate": 5.52176824297504e-07,
      "loss": 0.246,
      "num_input_tokens_seen": 70714048,
      "step": 37320
    },
    {
      "epoch": 6.088914267069092,
      "grad_norm": 0.37858644127845764,
      "learning_rate": 5.501267320850018e-07,
      "loss": 0.1945,
      "num_input_tokens_seen": 70722736,
      "step": 37325
    },
    {
      "epoch": 6.089729994289909,
      "grad_norm": 0.054346345365047455,
      "learning_rate": 5.480804103502157e-07,
      "loss": 0.163,
      "num_input_tokens_seen": 70732928,
      "step": 37330
    },
    {
      "epoch": 6.090545721510727,
      "grad_norm": 0.8420851230621338,
      "learning_rate": 5.460378594087101e-07,
      "loss": 0.1381,
      "num_input_tokens_seen": 70741984,
      "step": 37335
    },
    {
      "epoch": 6.091361448731544,
      "grad_norm": 0.15125180780887604,
      "learning_rate": 5.439990795754773e-07,
      "loss": 0.0767,
      "num_input_tokens_seen": 70750496,
      "step": 37340
    },
    {
      "epoch": 6.0921771759523615,
      "grad_norm": 1.3896713256835938,
      "learning_rate": 5.419640711649188e-07,
      "loss": 0.1474,
      "num_input_tokens_seen": 70759744,
      "step": 37345
    },
    {
      "epoch": 6.092992903173179,
      "grad_norm": 0.35130614042282104,
      "learning_rate": 5.399328344908583e-07,
      "loss": 0.1992,
      "num_input_tokens_seen": 70769312,
      "step": 37350
    },
    {
      "epoch": 6.093808630393997,
      "grad_norm": 0.3681441843509674,
      "learning_rate": 5.379053698665399e-07,
      "loss": 0.1119,
      "num_input_tokens_seen": 70778720,
      "step": 37355
    },
    {
      "epoch": 6.094624357614814,
      "grad_norm": 1.0048545598983765,
      "learning_rate": 5.358816776046216e-07,
      "loss": 0.1012,
      "num_input_tokens_seen": 70788480,
      "step": 37360
    },
    {
      "epoch": 6.095440084835631,
      "grad_norm": 1.551300048828125,
      "learning_rate": 5.338617580171817e-07,
      "loss": 0.1632,
      "num_input_tokens_seen": 70798752,
      "step": 37365
    },
    {
      "epoch": 6.096255812056448,
      "grad_norm": 3.06945538520813,
      "learning_rate": 5.318456114157239e-07,
      "loss": 0.1201,
      "num_input_tokens_seen": 70807600,
      "step": 37370
    },
    {
      "epoch": 6.097071539277266,
      "grad_norm": 0.932137131690979,
      "learning_rate": 5.298332381111576e-07,
      "loss": 0.0466,
      "num_input_tokens_seen": 70817376,
      "step": 37375
    },
    {
      "epoch": 6.097887266498083,
      "grad_norm": 0.6422590017318726,
      "learning_rate": 5.27824638413818e-07,
      "loss": 0.1318,
      "num_input_tokens_seen": 70827856,
      "step": 37380
    },
    {
      "epoch": 6.0987029937189,
      "grad_norm": 0.40814298391342163,
      "learning_rate": 5.258198126334546e-07,
      "loss": 0.1122,
      "num_input_tokens_seen": 70837056,
      "step": 37385
    },
    {
      "epoch": 6.0995187209397175,
      "grad_norm": 0.16205383837223053,
      "learning_rate": 5.238187610792367e-07,
      "loss": 0.2232,
      "num_input_tokens_seen": 70846992,
      "step": 37390
    },
    {
      "epoch": 6.1003344481605355,
      "grad_norm": 1.4031366109848022,
      "learning_rate": 5.218214840597563e-07,
      "loss": 0.2111,
      "num_input_tokens_seen": 70856160,
      "step": 37395
    },
    {
      "epoch": 6.101150175381353,
      "grad_norm": 0.830062210559845,
      "learning_rate": 5.198279818830115e-07,
      "loss": 0.0429,
      "num_input_tokens_seen": 70865792,
      "step": 37400
    },
    {
      "epoch": 6.101150175381353,
      "eval_loss": 0.142240509390831,
      "eval_runtime": 133.2403,
      "eval_samples_per_second": 20.452,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 70865792,
      "step": 37400
    },
    {
      "epoch": 6.10196590260217,
      "grad_norm": 0.16173061728477478,
      "learning_rate": 5.178382548564287e-07,
      "loss": 0.1109,
      "num_input_tokens_seen": 70874144,
      "step": 37405
    },
    {
      "epoch": 6.102781629822987,
      "grad_norm": 0.7772011160850525,
      "learning_rate": 5.15852303286854e-07,
      "loss": 0.2153,
      "num_input_tokens_seen": 70884144,
      "step": 37410
    },
    {
      "epoch": 6.103597357043805,
      "grad_norm": 0.7379999756813049,
      "learning_rate": 5.138701274805396e-07,
      "loss": 0.0973,
      "num_input_tokens_seen": 70894416,
      "step": 37415
    },
    {
      "epoch": 6.104413084264622,
      "grad_norm": 0.9930481910705566,
      "learning_rate": 5.118917277431606e-07,
      "loss": 0.2823,
      "num_input_tokens_seen": 70903152,
      "step": 37420
    },
    {
      "epoch": 6.105228811485439,
      "grad_norm": 0.04131598398089409,
      "learning_rate": 5.099171043798145e-07,
      "loss": 0.0893,
      "num_input_tokens_seen": 70911888,
      "step": 37425
    },
    {
      "epoch": 6.106044538706256,
      "grad_norm": 0.25752103328704834,
      "learning_rate": 5.079462576950133e-07,
      "loss": 0.1035,
      "num_input_tokens_seen": 70920544,
      "step": 37430
    },
    {
      "epoch": 6.106860265927074,
      "grad_norm": 0.300498902797699,
      "learning_rate": 5.059791879926862e-07,
      "loss": 0.0418,
      "num_input_tokens_seen": 70930064,
      "step": 37435
    },
    {
      "epoch": 6.1076759931478914,
      "grad_norm": 0.21560868620872498,
      "learning_rate": 5.040158955761793e-07,
      "loss": 0.0907,
      "num_input_tokens_seen": 70939808,
      "step": 37440
    },
    {
      "epoch": 6.108491720368709,
      "grad_norm": 0.06696972250938416,
      "learning_rate": 5.020563807482559e-07,
      "loss": 0.196,
      "num_input_tokens_seen": 70948032,
      "step": 37445
    },
    {
      "epoch": 6.109307447589526,
      "grad_norm": 1.3764002323150635,
      "learning_rate": 5.001006438110995e-07,
      "loss": 0.1873,
      "num_input_tokens_seen": 70956560,
      "step": 37450
    },
    {
      "epoch": 6.110123174810344,
      "grad_norm": 1.6262720823287964,
      "learning_rate": 4.981486850663075e-07,
      "loss": 0.1334,
      "num_input_tokens_seen": 70966064,
      "step": 37455
    },
    {
      "epoch": 6.110938902031161,
      "grad_norm": 0.3454717695713043,
      "learning_rate": 4.962005048149005e-07,
      "loss": 0.1353,
      "num_input_tokens_seen": 70976016,
      "step": 37460
    },
    {
      "epoch": 6.111754629251978,
      "grad_norm": 0.15658040344715118,
      "learning_rate": 4.942561033573073e-07,
      "loss": 0.143,
      "num_input_tokens_seen": 70985280,
      "step": 37465
    },
    {
      "epoch": 6.112570356472795,
      "grad_norm": 1.2963052988052368,
      "learning_rate": 4.923154809933827e-07,
      "loss": 0.1108,
      "num_input_tokens_seen": 70993616,
      "step": 37470
    },
    {
      "epoch": 6.113386083693613,
      "grad_norm": 0.06589144468307495,
      "learning_rate": 4.903786380223957e-07,
      "loss": 0.1159,
      "num_input_tokens_seen": 71002624,
      "step": 37475
    },
    {
      "epoch": 6.11420181091443,
      "grad_norm": 0.2300928831100464,
      "learning_rate": 4.884455747430266e-07,
      "loss": 0.169,
      "num_input_tokens_seen": 71012880,
      "step": 37480
    },
    {
      "epoch": 6.115017538135247,
      "grad_norm": 0.07361887395381927,
      "learning_rate": 4.865162914533816e-07,
      "loss": 0.2892,
      "num_input_tokens_seen": 71023536,
      "step": 37485
    },
    {
      "epoch": 6.1158332653560645,
      "grad_norm": 0.03558274358510971,
      "learning_rate": 4.845907884509809e-07,
      "loss": 0.0622,
      "num_input_tokens_seen": 71032992,
      "step": 37490
    },
    {
      "epoch": 6.1166489925768825,
      "grad_norm": 0.29515424370765686,
      "learning_rate": 4.82669066032762e-07,
      "loss": 0.1373,
      "num_input_tokens_seen": 71042032,
      "step": 37495
    },
    {
      "epoch": 6.1174647197977,
      "grad_norm": 1.0041766166687012,
      "learning_rate": 4.807511244950768e-07,
      "loss": 0.1804,
      "num_input_tokens_seen": 71051824,
      "step": 37500
    },
    {
      "epoch": 6.118280447018517,
      "grad_norm": 0.11466759443283081,
      "learning_rate": 4.788369641336943e-07,
      "loss": 0.0818,
      "num_input_tokens_seen": 71061328,
      "step": 37505
    },
    {
      "epoch": 6.119096174239334,
      "grad_norm": 0.0512528121471405,
      "learning_rate": 4.769265852438032e-07,
      "loss": 0.0703,
      "num_input_tokens_seen": 71070880,
      "step": 37510
    },
    {
      "epoch": 6.119911901460152,
      "grad_norm": 0.5155877470970154,
      "learning_rate": 4.750199881200124e-07,
      "loss": 0.1076,
      "num_input_tokens_seen": 71080736,
      "step": 37515
    },
    {
      "epoch": 6.120727628680969,
      "grad_norm": 1.5289673805236816,
      "learning_rate": 4.7311717305633664e-07,
      "loss": 0.191,
      "num_input_tokens_seen": 71089920,
      "step": 37520
    },
    {
      "epoch": 6.121543355901786,
      "grad_norm": 1.0573382377624512,
      "learning_rate": 4.7121814034621623e-07,
      "loss": 0.2222,
      "num_input_tokens_seen": 71100368,
      "step": 37525
    },
    {
      "epoch": 6.122359083122603,
      "grad_norm": 0.654712438583374,
      "learning_rate": 4.693228902825114e-07,
      "loss": 0.026,
      "num_input_tokens_seen": 71110288,
      "step": 37530
    },
    {
      "epoch": 6.123174810343421,
      "grad_norm": 0.03610575571656227,
      "learning_rate": 4.6743142315748277e-07,
      "loss": 0.0805,
      "num_input_tokens_seen": 71120880,
      "step": 37535
    },
    {
      "epoch": 6.1239905375642385,
      "grad_norm": 1.5384498834609985,
      "learning_rate": 4.655437392628276e-07,
      "loss": 0.2447,
      "num_input_tokens_seen": 71129744,
      "step": 37540
    },
    {
      "epoch": 6.124806264785056,
      "grad_norm": 0.8769631385803223,
      "learning_rate": 4.636598388896463e-07,
      "loss": 0.0912,
      "num_input_tokens_seen": 71139472,
      "step": 37545
    },
    {
      "epoch": 6.125621992005874,
      "grad_norm": 0.05115659907460213,
      "learning_rate": 4.6177972232845925e-07,
      "loss": 0.0551,
      "num_input_tokens_seen": 71148896,
      "step": 37550
    },
    {
      "epoch": 6.126437719226691,
      "grad_norm": 1.4552154541015625,
      "learning_rate": 4.5990338986920953e-07,
      "loss": 0.1525,
      "num_input_tokens_seen": 71158928,
      "step": 37555
    },
    {
      "epoch": 6.127253446447508,
      "grad_norm": 0.07892758399248123,
      "learning_rate": 4.5803084180124633e-07,
      "loss": 0.0494,
      "num_input_tokens_seen": 71168240,
      "step": 37560
    },
    {
      "epoch": 6.128069173668325,
      "grad_norm": 0.8649349808692932,
      "learning_rate": 4.561620784133386e-07,
      "loss": 0.0615,
      "num_input_tokens_seen": 71178832,
      "step": 37565
    },
    {
      "epoch": 6.128884900889143,
      "grad_norm": 0.13628444075584412,
      "learning_rate": 4.5429709999367796e-07,
      "loss": 0.1233,
      "num_input_tokens_seen": 71189600,
      "step": 37570
    },
    {
      "epoch": 6.12970062810996,
      "grad_norm": 0.3179175853729248,
      "learning_rate": 4.5243590682986223e-07,
      "loss": 0.0746,
      "num_input_tokens_seen": 71198192,
      "step": 37575
    },
    {
      "epoch": 6.130516355330777,
      "grad_norm": 0.07612282037734985,
      "learning_rate": 4.5057849920891735e-07,
      "loss": 0.0416,
      "num_input_tokens_seen": 71207184,
      "step": 37580
    },
    {
      "epoch": 6.1313320825515945,
      "grad_norm": 0.4873189330101013,
      "learning_rate": 4.487248774172698e-07,
      "loss": 0.0705,
      "num_input_tokens_seen": 71216896,
      "step": 37585
    },
    {
      "epoch": 6.1321478097724125,
      "grad_norm": 2.1104042530059814,
      "learning_rate": 4.4687504174077965e-07,
      "loss": 0.1658,
      "num_input_tokens_seen": 71226400,
      "step": 37590
    },
    {
      "epoch": 6.13296353699323,
      "grad_norm": 1.4761812686920166,
      "learning_rate": 4.450289924647133e-07,
      "loss": 0.2763,
      "num_input_tokens_seen": 71235424,
      "step": 37595
    },
    {
      "epoch": 6.133779264214047,
      "grad_norm": 0.5851972103118896,
      "learning_rate": 4.431867298737513e-07,
      "loss": 0.0431,
      "num_input_tokens_seen": 71244784,
      "step": 37600
    },
    {
      "epoch": 6.133779264214047,
      "eval_loss": 0.142141193151474,
      "eval_runtime": 133.0834,
      "eval_samples_per_second": 20.476,
      "eval_steps_per_second": 10.242,
      "num_input_tokens_seen": 71244784,
      "step": 37600
    },
    {
      "epoch": 6.134594991434864,
      "grad_norm": 1.2349660396575928,
      "learning_rate": 4.41348254251997e-07,
      "loss": 0.2096,
      "num_input_tokens_seen": 71253712,
      "step": 37605
    },
    {
      "epoch": 6.135410718655682,
      "grad_norm": 0.6126876473426819,
      "learning_rate": 4.395135658829652e-07,
      "loss": 0.1035,
      "num_input_tokens_seen": 71262624,
      "step": 37610
    },
    {
      "epoch": 6.136226445876499,
      "grad_norm": 1.0131255388259888,
      "learning_rate": 4.376826650495852e-07,
      "loss": 0.1354,
      "num_input_tokens_seen": 71272928,
      "step": 37615
    },
    {
      "epoch": 6.137042173097316,
      "grad_norm": 0.817962110042572,
      "learning_rate": 4.358555520342117e-07,
      "loss": 0.1325,
      "num_input_tokens_seen": 71281616,
      "step": 37620
    },
    {
      "epoch": 6.137857900318133,
      "grad_norm": 0.7859652042388916,
      "learning_rate": 4.3403222711860257e-07,
      "loss": 0.0628,
      "num_input_tokens_seen": 71291392,
      "step": 37625
    },
    {
      "epoch": 6.138673627538951,
      "grad_norm": 0.2131226360797882,
      "learning_rate": 4.3221269058394133e-07,
      "loss": 0.0212,
      "num_input_tokens_seen": 71302160,
      "step": 37630
    },
    {
      "epoch": 6.139489354759768,
      "grad_norm": 0.17753930389881134,
      "learning_rate": 4.303969427108173e-07,
      "loss": 0.1272,
      "num_input_tokens_seen": 71312640,
      "step": 37635
    },
    {
      "epoch": 6.1403050819805856,
      "grad_norm": 0.07793349772691727,
      "learning_rate": 4.2858498377924825e-07,
      "loss": 0.0449,
      "num_input_tokens_seen": 71323424,
      "step": 37640
    },
    {
      "epoch": 6.141120809201403,
      "grad_norm": 0.1422722339630127,
      "learning_rate": 4.267768140686579e-07,
      "loss": 0.0275,
      "num_input_tokens_seen": 71332880,
      "step": 37645
    },
    {
      "epoch": 6.141936536422221,
      "grad_norm": 1.7822301387786865,
      "learning_rate": 4.2497243385788975e-07,
      "loss": 0.2115,
      "num_input_tokens_seen": 71342080,
      "step": 37650
    },
    {
      "epoch": 6.142752263643038,
      "grad_norm": 0.24618089199066162,
      "learning_rate": 4.231718434251991e-07,
      "loss": 0.1129,
      "num_input_tokens_seen": 71352592,
      "step": 37655
    },
    {
      "epoch": 6.143567990863855,
      "grad_norm": 2.13999080657959,
      "learning_rate": 4.213750430482666e-07,
      "loss": 0.3158,
      "num_input_tokens_seen": 71361664,
      "step": 37660
    },
    {
      "epoch": 6.144383718084672,
      "grad_norm": 2.502864122390747,
      "learning_rate": 4.1958203300417054e-07,
      "loss": 0.2548,
      "num_input_tokens_seen": 71370976,
      "step": 37665
    },
    {
      "epoch": 6.14519944530549,
      "grad_norm": 0.3782382011413574,
      "learning_rate": 4.177928135694259e-07,
      "loss": 0.043,
      "num_input_tokens_seen": 71381456,
      "step": 37670
    },
    {
      "epoch": 6.146015172526307,
      "grad_norm": 1.2181041240692139,
      "learning_rate": 4.1600738501994807e-07,
      "loss": 0.1631,
      "num_input_tokens_seen": 71391264,
      "step": 37675
    },
    {
      "epoch": 6.146830899747124,
      "grad_norm": 0.5196859240531921,
      "learning_rate": 4.1422574763107237e-07,
      "loss": 0.1328,
      "num_input_tokens_seen": 71400816,
      "step": 37680
    },
    {
      "epoch": 6.1476466269679415,
      "grad_norm": 0.2842521071434021,
      "learning_rate": 4.124479016775512e-07,
      "loss": 0.1841,
      "num_input_tokens_seen": 71410816,
      "step": 37685
    },
    {
      "epoch": 6.1484623541887595,
      "grad_norm": 0.5721525549888611,
      "learning_rate": 4.106738474335514e-07,
      "loss": 0.1414,
      "num_input_tokens_seen": 71420416,
      "step": 37690
    },
    {
      "epoch": 6.149278081409577,
      "grad_norm": 1.0437536239624023,
      "learning_rate": 4.089035851726486e-07,
      "loss": 0.0914,
      "num_input_tokens_seen": 71429376,
      "step": 37695
    },
    {
      "epoch": 6.150093808630394,
      "grad_norm": 1.1785798072814941,
      "learning_rate": 4.0713711516784937e-07,
      "loss": 0.0964,
      "num_input_tokens_seen": 71439584,
      "step": 37700
    },
    {
      "epoch": 6.150909535851211,
      "grad_norm": 0.08193283528089523,
      "learning_rate": 4.05374437691558e-07,
      "loss": 0.113,
      "num_input_tokens_seen": 71448784,
      "step": 37705
    },
    {
      "epoch": 6.151725263072029,
      "grad_norm": 0.42902955412864685,
      "learning_rate": 4.036155530156044e-07,
      "loss": 0.1046,
      "num_input_tokens_seen": 71458400,
      "step": 37710
    },
    {
      "epoch": 6.152540990292846,
      "grad_norm": 2.8786427974700928,
      "learning_rate": 4.018604614112298e-07,
      "loss": 0.1913,
      "num_input_tokens_seen": 71468352,
      "step": 37715
    },
    {
      "epoch": 6.153356717513663,
      "grad_norm": 1.5820655822753906,
      "learning_rate": 4.0010916314908996e-07,
      "loss": 0.1738,
      "num_input_tokens_seen": 71478896,
      "step": 37720
    },
    {
      "epoch": 6.154172444734481,
      "grad_norm": 0.6425079703330994,
      "learning_rate": 3.983616584992578e-07,
      "loss": 0.2079,
      "num_input_tokens_seen": 71488752,
      "step": 37725
    },
    {
      "epoch": 6.154988171955298,
      "grad_norm": 0.9049738645553589,
      "learning_rate": 3.9661794773122595e-07,
      "loss": 0.0977,
      "num_input_tokens_seen": 71498112,
      "step": 37730
    },
    {
      "epoch": 6.1558038991761155,
      "grad_norm": 1.1701891422271729,
      "learning_rate": 3.9487803111388777e-07,
      "loss": 0.3275,
      "num_input_tokens_seen": 71507760,
      "step": 37735
    },
    {
      "epoch": 6.156619626396933,
      "grad_norm": 0.295661062002182,
      "learning_rate": 3.9314190891556747e-07,
      "loss": 0.1518,
      "num_input_tokens_seen": 71516560,
      "step": 37740
    },
    {
      "epoch": 6.157435353617751,
      "grad_norm": 1.6094647645950317,
      "learning_rate": 3.914095814039925e-07,
      "loss": 0.1231,
      "num_input_tokens_seen": 71525984,
      "step": 37745
    },
    {
      "epoch": 6.158251080838568,
      "grad_norm": 1.24552321434021,
      "learning_rate": 3.896810488463104e-07,
      "loss": 0.2069,
      "num_input_tokens_seen": 71536112,
      "step": 37750
    },
    {
      "epoch": 6.159066808059385,
      "grad_norm": 0.13560180366039276,
      "learning_rate": 3.8795631150908565e-07,
      "loss": 0.1151,
      "num_input_tokens_seen": 71546112,
      "step": 37755
    },
    {
      "epoch": 6.159882535280202,
      "grad_norm": 0.1542254239320755,
      "learning_rate": 3.862353696582888e-07,
      "loss": 0.3403,
      "num_input_tokens_seen": 71555040,
      "step": 37760
    },
    {
      "epoch": 6.16069826250102,
      "grad_norm": 0.24382680654525757,
      "learning_rate": 3.8451822355931313e-07,
      "loss": 0.1141,
      "num_input_tokens_seen": 71564560,
      "step": 37765
    },
    {
      "epoch": 6.161513989721837,
      "grad_norm": 2.0937347412109375,
      "learning_rate": 3.82804873476969e-07,
      "loss": 0.1035,
      "num_input_tokens_seen": 71574928,
      "step": 37770
    },
    {
      "epoch": 6.162329716942654,
      "grad_norm": 0.9122528433799744,
      "learning_rate": 3.810953196754702e-07,
      "loss": 0.1785,
      "num_input_tokens_seen": 71583872,
      "step": 37775
    },
    {
      "epoch": 6.163145444163471,
      "grad_norm": 0.8350627422332764,
      "learning_rate": 3.793895624184529e-07,
      "loss": 0.0824,
      "num_input_tokens_seen": 71593440,
      "step": 37780
    },
    {
      "epoch": 6.1639611713842895,
      "grad_norm": 0.5701081156730652,
      "learning_rate": 3.776876019689679e-07,
      "loss": 0.1385,
      "num_input_tokens_seen": 71603248,
      "step": 37785
    },
    {
      "epoch": 6.164776898605107,
      "grad_norm": 1.1921964883804321,
      "learning_rate": 3.7598943858947743e-07,
      "loss": 0.153,
      "num_input_tokens_seen": 71611904,
      "step": 37790
    },
    {
      "epoch": 6.165592625825924,
      "grad_norm": 1.21902334690094,
      "learning_rate": 3.742950725418637e-07,
      "loss": 0.1577,
      "num_input_tokens_seen": 71621584,
      "step": 37795
    },
    {
      "epoch": 6.166408353046741,
      "grad_norm": 0.15285412967205048,
      "learning_rate": 3.726045040874093e-07,
      "loss": 0.2317,
      "num_input_tokens_seen": 71630704,
      "step": 37800
    },
    {
      "epoch": 6.166408353046741,
      "eval_loss": 0.14202100038528442,
      "eval_runtime": 133.0932,
      "eval_samples_per_second": 20.474,
      "eval_steps_per_second": 10.241,
      "num_input_tokens_seen": 71630704,
      "step": 37800
    },
    {
      "epoch": 6.167224080267559,
      "grad_norm": 0.578126072883606,
      "learning_rate": 3.709177334868308e-07,
      "loss": 0.0676,
      "num_input_tokens_seen": 71640240,
      "step": 37805
    },
    {
      "epoch": 6.168039807488376,
      "grad_norm": 2.371779203414917,
      "learning_rate": 3.692347610002478e-07,
      "loss": 0.1835,
      "num_input_tokens_seen": 71649056,
      "step": 37810
    },
    {
      "epoch": 6.168855534709193,
      "grad_norm": 0.04830339178442955,
      "learning_rate": 3.675555868871916e-07,
      "loss": 0.2559,
      "num_input_tokens_seen": 71657552,
      "step": 37815
    },
    {
      "epoch": 6.16967126193001,
      "grad_norm": 0.4727558493614197,
      "learning_rate": 3.658802114066162e-07,
      "loss": 0.1286,
      "num_input_tokens_seen": 71667888,
      "step": 37820
    },
    {
      "epoch": 6.170486989150828,
      "grad_norm": 0.04783322662115097,
      "learning_rate": 3.6420863481688437e-07,
      "loss": 0.0459,
      "num_input_tokens_seen": 71677456,
      "step": 37825
    },
    {
      "epoch": 6.171302716371645,
      "grad_norm": 0.15039578080177307,
      "learning_rate": 3.625408573757705e-07,
      "loss": 0.073,
      "num_input_tokens_seen": 71687056,
      "step": 37830
    },
    {
      "epoch": 6.1721184435924625,
      "grad_norm": 0.9551626443862915,
      "learning_rate": 3.608768793404743e-07,
      "loss": 0.1094,
      "num_input_tokens_seen": 71696112,
      "step": 37835
    },
    {
      "epoch": 6.17293417081328,
      "grad_norm": 1.35858952999115,
      "learning_rate": 3.592167009675934e-07,
      "loss": 0.1303,
      "num_input_tokens_seen": 71705408,
      "step": 37840
    },
    {
      "epoch": 6.173749898034098,
      "grad_norm": 0.4193723797798157,
      "learning_rate": 3.575603225131563e-07,
      "loss": 0.1125,
      "num_input_tokens_seen": 71714624,
      "step": 37845
    },
    {
      "epoch": 6.174565625254915,
      "grad_norm": 0.23733043670654297,
      "learning_rate": 3.55907744232592e-07,
      "loss": 0.0894,
      "num_input_tokens_seen": 71724384,
      "step": 37850
    },
    {
      "epoch": 6.175381352475732,
      "grad_norm": 0.19763846695423126,
      "learning_rate": 3.5425896638075217e-07,
      "loss": 0.1943,
      "num_input_tokens_seen": 71733328,
      "step": 37855
    },
    {
      "epoch": 6.176197079696549,
      "grad_norm": 0.11355762183666229,
      "learning_rate": 3.5261398921189736e-07,
      "loss": 0.0141,
      "num_input_tokens_seen": 71741856,
      "step": 37860
    },
    {
      "epoch": 6.177012806917367,
      "grad_norm": 0.5789806842803955,
      "learning_rate": 3.509728129797024e-07,
      "loss": 0.096,
      "num_input_tokens_seen": 71751584,
      "step": 37865
    },
    {
      "epoch": 6.177828534138184,
      "grad_norm": 0.6497524976730347,
      "learning_rate": 3.4933543793725656e-07,
      "loss": 0.1467,
      "num_input_tokens_seen": 71761104,
      "step": 37870
    },
    {
      "epoch": 6.178644261359001,
      "grad_norm": 1.2301493883132935,
      "learning_rate": 3.4770186433707163e-07,
      "loss": 0.2159,
      "num_input_tokens_seen": 71770880,
      "step": 37875
    },
    {
      "epoch": 6.1794599885798185,
      "grad_norm": 2.3581128120422363,
      "learning_rate": 3.4607209243105453e-07,
      "loss": 0.1588,
      "num_input_tokens_seen": 71780384,
      "step": 37880
    },
    {
      "epoch": 6.1802757158006365,
      "grad_norm": 0.4834732711315155,
      "learning_rate": 3.444461224705431e-07,
      "loss": 0.1983,
      "num_input_tokens_seen": 71790016,
      "step": 37885
    },
    {
      "epoch": 6.181091443021454,
      "grad_norm": 0.1713651716709137,
      "learning_rate": 3.4282395470628116e-07,
      "loss": 0.0863,
      "num_input_tokens_seen": 71798704,
      "step": 37890
    },
    {
      "epoch": 6.181907170242271,
      "grad_norm": 0.28105196356773376,
      "learning_rate": 3.4120558938842417e-07,
      "loss": 0.0384,
      "num_input_tokens_seen": 71808448,
      "step": 37895
    },
    {
      "epoch": 6.182722897463089,
      "grad_norm": 1.7097055912017822,
      "learning_rate": 3.395910267665503e-07,
      "loss": 0.1603,
      "num_input_tokens_seen": 71818080,
      "step": 37900
    },
    {
      "epoch": 6.183538624683906,
      "grad_norm": 0.06527135521173477,
      "learning_rate": 3.3798026708964094e-07,
      "loss": 0.1722,
      "num_input_tokens_seen": 71827504,
      "step": 37905
    },
    {
      "epoch": 6.184354351904723,
      "grad_norm": 1.1969215869903564,
      "learning_rate": 3.3637331060609456e-07,
      "loss": 0.2699,
      "num_input_tokens_seen": 71837776,
      "step": 37910
    },
    {
      "epoch": 6.18517007912554,
      "grad_norm": 0.10933050513267517,
      "learning_rate": 3.3477015756372966e-07,
      "loss": 0.0813,
      "num_input_tokens_seen": 71846096,
      "step": 37915
    },
    {
      "epoch": 6.185985806346358,
      "grad_norm": 0.39872872829437256,
      "learning_rate": 3.3317080820976785e-07,
      "loss": 0.1001,
      "num_input_tokens_seen": 71855040,
      "step": 37920
    },
    {
      "epoch": 6.186801533567175,
      "grad_norm": 1.3030352592468262,
      "learning_rate": 3.315752627908508e-07,
      "loss": 0.0683,
      "num_input_tokens_seen": 71864336,
      "step": 37925
    },
    {
      "epoch": 6.1876172607879925,
      "grad_norm": 1.3912348747253418,
      "learning_rate": 3.299835215530317e-07,
      "loss": 0.1748,
      "num_input_tokens_seen": 71873472,
      "step": 37930
    },
    {
      "epoch": 6.18843298800881,
      "grad_norm": 1.321181058883667,
      "learning_rate": 3.2839558474177245e-07,
      "loss": 0.1614,
      "num_input_tokens_seen": 71883040,
      "step": 37935
    },
    {
      "epoch": 6.189248715229628,
      "grad_norm": 0.05989128723740578,
      "learning_rate": 3.2681145260196056e-07,
      "loss": 0.1803,
      "num_input_tokens_seen": 71891808,
      "step": 37940
    },
    {
      "epoch": 6.190064442450445,
      "grad_norm": 0.14857301115989685,
      "learning_rate": 3.252311253778839e-07,
      "loss": 0.1079,
      "num_input_tokens_seen": 71902288,
      "step": 37945
    },
    {
      "epoch": 6.190880169671262,
      "grad_norm": 0.46656492352485657,
      "learning_rate": 3.2365460331325034e-07,
      "loss": 0.0483,
      "num_input_tokens_seen": 71912544,
      "step": 37950
    },
    {
      "epoch": 6.191695896892079,
      "grad_norm": 0.06231873482465744,
      "learning_rate": 3.2208188665117934e-07,
      "loss": 0.0138,
      "num_input_tokens_seen": 71921840,
      "step": 37955
    },
    {
      "epoch": 6.192511624112897,
      "grad_norm": 0.17393912374973297,
      "learning_rate": 3.205129756342018e-07,
      "loss": 0.2151,
      "num_input_tokens_seen": 71929792,
      "step": 37960
    },
    {
      "epoch": 6.193327351333714,
      "grad_norm": 1.2719461917877197,
      "learning_rate": 3.189478705042659e-07,
      "loss": 0.1555,
      "num_input_tokens_seen": 71937264,
      "step": 37965
    },
    {
      "epoch": 6.194143078554531,
      "grad_norm": 0.48175373673439026,
      "learning_rate": 3.173865715027341e-07,
      "loss": 0.1467,
      "num_input_tokens_seen": 71946096,
      "step": 37970
    },
    {
      "epoch": 6.194958805775348,
      "grad_norm": 3.279243230819702,
      "learning_rate": 3.158290788703694e-07,
      "loss": 0.2651,
      "num_input_tokens_seen": 71954672,
      "step": 37975
    },
    {
      "epoch": 6.195774532996166,
      "grad_norm": 1.0198520421981812,
      "learning_rate": 3.1427539284736297e-07,
      "loss": 0.148,
      "num_input_tokens_seen": 71965056,
      "step": 37980
    },
    {
      "epoch": 6.196590260216984,
      "grad_norm": 1.1843591928482056,
      "learning_rate": 3.127255136733093e-07,
      "loss": 0.168,
      "num_input_tokens_seen": 71973936,
      "step": 37985
    },
    {
      "epoch": 6.197405987437801,
      "grad_norm": 0.16966338455677032,
      "learning_rate": 3.1117944158722544e-07,
      "loss": 0.0318,
      "num_input_tokens_seen": 71983936,
      "step": 37990
    },
    {
      "epoch": 6.198221714658618,
      "grad_norm": 0.36498942971229553,
      "learning_rate": 3.0963717682752635e-07,
      "loss": 0.1164,
      "num_input_tokens_seen": 71993264,
      "step": 37995
    },
    {
      "epoch": 6.199037441879436,
      "grad_norm": 0.08510050922632217,
      "learning_rate": 3.080987196320578e-07,
      "loss": 0.0573,
      "num_input_tokens_seen": 72002688,
      "step": 38000
    },
    {
      "epoch": 6.199037441879436,
      "eval_loss": 0.1414705514907837,
      "eval_runtime": 133.298,
      "eval_samples_per_second": 20.443,
      "eval_steps_per_second": 10.225,
      "num_input_tokens_seen": 72002688,
      "step": 38000
    },
    {
      "epoch": 6.199853169100253,
      "grad_norm": 0.4790327548980713,
      "learning_rate": 3.065640702380607e-07,
      "loss": 0.0766,
      "num_input_tokens_seen": 72012272,
      "step": 38005
    },
    {
      "epoch": 6.20066889632107,
      "grad_norm": 0.08068227022886276,
      "learning_rate": 3.050332288822011e-07,
      "loss": 0.1032,
      "num_input_tokens_seen": 72021328,
      "step": 38010
    },
    {
      "epoch": 6.201484623541887,
      "grad_norm": 0.2783799171447754,
      "learning_rate": 3.035061958005542e-07,
      "loss": 0.2636,
      "num_input_tokens_seen": 72032048,
      "step": 38015
    },
    {
      "epoch": 6.202300350762705,
      "grad_norm": 0.05281477048993111,
      "learning_rate": 3.019829712286093e-07,
      "loss": 0.1208,
      "num_input_tokens_seen": 72042560,
      "step": 38020
    },
    {
      "epoch": 6.203116077983522,
      "grad_norm": 0.12457580864429474,
      "learning_rate": 3.004635554012647e-07,
      "loss": 0.1038,
      "num_input_tokens_seen": 72051712,
      "step": 38025
    },
    {
      "epoch": 6.2039318052043395,
      "grad_norm": 2.8357467651367188,
      "learning_rate": 2.9894794855283017e-07,
      "loss": 0.1747,
      "num_input_tokens_seen": 72061936,
      "step": 38030
    },
    {
      "epoch": 6.204747532425157,
      "grad_norm": 0.24785834550857544,
      "learning_rate": 2.9743615091703816e-07,
      "loss": 0.1734,
      "num_input_tokens_seen": 72070752,
      "step": 38035
    },
    {
      "epoch": 6.205563259645975,
      "grad_norm": 0.05933262035250664,
      "learning_rate": 2.959281627270216e-07,
      "loss": 0.0603,
      "num_input_tokens_seen": 72079472,
      "step": 38040
    },
    {
      "epoch": 6.206378986866792,
      "grad_norm": 0.8043797612190247,
      "learning_rate": 2.944239842153362e-07,
      "loss": 0.2472,
      "num_input_tokens_seen": 72089120,
      "step": 38045
    },
    {
      "epoch": 6.207194714087609,
      "grad_norm": 0.21206800639629364,
      "learning_rate": 2.929236156139381e-07,
      "loss": 0.2269,
      "num_input_tokens_seen": 72098064,
      "step": 38050
    },
    {
      "epoch": 6.208010441308426,
      "grad_norm": 0.7323465943336487,
      "learning_rate": 2.9142705715420883e-07,
      "loss": 0.1497,
      "num_input_tokens_seen": 72107584,
      "step": 38055
    },
    {
      "epoch": 6.208826168529244,
      "grad_norm": 0.5642175078392029,
      "learning_rate": 2.8993430906693595e-07,
      "loss": 0.1037,
      "num_input_tokens_seen": 72117392,
      "step": 38060
    },
    {
      "epoch": 6.209641895750061,
      "grad_norm": 1.37083101272583,
      "learning_rate": 2.88445371582316e-07,
      "loss": 0.0782,
      "num_input_tokens_seen": 72126480,
      "step": 38065
    },
    {
      "epoch": 6.210457622970878,
      "grad_norm": 0.4858981966972351,
      "learning_rate": 2.8696024492996796e-07,
      "loss": 0.168,
      "num_input_tokens_seen": 72135792,
      "step": 38070
    },
    {
      "epoch": 6.211273350191696,
      "grad_norm": 0.26306429505348206,
      "learning_rate": 2.854789293389115e-07,
      "loss": 0.1144,
      "num_input_tokens_seen": 72145104,
      "step": 38075
    },
    {
      "epoch": 6.2120890774125135,
      "grad_norm": 1.4512543678283691,
      "learning_rate": 2.8400142503758606e-07,
      "loss": 0.1222,
      "num_input_tokens_seen": 72154640,
      "step": 38080
    },
    {
      "epoch": 6.212904804633331,
      "grad_norm": 1.0050443410873413,
      "learning_rate": 2.8252773225384276e-07,
      "loss": 0.059,
      "num_input_tokens_seen": 72164768,
      "step": 38085
    },
    {
      "epoch": 6.213720531854148,
      "grad_norm": 1.8486638069152832,
      "learning_rate": 2.8105785121494143e-07,
      "loss": 0.2031,
      "num_input_tokens_seen": 72175200,
      "step": 38090
    },
    {
      "epoch": 6.214536259074965,
      "grad_norm": 0.12297284603118896,
      "learning_rate": 2.795917821475563e-07,
      "loss": 0.0541,
      "num_input_tokens_seen": 72184512,
      "step": 38095
    },
    {
      "epoch": 6.215351986295783,
      "grad_norm": 0.8516993522644043,
      "learning_rate": 2.78129525277776e-07,
      "loss": 0.1696,
      "num_input_tokens_seen": 72194160,
      "step": 38100
    },
    {
      "epoch": 6.2161677135166,
      "grad_norm": 0.8147845268249512,
      "learning_rate": 2.766710808310952e-07,
      "loss": 0.1574,
      "num_input_tokens_seen": 72203568,
      "step": 38105
    },
    {
      "epoch": 6.216983440737417,
      "grad_norm": 0.10258424282073975,
      "learning_rate": 2.7521644903242827e-07,
      "loss": 0.2129,
      "num_input_tokens_seen": 72213072,
      "step": 38110
    },
    {
      "epoch": 6.217799167958235,
      "grad_norm": 1.716756820678711,
      "learning_rate": 2.7376563010609593e-07,
      "loss": 0.4277,
      "num_input_tokens_seen": 72221792,
      "step": 38115
    },
    {
      "epoch": 6.218614895179052,
      "grad_norm": 0.21596165001392365,
      "learning_rate": 2.72318624275833e-07,
      "loss": 0.1758,
      "num_input_tokens_seen": 72231216,
      "step": 38120
    },
    {
      "epoch": 6.219430622399869,
      "grad_norm": 0.8537912964820862,
      "learning_rate": 2.7087543176478324e-07,
      "loss": 0.1095,
      "num_input_tokens_seen": 72241200,
      "step": 38125
    },
    {
      "epoch": 6.220246349620687,
      "grad_norm": 1.6270934343338013,
      "learning_rate": 2.694360527955103e-07,
      "loss": 0.0997,
      "num_input_tokens_seen": 72250096,
      "step": 38130
    },
    {
      "epoch": 6.221062076841505,
      "grad_norm": 1.9839751720428467,
      "learning_rate": 2.680004875899811e-07,
      "loss": 0.1042,
      "num_input_tokens_seen": 72260176,
      "step": 38135
    },
    {
      "epoch": 6.221877804062322,
      "grad_norm": 1.4074797630310059,
      "learning_rate": 2.665687363695768e-07,
      "loss": 0.1366,
      "num_input_tokens_seen": 72269200,
      "step": 38140
    },
    {
      "epoch": 6.222693531283139,
      "grad_norm": 2.8455114364624023,
      "learning_rate": 2.6514079935509584e-07,
      "loss": 0.2678,
      "num_input_tokens_seen": 72278752,
      "step": 38145
    },
    {
      "epoch": 6.223509258503956,
      "grad_norm": 1.8379602432250977,
      "learning_rate": 2.6371667676673983e-07,
      "loss": 0.1435,
      "num_input_tokens_seen": 72288464,
      "step": 38150
    },
    {
      "epoch": 6.224324985724774,
      "grad_norm": 0.7032281756401062,
      "learning_rate": 2.6229636882412755e-07,
      "loss": 0.1105,
      "num_input_tokens_seen": 72298144,
      "step": 38155
    },
    {
      "epoch": 6.225140712945591,
      "grad_norm": 1.0240163803100586,
      "learning_rate": 2.6087987574628935e-07,
      "loss": 0.038,
      "num_input_tokens_seen": 72307824,
      "step": 38160
    },
    {
      "epoch": 6.225956440166408,
      "grad_norm": 0.2201271802186966,
      "learning_rate": 2.5946719775166437e-07,
      "loss": 0.0618,
      "num_input_tokens_seen": 72316880,
      "step": 38165
    },
    {
      "epoch": 6.226772167387225,
      "grad_norm": 0.48371613025665283,
      "learning_rate": 2.5805833505810616e-07,
      "loss": 0.1834,
      "num_input_tokens_seen": 72327056,
      "step": 38170
    },
    {
      "epoch": 6.227587894608043,
      "grad_norm": 0.11340699344873428,
      "learning_rate": 2.566532878828798e-07,
      "loss": 0.0904,
      "num_input_tokens_seen": 72337296,
      "step": 38175
    },
    {
      "epoch": 6.2284036218288605,
      "grad_norm": 0.17893923819065094,
      "learning_rate": 2.552520564426619e-07,
      "loss": 0.0881,
      "num_input_tokens_seen": 72347568,
      "step": 38180
    },
    {
      "epoch": 6.229219349049678,
      "grad_norm": 1.0496026277542114,
      "learning_rate": 2.5385464095353803e-07,
      "loss": 0.1474,
      "num_input_tokens_seen": 72356608,
      "step": 38185
    },
    {
      "epoch": 6.230035076270495,
      "grad_norm": 0.2232980579137802,
      "learning_rate": 2.5246104163100804e-07,
      "loss": 0.1745,
      "num_input_tokens_seen": 72366048,
      "step": 38190
    },
    {
      "epoch": 6.230850803491313,
      "grad_norm": 0.32148173451423645,
      "learning_rate": 2.510712586899833e-07,
      "loss": 0.1077,
      "num_input_tokens_seen": 72376112,
      "step": 38195
    },
    {
      "epoch": 6.23166653071213,
      "grad_norm": 1.8081754446029663,
      "learning_rate": 2.4968529234478124e-07,
      "loss": 0.3545,
      "num_input_tokens_seen": 72385776,
      "step": 38200
    },
    {
      "epoch": 6.23166653071213,
      "eval_loss": 0.14212387800216675,
      "eval_runtime": 133.2353,
      "eval_samples_per_second": 20.453,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 72385776,
      "step": 38200
    },
    {
      "epoch": 6.232482257932947,
      "grad_norm": 0.23326674103736877,
      "learning_rate": 2.483031428091448e-07,
      "loss": 0.0804,
      "num_input_tokens_seen": 72395168,
      "step": 38205
    },
    {
      "epoch": 6.233297985153764,
      "grad_norm": 0.13333798944950104,
      "learning_rate": 2.469248102962091e-07,
      "loss": 0.1274,
      "num_input_tokens_seen": 72405088,
      "step": 38210
    },
    {
      "epoch": 6.234113712374582,
      "grad_norm": 0.048954401165246964,
      "learning_rate": 2.4555029501853455e-07,
      "loss": 0.1741,
      "num_input_tokens_seen": 72414816,
      "step": 38215
    },
    {
      "epoch": 6.234929439595399,
      "grad_norm": 0.03028573840856552,
      "learning_rate": 2.441795971880906e-07,
      "loss": 0.1506,
      "num_input_tokens_seen": 72424528,
      "step": 38220
    },
    {
      "epoch": 6.2357451668162165,
      "grad_norm": 0.6273635029792786,
      "learning_rate": 2.4281271701625255e-07,
      "loss": 0.0691,
      "num_input_tokens_seen": 72434208,
      "step": 38225
    },
    {
      "epoch": 6.236560894037034,
      "grad_norm": 0.3699606955051422,
      "learning_rate": 2.4144965471381007e-07,
      "loss": 0.1744,
      "num_input_tokens_seen": 72444320,
      "step": 38230
    },
    {
      "epoch": 6.237376621257852,
      "grad_norm": 0.09420964866876602,
      "learning_rate": 2.400904104909674e-07,
      "loss": 0.0596,
      "num_input_tokens_seen": 72453984,
      "step": 38235
    },
    {
      "epoch": 6.238192348478669,
      "grad_norm": 0.8835181593894958,
      "learning_rate": 2.3873498455733725e-07,
      "loss": 0.1696,
      "num_input_tokens_seen": 72463328,
      "step": 38240
    },
    {
      "epoch": 6.239008075699486,
      "grad_norm": 0.6612083315849304,
      "learning_rate": 2.3738337712194137e-07,
      "loss": 0.1233,
      "num_input_tokens_seen": 72472560,
      "step": 38245
    },
    {
      "epoch": 6.239823802920303,
      "grad_norm": 0.02799399010837078,
      "learning_rate": 2.3603558839321305e-07,
      "loss": 0.0744,
      "num_input_tokens_seen": 72482448,
      "step": 38250
    },
    {
      "epoch": 6.240639530141121,
      "grad_norm": 1.9031734466552734,
      "learning_rate": 2.3469161857900267e-07,
      "loss": 0.1909,
      "num_input_tokens_seen": 72491920,
      "step": 38255
    },
    {
      "epoch": 6.241455257361938,
      "grad_norm": 2.2989277839660645,
      "learning_rate": 2.3335146788656393e-07,
      "loss": 0.186,
      "num_input_tokens_seen": 72502064,
      "step": 38260
    },
    {
      "epoch": 6.242270984582755,
      "grad_norm": 0.06889757513999939,
      "learning_rate": 2.3201513652256757e-07,
      "loss": 0.0376,
      "num_input_tokens_seen": 72511328,
      "step": 38265
    },
    {
      "epoch": 6.243086711803572,
      "grad_norm": 1.1497547626495361,
      "learning_rate": 2.3068262469308766e-07,
      "loss": 0.1539,
      "num_input_tokens_seen": 72519856,
      "step": 38270
    },
    {
      "epoch": 6.2439024390243905,
      "grad_norm": 0.48594847321510315,
      "learning_rate": 2.2935393260362093e-07,
      "loss": 0.1528,
      "num_input_tokens_seen": 72529648,
      "step": 38275
    },
    {
      "epoch": 6.244718166245208,
      "grad_norm": 0.0698729157447815,
      "learning_rate": 2.2802906045906458e-07,
      "loss": 0.1152,
      "num_input_tokens_seen": 72538384,
      "step": 38280
    },
    {
      "epoch": 6.245533893466025,
      "grad_norm": 0.47095203399658203,
      "learning_rate": 2.2670800846373018e-07,
      "loss": 0.0731,
      "num_input_tokens_seen": 72546512,
      "step": 38285
    },
    {
      "epoch": 6.246349620686843,
      "grad_norm": 0.4885791838169098,
      "learning_rate": 2.2539077682134367e-07,
      "loss": 0.164,
      "num_input_tokens_seen": 72556816,
      "step": 38290
    },
    {
      "epoch": 6.24716534790766,
      "grad_norm": 0.6998597383499146,
      "learning_rate": 2.2407736573503423e-07,
      "loss": 0.0838,
      "num_input_tokens_seen": 72566880,
      "step": 38295
    },
    {
      "epoch": 6.247981075128477,
      "grad_norm": 0.18221309781074524,
      "learning_rate": 2.2276777540735093e-07,
      "loss": 0.0583,
      "num_input_tokens_seen": 72577376,
      "step": 38300
    },
    {
      "epoch": 6.248796802349294,
      "grad_norm": 0.690430223941803,
      "learning_rate": 2.2146200604024613e-07,
      "loss": 0.0876,
      "num_input_tokens_seen": 72587664,
      "step": 38305
    },
    {
      "epoch": 6.249612529570112,
      "grad_norm": 0.22779859602451324,
      "learning_rate": 2.2016005783508375e-07,
      "loss": 0.043,
      "num_input_tokens_seen": 72596784,
      "step": 38310
    },
    {
      "epoch": 6.250428256790929,
      "grad_norm": 0.45053011178970337,
      "learning_rate": 2.1886193099264763e-07,
      "loss": 0.1825,
      "num_input_tokens_seen": 72605920,
      "step": 38315
    },
    {
      "epoch": 6.251243984011746,
      "grad_norm": 0.14100301265716553,
      "learning_rate": 2.175676257131165e-07,
      "loss": 0.103,
      "num_input_tokens_seen": 72616672,
      "step": 38320
    },
    {
      "epoch": 6.2520597112325635,
      "grad_norm": 1.333191990852356,
      "learning_rate": 2.162771421960974e-07,
      "loss": 0.1558,
      "num_input_tokens_seen": 72626176,
      "step": 38325
    },
    {
      "epoch": 6.252875438453382,
      "grad_norm": 0.6776754260063171,
      "learning_rate": 2.1499048064059224e-07,
      "loss": 0.084,
      "num_input_tokens_seen": 72635792,
      "step": 38330
    },
    {
      "epoch": 6.253691165674199,
      "grad_norm": 0.2874036729335785,
      "learning_rate": 2.1370764124502285e-07,
      "loss": 0.0807,
      "num_input_tokens_seen": 72643616,
      "step": 38335
    },
    {
      "epoch": 6.254506892895016,
      "grad_norm": 1.0367063283920288,
      "learning_rate": 2.1242862420721988e-07,
      "loss": 0.1056,
      "num_input_tokens_seen": 72653952,
      "step": 38340
    },
    {
      "epoch": 6.255322620115833,
      "grad_norm": 1.810220718383789,
      "learning_rate": 2.1115342972442276e-07,
      "loss": 0.1613,
      "num_input_tokens_seen": 72665152,
      "step": 38345
    },
    {
      "epoch": 6.256138347336651,
      "grad_norm": 0.9676206707954407,
      "learning_rate": 2.0988205799328252e-07,
      "loss": 0.0806,
      "num_input_tokens_seen": 72674208,
      "step": 38350
    },
    {
      "epoch": 6.256954074557468,
      "grad_norm": 0.24115803837776184,
      "learning_rate": 2.0861450920986182e-07,
      "loss": 0.1077,
      "num_input_tokens_seen": 72684112,
      "step": 38355
    },
    {
      "epoch": 6.257769801778285,
      "grad_norm": 0.1419493705034256,
      "learning_rate": 2.07350783569632e-07,
      "loss": 0.024,
      "num_input_tokens_seen": 72694208,
      "step": 38360
    },
    {
      "epoch": 6.258585528999102,
      "grad_norm": 0.31126636266708374,
      "learning_rate": 2.060908812674761e-07,
      "loss": 0.0415,
      "num_input_tokens_seen": 72703888,
      "step": 38365
    },
    {
      "epoch": 6.25940125621992,
      "grad_norm": 1.4864803552627563,
      "learning_rate": 2.0483480249768317e-07,
      "loss": 0.148,
      "num_input_tokens_seen": 72713424,
      "step": 38370
    },
    {
      "epoch": 6.2602169834407375,
      "grad_norm": 0.3697473704814911,
      "learning_rate": 2.035825474539621e-07,
      "loss": 0.0772,
      "num_input_tokens_seen": 72723520,
      "step": 38375
    },
    {
      "epoch": 6.261032710661555,
      "grad_norm": 1.2590492963790894,
      "learning_rate": 2.0233411632942235e-07,
      "loss": 0.1079,
      "num_input_tokens_seen": 72732304,
      "step": 38380
    },
    {
      "epoch": 6.261848437882372,
      "grad_norm": 0.4732646644115448,
      "learning_rate": 2.0108950931658764e-07,
      "loss": 0.0428,
      "num_input_tokens_seen": 72743184,
      "step": 38385
    },
    {
      "epoch": 6.26266416510319,
      "grad_norm": 1.2285159826278687,
      "learning_rate": 1.998487266073934e-07,
      "loss": 0.1947,
      "num_input_tokens_seen": 72752496,
      "step": 38390
    },
    {
      "epoch": 6.263479892324007,
      "grad_norm": 0.09808298945426941,
      "learning_rate": 1.986117683931865e-07,
      "loss": 0.0569,
      "num_input_tokens_seen": 72762608,
      "step": 38395
    },
    {
      "epoch": 6.264295619544824,
      "grad_norm": 0.5511799454689026,
      "learning_rate": 1.9737863486471442e-07,
      "loss": 0.1921,
      "num_input_tokens_seen": 72773152,
      "step": 38400
    },
    {
      "epoch": 6.264295619544824,
      "eval_loss": 0.14221666753292084,
      "eval_runtime": 133.2701,
      "eval_samples_per_second": 20.447,
      "eval_steps_per_second": 10.227,
      "num_input_tokens_seen": 72773152,
      "step": 38400
    },
    {
      "epoch": 6.265111346765641,
      "grad_norm": 0.5776437520980835,
      "learning_rate": 1.9614932621215e-07,
      "loss": 0.0773,
      "num_input_tokens_seen": 72781648,
      "step": 38405
    },
    {
      "epoch": 6.265927073986459,
      "grad_norm": 0.9981775879859924,
      "learning_rate": 1.9492384262506102e-07,
      "loss": 0.1768,
      "num_input_tokens_seen": 72790384,
      "step": 38410
    },
    {
      "epoch": 6.266742801207276,
      "grad_norm": 0.21877075731754303,
      "learning_rate": 1.9370218429243524e-07,
      "loss": 0.0666,
      "num_input_tokens_seen": 72800256,
      "step": 38415
    },
    {
      "epoch": 6.2675585284280935,
      "grad_norm": 1.4226486682891846,
      "learning_rate": 1.9248435140267197e-07,
      "loss": 0.2054,
      "num_input_tokens_seen": 72810016,
      "step": 38420
    },
    {
      "epoch": 6.268374255648911,
      "grad_norm": 0.7326995730400085,
      "learning_rate": 1.9127034414356814e-07,
      "loss": 0.1341,
      "num_input_tokens_seen": 72819008,
      "step": 38425
    },
    {
      "epoch": 6.269189982869729,
      "grad_norm": 0.5726404786109924,
      "learning_rate": 1.9006016270234627e-07,
      "loss": 0.1171,
      "num_input_tokens_seen": 72827888,
      "step": 38430
    },
    {
      "epoch": 6.270005710090546,
      "grad_norm": 0.9986951947212219,
      "learning_rate": 1.888538072656293e-07,
      "loss": 0.1239,
      "num_input_tokens_seen": 72838752,
      "step": 38435
    },
    {
      "epoch": 6.270821437311363,
      "grad_norm": 0.9122962951660156,
      "learning_rate": 1.8765127801944893e-07,
      "loss": 0.0638,
      "num_input_tokens_seen": 72847648,
      "step": 38440
    },
    {
      "epoch": 6.27163716453218,
      "grad_norm": 1.4400498867034912,
      "learning_rate": 1.8645257514925406e-07,
      "loss": 0.1296,
      "num_input_tokens_seen": 72857344,
      "step": 38445
    },
    {
      "epoch": 6.272452891752998,
      "grad_norm": 0.09006964415311813,
      "learning_rate": 1.8525769883989685e-07,
      "loss": 0.0696,
      "num_input_tokens_seen": 72867232,
      "step": 38450
    },
    {
      "epoch": 6.273268618973815,
      "grad_norm": 0.6458718180656433,
      "learning_rate": 1.8406664927564654e-07,
      "loss": 0.1611,
      "num_input_tokens_seen": 72875664,
      "step": 38455
    },
    {
      "epoch": 6.274084346194632,
      "grad_norm": 0.5878198742866516,
      "learning_rate": 1.8287942664017566e-07,
      "loss": 0.2022,
      "num_input_tokens_seen": 72884992,
      "step": 38460
    },
    {
      "epoch": 6.27490007341545,
      "grad_norm": 0.40876176953315735,
      "learning_rate": 1.8169603111656552e-07,
      "loss": 0.1933,
      "num_input_tokens_seen": 72893536,
      "step": 38465
    },
    {
      "epoch": 6.275715800636267,
      "grad_norm": 0.7417309284210205,
      "learning_rate": 1.805164628873146e-07,
      "loss": 0.2318,
      "num_input_tokens_seen": 72902768,
      "step": 38470
    },
    {
      "epoch": 6.276531527857085,
      "grad_norm": 0.3236558139324188,
      "learning_rate": 1.793407221343274e-07,
      "loss": 0.2454,
      "num_input_tokens_seen": 72912240,
      "step": 38475
    },
    {
      "epoch": 6.277347255077902,
      "grad_norm": 0.7260608673095703,
      "learning_rate": 1.781688090389172e-07,
      "loss": 0.1078,
      "num_input_tokens_seen": 72920368,
      "step": 38480
    },
    {
      "epoch": 6.27816298229872,
      "grad_norm": 0.3197302520275116,
      "learning_rate": 1.770007237818061e-07,
      "loss": 0.1093,
      "num_input_tokens_seen": 72930720,
      "step": 38485
    },
    {
      "epoch": 6.278978709519537,
      "grad_norm": 0.9360746741294861,
      "learning_rate": 1.7583646654313059e-07,
      "loss": 0.209,
      "num_input_tokens_seen": 72940048,
      "step": 38490
    },
    {
      "epoch": 6.279794436740354,
      "grad_norm": 0.5055411458015442,
      "learning_rate": 1.7467603750242757e-07,
      "loss": 0.1877,
      "num_input_tokens_seen": 72949552,
      "step": 38495
    },
    {
      "epoch": 6.280610163961171,
      "grad_norm": 0.6629536747932434,
      "learning_rate": 1.7351943683865944e-07,
      "loss": 0.0716,
      "num_input_tokens_seen": 72960256,
      "step": 38500
    },
    {
      "epoch": 6.281425891181989,
      "grad_norm": 0.16449423134326935,
      "learning_rate": 1.723666647301808e-07,
      "loss": 0.0876,
      "num_input_tokens_seen": 72969680,
      "step": 38505
    },
    {
      "epoch": 6.282241618402806,
      "grad_norm": 1.1054103374481201,
      "learning_rate": 1.712177213547661e-07,
      "loss": 0.1374,
      "num_input_tokens_seen": 72978272,
      "step": 38510
    },
    {
      "epoch": 6.283057345623623,
      "grad_norm": 0.7971771955490112,
      "learning_rate": 1.7007260688959581e-07,
      "loss": 0.1042,
      "num_input_tokens_seen": 72988112,
      "step": 38515
    },
    {
      "epoch": 6.2838730728444405,
      "grad_norm": 2.1793344020843506,
      "learning_rate": 1.68931321511262e-07,
      "loss": 0.1756,
      "num_input_tokens_seen": 72997104,
      "step": 38520
    },
    {
      "epoch": 6.2846888000652585,
      "grad_norm": 0.7732468247413635,
      "learning_rate": 1.6779386539576835e-07,
      "loss": 0.1508,
      "num_input_tokens_seen": 73005984,
      "step": 38525
    },
    {
      "epoch": 6.285504527286076,
      "grad_norm": 0.23688241839408875,
      "learning_rate": 1.666602387185162e-07,
      "loss": 0.0867,
      "num_input_tokens_seen": 73015168,
      "step": 38530
    },
    {
      "epoch": 6.286320254506893,
      "grad_norm": 0.9964271783828735,
      "learning_rate": 1.655304416543352e-07,
      "loss": 0.2855,
      "num_input_tokens_seen": 73023952,
      "step": 38535
    },
    {
      "epoch": 6.28713598172771,
      "grad_norm": 0.8577680587768555,
      "learning_rate": 1.6440447437744698e-07,
      "loss": 0.1345,
      "num_input_tokens_seen": 73033296,
      "step": 38540
    },
    {
      "epoch": 6.287951708948528,
      "grad_norm": 0.9155134558677673,
      "learning_rate": 1.6328233706149332e-07,
      "loss": 0.2502,
      "num_input_tokens_seen": 73043280,
      "step": 38545
    },
    {
      "epoch": 6.288767436169345,
      "grad_norm": 0.05405459180474281,
      "learning_rate": 1.6216402987951906e-07,
      "loss": 0.1917,
      "num_input_tokens_seen": 73052192,
      "step": 38550
    },
    {
      "epoch": 6.289583163390162,
      "grad_norm": 0.7410343289375305,
      "learning_rate": 1.6104955300398627e-07,
      "loss": 0.085,
      "num_input_tokens_seen": 73062128,
      "step": 38555
    },
    {
      "epoch": 6.290398890610979,
      "grad_norm": 0.1767803579568863,
      "learning_rate": 1.5993890660675748e-07,
      "loss": 0.1642,
      "num_input_tokens_seen": 73071792,
      "step": 38560
    },
    {
      "epoch": 6.291214617831797,
      "grad_norm": 2.0544121265411377,
      "learning_rate": 1.5883209085910678e-07,
      "loss": 0.2384,
      "num_input_tokens_seen": 73081760,
      "step": 38565
    },
    {
      "epoch": 6.2920303450526145,
      "grad_norm": 0.0878584086894989,
      "learning_rate": 1.5772910593172264e-07,
      "loss": 0.1324,
      "num_input_tokens_seen": 73091120,
      "step": 38570
    },
    {
      "epoch": 6.292846072273432,
      "grad_norm": 0.3535337448120117,
      "learning_rate": 1.5662995199469954e-07,
      "loss": 0.1252,
      "num_input_tokens_seen": 73100640,
      "step": 38575
    },
    {
      "epoch": 6.293661799494249,
      "grad_norm": 0.0826416090130806,
      "learning_rate": 1.5553462921753802e-07,
      "loss": 0.0179,
      "num_input_tokens_seen": 73109392,
      "step": 38580
    },
    {
      "epoch": 6.294477526715067,
      "grad_norm": 1.4827011823654175,
      "learning_rate": 1.544431377691502e-07,
      "loss": 0.2572,
      "num_input_tokens_seen": 73118320,
      "step": 38585
    },
    {
      "epoch": 6.295293253935884,
      "grad_norm": 0.03591214865446091,
      "learning_rate": 1.5335547781785975e-07,
      "loss": 0.0573,
      "num_input_tokens_seen": 73128000,
      "step": 38590
    },
    {
      "epoch": 6.296108981156701,
      "grad_norm": 0.8878619074821472,
      "learning_rate": 1.5227164953139917e-07,
      "loss": 0.2867,
      "num_input_tokens_seen": 73139600,
      "step": 38595
    },
    {
      "epoch": 6.296924708377518,
      "grad_norm": 0.3681079149246216,
      "learning_rate": 1.511916530769042e-07,
      "loss": 0.1193,
      "num_input_tokens_seen": 73149584,
      "step": 38600
    },
    {
      "epoch": 6.296924708377518,
      "eval_loss": 0.1420513540506363,
      "eval_runtime": 133.3736,
      "eval_samples_per_second": 20.431,
      "eval_steps_per_second": 10.219,
      "num_input_tokens_seen": 73149584,
      "step": 38600
    },
    {
      "epoch": 6.297740435598336,
      "grad_norm": 0.545731246471405,
      "learning_rate": 1.5011548862092773e-07,
      "loss": 0.1522,
      "num_input_tokens_seen": 73158400,
      "step": 38605
    },
    {
      "epoch": 6.298556162819153,
      "grad_norm": 1.1319868564605713,
      "learning_rate": 1.490431563294231e-07,
      "loss": 0.1198,
      "num_input_tokens_seen": 73168528,
      "step": 38610
    },
    {
      "epoch": 6.2993718900399704,
      "grad_norm": 0.046994391828775406,
      "learning_rate": 1.4797465636776365e-07,
      "loss": 0.1705,
      "num_input_tokens_seen": 73178320,
      "step": 38615
    },
    {
      "epoch": 6.300187617260788,
      "grad_norm": 1.478247880935669,
      "learning_rate": 1.4690998890072027e-07,
      "loss": 0.1798,
      "num_input_tokens_seen": 73188480,
      "step": 38620
    },
    {
      "epoch": 6.301003344481606,
      "grad_norm": 2.0431487560272217,
      "learning_rate": 1.4584915409248112e-07,
      "loss": 0.1725,
      "num_input_tokens_seen": 73197360,
      "step": 38625
    },
    {
      "epoch": 6.301819071702423,
      "grad_norm": 1.1662578582763672,
      "learning_rate": 1.4479215210663754e-07,
      "loss": 0.1346,
      "num_input_tokens_seen": 73207184,
      "step": 38630
    },
    {
      "epoch": 6.30263479892324,
      "grad_norm": 0.25793272256851196,
      "learning_rate": 1.4373898310619528e-07,
      "loss": 0.0992,
      "num_input_tokens_seen": 73215952,
      "step": 38635
    },
    {
      "epoch": 6.303450526144058,
      "grad_norm": 0.20443491637706757,
      "learning_rate": 1.4268964725356604e-07,
      "loss": 0.2594,
      "num_input_tokens_seen": 73225552,
      "step": 38640
    },
    {
      "epoch": 6.304266253364875,
      "grad_norm": 0.5270785689353943,
      "learning_rate": 1.4164414471056764e-07,
      "loss": 0.0927,
      "num_input_tokens_seen": 73234048,
      "step": 38645
    },
    {
      "epoch": 6.305081980585692,
      "grad_norm": 0.027528854086995125,
      "learning_rate": 1.4060247563843497e-07,
      "loss": 0.0787,
      "num_input_tokens_seen": 73243600,
      "step": 38650
    },
    {
      "epoch": 6.305897707806509,
      "grad_norm": 0.39360591769218445,
      "learning_rate": 1.3956464019780068e-07,
      "loss": 0.1009,
      "num_input_tokens_seen": 73252624,
      "step": 38655
    },
    {
      "epoch": 6.306713435027326,
      "grad_norm": 0.15090441703796387,
      "learning_rate": 1.385306385487145e-07,
      "loss": 0.055,
      "num_input_tokens_seen": 73261280,
      "step": 38660
    },
    {
      "epoch": 6.307529162248144,
      "grad_norm": 0.09036707133054733,
      "learning_rate": 1.3750047085063222e-07,
      "loss": 0.0706,
      "num_input_tokens_seen": 73270912,
      "step": 38665
    },
    {
      "epoch": 6.3083448894689615,
      "grad_norm": 0.211425319314003,
      "learning_rate": 1.3647413726242119e-07,
      "loss": 0.0972,
      "num_input_tokens_seen": 73280000,
      "step": 38670
    },
    {
      "epoch": 6.309160616689779,
      "grad_norm": 2.918095827102661,
      "learning_rate": 1.3545163794235205e-07,
      "loss": 0.3248,
      "num_input_tokens_seen": 73289936,
      "step": 38675
    },
    {
      "epoch": 6.309976343910597,
      "grad_norm": 0.1975269466638565,
      "learning_rate": 1.3443297304810698e-07,
      "loss": 0.0349,
      "num_input_tokens_seen": 73299360,
      "step": 38680
    },
    {
      "epoch": 6.310792071131414,
      "grad_norm": 0.05121452733874321,
      "learning_rate": 1.3341814273677977e-07,
      "loss": 0.1243,
      "num_input_tokens_seen": 73307696,
      "step": 38685
    },
    {
      "epoch": 6.311607798352231,
      "grad_norm": 1.894829273223877,
      "learning_rate": 1.324071471648647e-07,
      "loss": 0.1754,
      "num_input_tokens_seen": 73317088,
      "step": 38690
    },
    {
      "epoch": 6.312423525573048,
      "grad_norm": 0.871855616569519,
      "learning_rate": 1.3139998648827312e-07,
      "loss": 0.1066,
      "num_input_tokens_seen": 73326592,
      "step": 38695
    },
    {
      "epoch": 6.313239252793866,
      "grad_norm": 0.2161187380552292,
      "learning_rate": 1.3039666086232526e-07,
      "loss": 0.1588,
      "num_input_tokens_seen": 73337040,
      "step": 38700
    },
    {
      "epoch": 6.314054980014683,
      "grad_norm": 0.6917955279350281,
      "learning_rate": 1.2939717044174183e-07,
      "loss": 0.067,
      "num_input_tokens_seen": 73345600,
      "step": 38705
    },
    {
      "epoch": 6.3148707072355,
      "grad_norm": 0.7773209810256958,
      "learning_rate": 1.284015153806578e-07,
      "loss": 0.1102,
      "num_input_tokens_seen": 73352912,
      "step": 38710
    },
    {
      "epoch": 6.3156864344563175,
      "grad_norm": 0.07757235318422318,
      "learning_rate": 1.274096958326171e-07,
      "loss": 0.0705,
      "num_input_tokens_seen": 73362384,
      "step": 38715
    },
    {
      "epoch": 6.3165021616771355,
      "grad_norm": 0.05278458818793297,
      "learning_rate": 1.2642171195056952e-07,
      "loss": 0.1208,
      "num_input_tokens_seen": 73370848,
      "step": 38720
    },
    {
      "epoch": 6.317317888897953,
      "grad_norm": 0.3576918840408325,
      "learning_rate": 1.2543756388687377e-07,
      "loss": 0.0741,
      "num_input_tokens_seen": 73380192,
      "step": 38725
    },
    {
      "epoch": 6.31813361611877,
      "grad_norm": 1.5314401388168335,
      "learning_rate": 1.2445725179330014e-07,
      "loss": 0.0928,
      "num_input_tokens_seen": 73389936,
      "step": 38730
    },
    {
      "epoch": 6.318949343339587,
      "grad_norm": 2.13623046875,
      "learning_rate": 1.2348077582102212e-07,
      "loss": 0.0676,
      "num_input_tokens_seen": 73399536,
      "step": 38735
    },
    {
      "epoch": 6.319765070560405,
      "grad_norm": 0.5554681420326233,
      "learning_rate": 1.2250813612062762e-07,
      "loss": 0.1019,
      "num_input_tokens_seen": 73409168,
      "step": 38740
    },
    {
      "epoch": 6.320580797781222,
      "grad_norm": 0.18778342008590698,
      "learning_rate": 1.215393328421105e-07,
      "loss": 0.0203,
      "num_input_tokens_seen": 73418272,
      "step": 38745
    },
    {
      "epoch": 6.321396525002039,
      "grad_norm": 0.5777256488800049,
      "learning_rate": 1.2057436613486796e-07,
      "loss": 0.128,
      "num_input_tokens_seen": 73426992,
      "step": 38750
    },
    {
      "epoch": 6.322212252222856,
      "grad_norm": 0.7211507558822632,
      "learning_rate": 1.1961323614771424e-07,
      "loss": 0.081,
      "num_input_tokens_seen": 73436784,
      "step": 38755
    },
    {
      "epoch": 6.323027979443674,
      "grad_norm": 0.45646482706069946,
      "learning_rate": 1.1865594302886418e-07,
      "loss": 0.1043,
      "num_input_tokens_seen": 73446096,
      "step": 38760
    },
    {
      "epoch": 6.3238437066644915,
      "grad_norm": 0.09077994525432587,
      "learning_rate": 1.1770248692594687e-07,
      "loss": 0.0632,
      "num_input_tokens_seen": 73454304,
      "step": 38765
    },
    {
      "epoch": 6.324659433885309,
      "grad_norm": 0.12476957589387894,
      "learning_rate": 1.167528679859975e-07,
      "loss": 0.1115,
      "num_input_tokens_seen": 73463264,
      "step": 38770
    },
    {
      "epoch": 6.325475161106126,
      "grad_norm": 0.4043148159980774,
      "learning_rate": 1.1580708635545446e-07,
      "loss": 0.0683,
      "num_input_tokens_seen": 73472688,
      "step": 38775
    },
    {
      "epoch": 6.326290888326944,
      "grad_norm": 0.21676714718341827,
      "learning_rate": 1.1486514218017885e-07,
      "loss": 0.2176,
      "num_input_tokens_seen": 73482592,
      "step": 38780
    },
    {
      "epoch": 6.327106615547761,
      "grad_norm": 0.6733180284500122,
      "learning_rate": 1.1392703560542117e-07,
      "loss": 0.0578,
      "num_input_tokens_seen": 73492112,
      "step": 38785
    },
    {
      "epoch": 6.327922342768578,
      "grad_norm": 1.3904603719711304,
      "learning_rate": 1.129927667758518e-07,
      "loss": 0.2161,
      "num_input_tokens_seen": 73501568,
      "step": 38790
    },
    {
      "epoch": 6.328738069989395,
      "grad_norm": 1.0160056352615356,
      "learning_rate": 1.1206233583554992e-07,
      "loss": 0.296,
      "num_input_tokens_seen": 73511136,
      "step": 38795
    },
    {
      "epoch": 6.329553797210213,
      "grad_norm": 0.2057114541530609,
      "learning_rate": 1.1113574292799523e-07,
      "loss": 0.187,
      "num_input_tokens_seen": 73519536,
      "step": 38800
    },
    {
      "epoch": 6.329553797210213,
      "eval_loss": 0.14192408323287964,
      "eval_runtime": 133.2223,
      "eval_samples_per_second": 20.455,
      "eval_steps_per_second": 10.231,
      "num_input_tokens_seen": 73519536,
      "step": 38800
    },
    {
      "epoch": 6.33036952443103,
      "grad_norm": 1.1585243940353394,
      "learning_rate": 1.1021298819608449e-07,
      "loss": 0.1634,
      "num_input_tokens_seen": 73529360,
      "step": 38805
    },
    {
      "epoch": 6.331185251651847,
      "grad_norm": 0.06416064500808716,
      "learning_rate": 1.0929407178211226e-07,
      "loss": 0.1532,
      "num_input_tokens_seen": 73538288,
      "step": 38810
    },
    {
      "epoch": 6.332000978872665,
      "grad_norm": 0.048609793186187744,
      "learning_rate": 1.0837899382779293e-07,
      "loss": 0.1029,
      "num_input_tokens_seen": 73547840,
      "step": 38815
    },
    {
      "epoch": 6.332816706093483,
      "grad_norm": 0.714495062828064,
      "learning_rate": 1.0746775447423862e-07,
      "loss": 0.2121,
      "num_input_tokens_seen": 73556848,
      "step": 38820
    },
    {
      "epoch": 6.3336324333143,
      "grad_norm": 0.14860786497592926,
      "learning_rate": 1.0656035386197583e-07,
      "loss": 0.0907,
      "num_input_tokens_seen": 73567696,
      "step": 38825
    },
    {
      "epoch": 6.334448160535117,
      "grad_norm": 0.3958728611469269,
      "learning_rate": 1.0565679213093982e-07,
      "loss": 0.1268,
      "num_input_tokens_seen": 73577344,
      "step": 38830
    },
    {
      "epoch": 6.335263887755934,
      "grad_norm": 0.17576946318149567,
      "learning_rate": 1.0475706942046638e-07,
      "loss": 0.0299,
      "num_input_tokens_seen": 73588320,
      "step": 38835
    },
    {
      "epoch": 6.336079614976752,
      "grad_norm": 0.7411639094352722,
      "learning_rate": 1.0386118586930282e-07,
      "loss": 0.1891,
      "num_input_tokens_seen": 73597664,
      "step": 38840
    },
    {
      "epoch": 6.336895342197569,
      "grad_norm": 1.9227042198181152,
      "learning_rate": 1.0296914161561367e-07,
      "loss": 0.1378,
      "num_input_tokens_seen": 73608768,
      "step": 38845
    },
    {
      "epoch": 6.337711069418386,
      "grad_norm": 1.153663992881775,
      "learning_rate": 1.0208093679695552e-07,
      "loss": 0.1789,
      "num_input_tokens_seen": 73619392,
      "step": 38850
    },
    {
      "epoch": 6.338526796639204,
      "grad_norm": 0.2032003104686737,
      "learning_rate": 1.0119657155030493e-07,
      "loss": 0.1303,
      "num_input_tokens_seen": 73629312,
      "step": 38855
    },
    {
      "epoch": 6.339342523860021,
      "grad_norm": 0.17107754945755005,
      "learning_rate": 1.003160460120417e-07,
      "loss": 0.0399,
      "num_input_tokens_seen": 73639488,
      "step": 38860
    },
    {
      "epoch": 6.3401582510808385,
      "grad_norm": 1.3300440311431885,
      "learning_rate": 9.943936031795165e-08,
      "loss": 0.2102,
      "num_input_tokens_seen": 73649808,
      "step": 38865
    },
    {
      "epoch": 6.340973978301656,
      "grad_norm": 0.18885177373886108,
      "learning_rate": 9.856651460323219e-08,
      "loss": 0.1085,
      "num_input_tokens_seen": 73659680,
      "step": 38870
    },
    {
      "epoch": 6.341789705522474,
      "grad_norm": 1.1478657722473145,
      "learning_rate": 9.769750900248953e-08,
      "loss": 0.1205,
      "num_input_tokens_seen": 73668928,
      "step": 38875
    },
    {
      "epoch": 6.342605432743291,
      "grad_norm": 1.07404625415802,
      "learning_rate": 9.683234364973038e-08,
      "loss": 0.1219,
      "num_input_tokens_seen": 73676640,
      "step": 38880
    },
    {
      "epoch": 6.343421159964108,
      "grad_norm": 0.29480281472206116,
      "learning_rate": 9.597101867837854e-08,
      "loss": 0.0532,
      "num_input_tokens_seen": 73686784,
      "step": 38885
    },
    {
      "epoch": 6.344236887184925,
      "grad_norm": 0.28320470452308655,
      "learning_rate": 9.511353422125835e-08,
      "loss": 0.0688,
      "num_input_tokens_seen": 73695952,
      "step": 38890
    },
    {
      "epoch": 6.345052614405743,
      "grad_norm": 0.25711703300476074,
      "learning_rate": 9.42598904106029e-08,
      "loss": 0.0439,
      "num_input_tokens_seen": 73706640,
      "step": 38895
    },
    {
      "epoch": 6.34586834162656,
      "grad_norm": 0.18404114246368408,
      "learning_rate": 9.341008737806245e-08,
      "loss": 0.0586,
      "num_input_tokens_seen": 73716016,
      "step": 38900
    },
    {
      "epoch": 6.346684068847377,
      "grad_norm": 0.4460301995277405,
      "learning_rate": 9.256412525467661e-08,
      "loss": 0.0771,
      "num_input_tokens_seen": 73725856,
      "step": 38905
    },
    {
      "epoch": 6.3474997960681945,
      "grad_norm": 0.7360022068023682,
      "learning_rate": 9.172200417091326e-08,
      "loss": 0.1346,
      "num_input_tokens_seen": 73735984,
      "step": 38910
    },
    {
      "epoch": 6.3483155232890125,
      "grad_norm": 0.0724109634757042,
      "learning_rate": 9.088372425663239e-08,
      "loss": 0.0626,
      "num_input_tokens_seen": 73744464,
      "step": 38915
    },
    {
      "epoch": 6.34913125050983,
      "grad_norm": 0.022725611925125122,
      "learning_rate": 9.004928564110837e-08,
      "loss": 0.0759,
      "num_input_tokens_seen": 73753600,
      "step": 38920
    },
    {
      "epoch": 6.349946977730647,
      "grad_norm": 0.06582256406545639,
      "learning_rate": 8.92186884530244e-08,
      "loss": 0.1939,
      "num_input_tokens_seen": 73762736,
      "step": 38925
    },
    {
      "epoch": 6.350762704951464,
      "grad_norm": 1.1699482202529907,
      "learning_rate": 8.83919328204641e-08,
      "loss": 0.1351,
      "num_input_tokens_seen": 73772816,
      "step": 38930
    },
    {
      "epoch": 6.351578432172282,
      "grad_norm": 0.34690961241722107,
      "learning_rate": 8.756901887093105e-08,
      "loss": 0.1581,
      "num_input_tokens_seen": 73782896,
      "step": 38935
    },
    {
      "epoch": 6.352394159393099,
      "grad_norm": 1.406929850578308,
      "learning_rate": 8.674994673132098e-08,
      "loss": 0.0825,
      "num_input_tokens_seen": 73793216,
      "step": 38940
    },
    {
      "epoch": 6.353209886613916,
      "grad_norm": 0.9910104870796204,
      "learning_rate": 8.593471652794949e-08,
      "loss": 0.1186,
      "num_input_tokens_seen": 73802896,
      "step": 38945
    },
    {
      "epoch": 6.354025613834733,
      "grad_norm": 1.6288515329360962,
      "learning_rate": 8.512332838653548e-08,
      "loss": 0.2096,
      "num_input_tokens_seen": 73811440,
      "step": 38950
    },
    {
      "epoch": 6.354841341055551,
      "grad_norm": 0.22308458387851715,
      "learning_rate": 8.431578243220106e-08,
      "loss": 0.0435,
      "num_input_tokens_seen": 73819904,
      "step": 38955
    },
    {
      "epoch": 6.3556570682763684,
      "grad_norm": 0.9048522710800171,
      "learning_rate": 8.351207878948552e-08,
      "loss": 0.1986,
      "num_input_tokens_seen": 73828640,
      "step": 38960
    },
    {
      "epoch": 6.356472795497186,
      "grad_norm": 0.27561166882514954,
      "learning_rate": 8.271221758232583e-08,
      "loss": 0.2008,
      "num_input_tokens_seen": 73838320,
      "step": 38965
    },
    {
      "epoch": 6.357288522718003,
      "grad_norm": 0.4533836245536804,
      "learning_rate": 8.191619893407332e-08,
      "loss": 0.0365,
      "num_input_tokens_seen": 73848608,
      "step": 38970
    },
    {
      "epoch": 6.358104249938821,
      "grad_norm": 1.5957890748977661,
      "learning_rate": 8.112402296748534e-08,
      "loss": 0.1132,
      "num_input_tokens_seen": 73857488,
      "step": 38975
    },
    {
      "epoch": 6.358919977159638,
      "grad_norm": 2.165804386138916,
      "learning_rate": 8.033568980471973e-08,
      "loss": 0.257,
      "num_input_tokens_seen": 73865232,
      "step": 38980
    },
    {
      "epoch": 6.359735704380455,
      "grad_norm": 0.6997953057289124,
      "learning_rate": 7.955119956735146e-08,
      "loss": 0.0874,
      "num_input_tokens_seen": 73875200,
      "step": 38985
    },
    {
      "epoch": 6.360551431601272,
      "grad_norm": 1.2168582677841187,
      "learning_rate": 7.877055237636155e-08,
      "loss": 0.1028,
      "num_input_tokens_seen": 73885312,
      "step": 38990
    },
    {
      "epoch": 6.36136715882209,
      "grad_norm": 0.07564231008291245,
      "learning_rate": 7.79937483521287e-08,
      "loss": 0.1305,
      "num_input_tokens_seen": 73894768,
      "step": 38995
    },
    {
      "epoch": 6.362182886042907,
      "grad_norm": 0.14534322917461395,
      "learning_rate": 7.722078761444873e-08,
      "loss": 0.2218,
      "num_input_tokens_seen": 73902896,
      "step": 39000
    },
    {
      "epoch": 6.362182886042907,
      "eval_loss": 0.14146137237548828,
      "eval_runtime": 133.2248,
      "eval_samples_per_second": 20.454,
      "eval_steps_per_second": 10.231,
      "num_input_tokens_seen": 73902896,
      "step": 39000
    },
    {
      "epoch": 6.362998613263724,
      "grad_norm": 1.2341352701187134,
      "learning_rate": 7.645167028252631e-08,
      "loss": 0.1791,
      "num_input_tokens_seen": 73912688,
      "step": 39005
    },
    {
      "epoch": 6.3638143404845415,
      "grad_norm": 0.06972784548997879,
      "learning_rate": 7.568639647496379e-08,
      "loss": 0.054,
      "num_input_tokens_seen": 73921872,
      "step": 39010
    },
    {
      "epoch": 6.3646300677053596,
      "grad_norm": 0.5730539560317993,
      "learning_rate": 7.492496630977508e-08,
      "loss": 0.0615,
      "num_input_tokens_seen": 73931664,
      "step": 39015
    },
    {
      "epoch": 6.365445794926177,
      "grad_norm": 0.04938632994890213,
      "learning_rate": 7.416737990438571e-08,
      "loss": 0.1526,
      "num_input_tokens_seen": 73941344,
      "step": 39020
    },
    {
      "epoch": 6.366261522146994,
      "grad_norm": 0.11422107368707657,
      "learning_rate": 7.341363737562445e-08,
      "loss": 0.1393,
      "num_input_tokens_seen": 73950032,
      "step": 39025
    },
    {
      "epoch": 6.367077249367812,
      "grad_norm": 0.42052459716796875,
      "learning_rate": 7.266373883972887e-08,
      "loss": 0.027,
      "num_input_tokens_seen": 73959312,
      "step": 39030
    },
    {
      "epoch": 6.367892976588629,
      "grad_norm": 0.18611399829387665,
      "learning_rate": 7.191768441233981e-08,
      "loss": 0.0424,
      "num_input_tokens_seen": 73969152,
      "step": 39035
    },
    {
      "epoch": 6.368708703809446,
      "grad_norm": 0.6012230515480042,
      "learning_rate": 7.11754742085069e-08,
      "loss": 0.079,
      "num_input_tokens_seen": 73977472,
      "step": 39040
    },
    {
      "epoch": 6.369524431030263,
      "grad_norm": 0.8998752236366272,
      "learning_rate": 7.043710834269413e-08,
      "loss": 0.2546,
      "num_input_tokens_seen": 73987104,
      "step": 39045
    },
    {
      "epoch": 6.370340158251081,
      "grad_norm": 0.8639478087425232,
      "learning_rate": 6.970258692876319e-08,
      "loss": 0.186,
      "num_input_tokens_seen": 73995248,
      "step": 39050
    },
    {
      "epoch": 6.371155885471898,
      "grad_norm": 1.842309594154358,
      "learning_rate": 6.897191007998738e-08,
      "loss": 0.2102,
      "num_input_tokens_seen": 74004992,
      "step": 39055
    },
    {
      "epoch": 6.3719716126927155,
      "grad_norm": 0.18039393424987793,
      "learning_rate": 6.824507790904599e-08,
      "loss": 0.1628,
      "num_input_tokens_seen": 74014768,
      "step": 39060
    },
    {
      "epoch": 6.372787339913533,
      "grad_norm": 0.153465136885643,
      "learning_rate": 6.752209052802439e-08,
      "loss": 0.0585,
      "num_input_tokens_seen": 74024720,
      "step": 39065
    },
    {
      "epoch": 6.373603067134351,
      "grad_norm": 0.29446008801460266,
      "learning_rate": 6.680294804841946e-08,
      "loss": 0.1296,
      "num_input_tokens_seen": 74033920,
      "step": 39070
    },
    {
      "epoch": 6.374418794355168,
      "grad_norm": 0.25956887006759644,
      "learning_rate": 6.608765058112865e-08,
      "loss": 0.0477,
      "num_input_tokens_seen": 74042960,
      "step": 39075
    },
    {
      "epoch": 6.375234521575985,
      "grad_norm": 0.13848471641540527,
      "learning_rate": 6.537619823646368e-08,
      "loss": 0.0413,
      "num_input_tokens_seen": 74052928,
      "step": 39080
    },
    {
      "epoch": 6.376050248796802,
      "grad_norm": 0.0990334004163742,
      "learning_rate": 6.466859112413404e-08,
      "loss": 0.0622,
      "num_input_tokens_seen": 74062000,
      "step": 39085
    },
    {
      "epoch": 6.37686597601762,
      "grad_norm": 0.07827379554510117,
      "learning_rate": 6.39648293532663e-08,
      "loss": 0.0404,
      "num_input_tokens_seen": 74071456,
      "step": 39090
    },
    {
      "epoch": 6.377681703238437,
      "grad_norm": 0.3951127231121063,
      "learning_rate": 6.32649130323848e-08,
      "loss": 0.214,
      "num_input_tokens_seen": 74079728,
      "step": 39095
    },
    {
      "epoch": 6.378497430459254,
      "grad_norm": 0.11218845099210739,
      "learning_rate": 6.256884226943094e-08,
      "loss": 0.0329,
      "num_input_tokens_seen": 74088048,
      "step": 39100
    },
    {
      "epoch": 6.3793131576800715,
      "grad_norm": 0.05724416300654411,
      "learning_rate": 6.187661717174386e-08,
      "loss": 0.0357,
      "num_input_tokens_seen": 74098256,
      "step": 39105
    },
    {
      "epoch": 6.3801288849008895,
      "grad_norm": 0.3253237307071686,
      "learning_rate": 6.118823784607708e-08,
      "loss": 0.0358,
      "num_input_tokens_seen": 74106704,
      "step": 39110
    },
    {
      "epoch": 6.380944612121707,
      "grad_norm": 0.1655152291059494,
      "learning_rate": 6.050370439858178e-08,
      "loss": 0.0477,
      "num_input_tokens_seen": 74115712,
      "step": 39115
    },
    {
      "epoch": 6.381760339342524,
      "grad_norm": 0.7901779413223267,
      "learning_rate": 5.98230169348235e-08,
      "loss": 0.1085,
      "num_input_tokens_seen": 74125008,
      "step": 39120
    },
    {
      "epoch": 6.382576066563341,
      "grad_norm": 1.2860901355743408,
      "learning_rate": 5.914617555977664e-08,
      "loss": 0.2377,
      "num_input_tokens_seen": 74134672,
      "step": 39125
    },
    {
      "epoch": 6.383391793784159,
      "grad_norm": 0.410899817943573,
      "learning_rate": 5.8473180377816017e-08,
      "loss": 0.0662,
      "num_input_tokens_seen": 74144912,
      "step": 39130
    },
    {
      "epoch": 6.384207521004976,
      "grad_norm": 0.5425228476524353,
      "learning_rate": 5.780403149272251e-08,
      "loss": 0.1149,
      "num_input_tokens_seen": 74155456,
      "step": 39135
    },
    {
      "epoch": 6.385023248225793,
      "grad_norm": 0.40173497796058655,
      "learning_rate": 5.7138729007694126e-08,
      "loss": 0.0644,
      "num_input_tokens_seen": 74165424,
      "step": 39140
    },
    {
      "epoch": 6.38583897544661,
      "grad_norm": 0.03374416381120682,
      "learning_rate": 5.64772730253238e-08,
      "loss": 0.2714,
      "num_input_tokens_seen": 74175552,
      "step": 39145
    },
    {
      "epoch": 6.386654702667428,
      "grad_norm": 1.0236141681671143,
      "learning_rate": 5.5819663647618814e-08,
      "loss": 0.0393,
      "num_input_tokens_seen": 74184960,
      "step": 39150
    },
    {
      "epoch": 6.387470429888245,
      "grad_norm": 0.4989677369594574,
      "learning_rate": 5.5165900975989723e-08,
      "loss": 0.2311,
      "num_input_tokens_seen": 74194560,
      "step": 39155
    },
    {
      "epoch": 6.388286157109063,
      "grad_norm": 0.10159284621477127,
      "learning_rate": 5.451598511125311e-08,
      "loss": 0.1314,
      "num_input_tokens_seen": 74203568,
      "step": 39160
    },
    {
      "epoch": 6.38910188432988,
      "grad_norm": 0.4912877082824707,
      "learning_rate": 5.3869916153637124e-08,
      "loss": 0.0544,
      "num_input_tokens_seen": 74213680,
      "step": 39165
    },
    {
      "epoch": 6.389917611550698,
      "grad_norm": 0.8289280533790588,
      "learning_rate": 5.322769420277318e-08,
      "loss": 0.1332,
      "num_input_tokens_seen": 74223376,
      "step": 39170
    },
    {
      "epoch": 6.390733338771515,
      "grad_norm": 0.32157692313194275,
      "learning_rate": 5.258931935769873e-08,
      "loss": 0.1188,
      "num_input_tokens_seen": 74233056,
      "step": 39175
    },
    {
      "epoch": 6.391549065992332,
      "grad_norm": 0.728066086769104,
      "learning_rate": 5.19547917168628e-08,
      "loss": 0.0405,
      "num_input_tokens_seen": 74242352,
      "step": 39180
    },
    {
      "epoch": 6.392364793213149,
      "grad_norm": 1.4189047813415527,
      "learning_rate": 5.13241113781121e-08,
      "loss": 0.3479,
      "num_input_tokens_seen": 74252096,
      "step": 39185
    },
    {
      "epoch": 6.393180520433967,
      "grad_norm": 1.5357471704483032,
      "learning_rate": 5.0697278438707755e-08,
      "loss": 0.1141,
      "num_input_tokens_seen": 74260400,
      "step": 39190
    },
    {
      "epoch": 6.393996247654784,
      "grad_norm": 0.34252968430519104,
      "learning_rate": 5.0074292995316854e-08,
      "loss": 0.2141,
      "num_input_tokens_seen": 74269968,
      "step": 39195
    },
    {
      "epoch": 6.394811974875601,
      "grad_norm": 1.1632628440856934,
      "learning_rate": 4.945515514400978e-08,
      "loss": 0.1392,
      "num_input_tokens_seen": 74278960,
      "step": 39200
    },
    {
      "epoch": 6.394811974875601,
      "eval_loss": 0.14225821197032928,
      "eval_runtime": 133.239,
      "eval_samples_per_second": 20.452,
      "eval_steps_per_second": 10.23,
      "num_input_tokens_seen": 74278960,
      "step": 39200
    },
    {
      "epoch": 6.395627702096419,
      "grad_norm": 1.0909461975097656,
      "learning_rate": 4.883986498026571e-08,
      "loss": 0.0962,
      "num_input_tokens_seen": 74288016,
      "step": 39205
    },
    {
      "epoch": 6.3964434293172365,
      "grad_norm": 1.0713921785354614,
      "learning_rate": 4.822842259896987e-08,
      "loss": 0.2476,
      "num_input_tokens_seen": 74297856,
      "step": 39210
    },
    {
      "epoch": 6.397259156538054,
      "grad_norm": 0.11188636720180511,
      "learning_rate": 4.762082809441626e-08,
      "loss": 0.0727,
      "num_input_tokens_seen": 74307744,
      "step": 39215
    },
    {
      "epoch": 6.398074883758871,
      "grad_norm": 1.4354768991470337,
      "learning_rate": 4.7017081560302156e-08,
      "loss": 0.1302,
      "num_input_tokens_seen": 74317536,
      "step": 39220
    },
    {
      "epoch": 6.398890610979688,
      "grad_norm": 0.688845157623291,
      "learning_rate": 4.6417183089730866e-08,
      "loss": 0.15,
      "num_input_tokens_seen": 74325920,
      "step": 39225
    },
    {
      "epoch": 6.399706338200506,
      "grad_norm": 0.2135213166475296,
      "learning_rate": 4.5821132775217265e-08,
      "loss": 0.0692,
      "num_input_tokens_seen": 74334592,
      "step": 39230
    },
    {
      "epoch": 6.400522065421323,
      "grad_norm": 0.817013144493103,
      "learning_rate": 4.5228930708679504e-08,
      "loss": 0.1821,
      "num_input_tokens_seen": 74342704,
      "step": 39235
    },
    {
      "epoch": 6.40133779264214,
      "grad_norm": 0.16184262931346893,
      "learning_rate": 4.464057698144175e-08,
      "loss": 0.1252,
      "num_input_tokens_seen": 74352000,
      "step": 39240
    },
    {
      "epoch": 6.402153519862958,
      "grad_norm": 0.5434083938598633,
      "learning_rate": 4.4056071684236974e-08,
      "loss": 0.1995,
      "num_input_tokens_seen": 74361040,
      "step": 39245
    },
    {
      "epoch": 6.402969247083775,
      "grad_norm": 1.7704178094863892,
      "learning_rate": 4.347541490719864e-08,
      "loss": 0.2789,
      "num_input_tokens_seen": 74370528,
      "step": 39250
    },
    {
      "epoch": 6.4037849743045925,
      "grad_norm": 2.12982177734375,
      "learning_rate": 4.2898606739877336e-08,
      "loss": 0.1531,
      "num_input_tokens_seen": 74380544,
      "step": 39255
    },
    {
      "epoch": 6.40460070152541,
      "grad_norm": 0.5042626261711121,
      "learning_rate": 4.232564727122135e-08,
      "loss": 0.0348,
      "num_input_tokens_seen": 74389856,
      "step": 39260
    },
    {
      "epoch": 6.405416428746228,
      "grad_norm": 1.6335585117340088,
      "learning_rate": 4.1756536589585004e-08,
      "loss": 0.2341,
      "num_input_tokens_seen": 74398928,
      "step": 39265
    },
    {
      "epoch": 6.406232155967045,
      "grad_norm": 0.9876236915588379,
      "learning_rate": 4.119127478273976e-08,
      "loss": 0.1864,
      "num_input_tokens_seen": 74408512,
      "step": 39270
    },
    {
      "epoch": 6.407047883187862,
      "grad_norm": 0.38531798124313354,
      "learning_rate": 4.062986193784923e-08,
      "loss": 0.1493,
      "num_input_tokens_seen": 74416576,
      "step": 39275
    },
    {
      "epoch": 6.407863610408679,
      "grad_norm": 1.4174330234527588,
      "learning_rate": 4.007229814149416e-08,
      "loss": 0.2027,
      "num_input_tokens_seen": 74425952,
      "step": 39280
    },
    {
      "epoch": 6.408679337629497,
      "grad_norm": 0.3895745575428009,
      "learning_rate": 3.951858347965576e-08,
      "loss": 0.1809,
      "num_input_tokens_seen": 74433776,
      "step": 39285
    },
    {
      "epoch": 6.409495064850314,
      "grad_norm": 0.23986995220184326,
      "learning_rate": 3.896871803772684e-08,
      "loss": 0.0903,
      "num_input_tokens_seen": 74443552,
      "step": 39290
    },
    {
      "epoch": 6.410310792071131,
      "grad_norm": 0.12729506194591522,
      "learning_rate": 3.842270190050068e-08,
      "loss": 0.2182,
      "num_input_tokens_seen": 74452672,
      "step": 39295
    },
    {
      "epoch": 6.411126519291948,
      "grad_norm": 1.0751612186431885,
      "learning_rate": 3.7880535152179376e-08,
      "loss": 0.2655,
      "num_input_tokens_seen": 74461696,
      "step": 39300
    },
    {
      "epoch": 6.4119422465127665,
      "grad_norm": 1.4521493911743164,
      "learning_rate": 3.734221787637382e-08,
      "loss": 0.0846,
      "num_input_tokens_seen": 74470480,
      "step": 39305
    },
    {
      "epoch": 6.412757973733584,
      "grad_norm": 0.5466761589050293,
      "learning_rate": 3.680775015609817e-08,
      "loss": 0.2706,
      "num_input_tokens_seen": 74481376,
      "step": 39310
    },
    {
      "epoch": 6.413573700954401,
      "grad_norm": 0.3339051604270935,
      "learning_rate": 3.627713207377537e-08,
      "loss": 0.0685,
      "num_input_tokens_seen": 74490464,
      "step": 39315
    },
    {
      "epoch": 6.414389428175218,
      "grad_norm": 0.10046771168708801,
      "learning_rate": 3.575036371123164e-08,
      "loss": 0.0795,
      "num_input_tokens_seen": 74501184,
      "step": 39320
    },
    {
      "epoch": 6.415205155396036,
      "grad_norm": 0.9792365431785583,
      "learning_rate": 3.5227445149704776e-08,
      "loss": 0.192,
      "num_input_tokens_seen": 74510368,
      "step": 39325
    },
    {
      "epoch": 6.416020882616853,
      "grad_norm": 0.7388211488723755,
      "learning_rate": 3.470837646983027e-08,
      "loss": 0.2249,
      "num_input_tokens_seen": 74520336,
      "step": 39330
    },
    {
      "epoch": 6.41683660983767,
      "grad_norm": 1.4633920192718506,
      "learning_rate": 3.419315775165799e-08,
      "loss": 0.1559,
      "num_input_tokens_seen": 74529952,
      "step": 39335
    },
    {
      "epoch": 6.417652337058487,
      "grad_norm": 0.502022385597229,
      "learning_rate": 3.368178907464103e-08,
      "loss": 0.0506,
      "num_input_tokens_seen": 74540288,
      "step": 39340
    },
    {
      "epoch": 6.418468064279305,
      "grad_norm": 2.4117369651794434,
      "learning_rate": 3.317427051763855e-08,
      "loss": 0.1194,
      "num_input_tokens_seen": 74549424,
      "step": 39345
    },
    {
      "epoch": 6.419283791500122,
      "grad_norm": 0.07802785933017731,
      "learning_rate": 3.267060215891571e-08,
      "loss": 0.0759,
      "num_input_tokens_seen": 74559760,
      "step": 39350
    },
    {
      "epoch": 6.4200995187209395,
      "grad_norm": 0.563725471496582,
      "learning_rate": 3.217078407614649e-08,
      "loss": 0.2208,
      "num_input_tokens_seen": 74568208,
      "step": 39355
    },
    {
      "epoch": 6.420915245941757,
      "grad_norm": 0.09188076853752136,
      "learning_rate": 3.1674816346405345e-08,
      "loss": 0.0868,
      "num_input_tokens_seen": 74576976,
      "step": 39360
    },
    {
      "epoch": 6.421730973162575,
      "grad_norm": 0.1395782232284546,
      "learning_rate": 3.11826990461811e-08,
      "loss": 0.1267,
      "num_input_tokens_seen": 74587984,
      "step": 39365
    },
    {
      "epoch": 6.422546700383392,
      "grad_norm": 0.46644124388694763,
      "learning_rate": 3.069443225136304e-08,
      "loss": 0.1669,
      "num_input_tokens_seen": 74598064,
      "step": 39370
    },
    {
      "epoch": 6.423362427604209,
      "grad_norm": 0.08616316318511963,
      "learning_rate": 3.021001603724372e-08,
      "loss": 0.0531,
      "num_input_tokens_seen": 74608176,
      "step": 39375
    },
    {
      "epoch": 6.424178154825027,
      "grad_norm": 1.0720852613449097,
      "learning_rate": 2.9729450478532818e-08,
      "loss": 0.1136,
      "num_input_tokens_seen": 74617776,
      "step": 39380
    },
    {
      "epoch": 6.424993882045844,
      "grad_norm": 0.5030749440193176,
      "learning_rate": 2.9252735649337726e-08,
      "loss": 0.0562,
      "num_input_tokens_seen": 74627488,
      "step": 39385
    },
    {
      "epoch": 6.425809609266661,
      "grad_norm": 0.36694473028182983,
      "learning_rate": 2.8779871623171863e-08,
      "loss": 0.1063,
      "num_input_tokens_seen": 74636672,
      "step": 39390
    },
    {
      "epoch": 6.426625336487478,
      "grad_norm": 1.2978640794754028,
      "learning_rate": 2.8310858472957448e-08,
      "loss": 0.1784,
      "num_input_tokens_seen": 74645856,
      "step": 39395
    },
    {
      "epoch": 6.4274410637082955,
      "grad_norm": 0.6061780452728271,
      "learning_rate": 2.784569627101996e-08,
      "loss": 0.2032,
      "num_input_tokens_seen": 74655728,
      "step": 39400
    },
    {
      "epoch": 6.4274410637082955,
      "eval_loss": 0.14245256781578064,
      "eval_runtime": 133.0492,
      "eval_samples_per_second": 20.481,
      "eval_steps_per_second": 10.244,
      "num_input_tokens_seen": 74655728,
      "step": 39400
    },
    {
      "epoch": 6.4282567909291135,
      "grad_norm": 1.2031853199005127,
      "learning_rate": 2.738438508909924e-08,
      "loss": 0.1224,
      "num_input_tokens_seen": 74665184,
      "step": 39405
    },
    {
      "epoch": 6.429072518149931,
      "grad_norm": 0.1164856106042862,
      "learning_rate": 2.692692499833005e-08,
      "loss": 0.1099,
      "num_input_tokens_seen": 74675024,
      "step": 39410
    },
    {
      "epoch": 6.429888245370748,
      "grad_norm": 0.046646203845739365,
      "learning_rate": 2.647331606926151e-08,
      "loss": 0.1534,
      "num_input_tokens_seen": 74684736,
      "step": 39415
    },
    {
      "epoch": 6.430703972591566,
      "grad_norm": 0.7230230569839478,
      "learning_rate": 2.6023558371843225e-08,
      "loss": 0.1282,
      "num_input_tokens_seen": 74693536,
      "step": 39420
    },
    {
      "epoch": 6.431519699812383,
      "grad_norm": 0.14545784890651703,
      "learning_rate": 2.557765197543638e-08,
      "loss": 0.0214,
      "num_input_tokens_seen": 74703680,
      "step": 39425
    },
    {
      "epoch": 6.4323354270332,
      "grad_norm": 0.4616521894931793,
      "learning_rate": 2.513559694880263e-08,
      "loss": 0.1595,
      "num_input_tokens_seen": 74711664,
      "step": 39430
    },
    {
      "epoch": 6.433151154254017,
      "grad_norm": 0.8860753178596497,
      "learning_rate": 2.469739336011523e-08,
      "loss": 0.1029,
      "num_input_tokens_seen": 74719616,
      "step": 39435
    },
    {
      "epoch": 6.433966881474835,
      "grad_norm": 0.5471423864364624,
      "learning_rate": 2.4263041276947894e-08,
      "loss": 0.1055,
      "num_input_tokens_seen": 74727472,
      "step": 39440
    },
    {
      "epoch": 6.434782608695652,
      "grad_norm": 0.7691605091094971,
      "learning_rate": 2.3832540766283164e-08,
      "loss": 0.1748,
      "num_input_tokens_seen": 74736704,
      "step": 39445
    },
    {
      "epoch": 6.4355983359164695,
      "grad_norm": 1.3011064529418945,
      "learning_rate": 2.3405891894512366e-08,
      "loss": 0.0998,
      "num_input_tokens_seen": 74744784,
      "step": 39450
    },
    {
      "epoch": 6.436414063137287,
      "grad_norm": 0.5810723900794983,
      "learning_rate": 2.29830947274301e-08,
      "loss": 0.1207,
      "num_input_tokens_seen": 74755104,
      "step": 39455
    },
    {
      "epoch": 6.437229790358105,
      "grad_norm": 0.21556571125984192,
      "learning_rate": 2.2564149330231432e-08,
      "loss": 0.0599,
      "num_input_tokens_seen": 74766288,
      "step": 39460
    },
    {
      "epoch": 6.438045517578922,
      "grad_norm": 0.8068953156471252,
      "learning_rate": 2.2149055767528572e-08,
      "loss": 0.1065,
      "num_input_tokens_seen": 74775120,
      "step": 39465
    },
    {
      "epoch": 6.438861244799739,
      "grad_norm": 0.33618593215942383,
      "learning_rate": 2.1737814103334197e-08,
      "loss": 0.1351,
      "num_input_tokens_seen": 74783968,
      "step": 39470
    },
    {
      "epoch": 6.439676972020556,
      "grad_norm": 0.2648180425167084,
      "learning_rate": 2.1330424401064253e-08,
      "loss": 0.0443,
      "num_input_tokens_seen": 74792480,
      "step": 39475
    },
    {
      "epoch": 6.440492699241374,
      "grad_norm": 0.4347954988479614,
      "learning_rate": 2.092688672354348e-08,
      "loss": 0.1209,
      "num_input_tokens_seen": 74802016,
      "step": 39480
    },
    {
      "epoch": 6.441308426462191,
      "grad_norm": 0.3307605981826782,
      "learning_rate": 2.0527201133005435e-08,
      "loss": 0.0882,
      "num_input_tokens_seen": 74810080,
      "step": 39485
    },
    {
      "epoch": 6.442124153683008,
      "grad_norm": 0.13669589161872864,
      "learning_rate": 2.0131367691084148e-08,
      "loss": 0.1405,
      "num_input_tokens_seen": 74819040,
      "step": 39490
    },
    {
      "epoch": 6.442939880903825,
      "grad_norm": 1.3223861455917358,
      "learning_rate": 1.9739386458819675e-08,
      "loss": 0.1909,
      "num_input_tokens_seen": 74828128,
      "step": 39495
    },
    {
      "epoch": 6.443755608124643,
      "grad_norm": 0.5205034613609314,
      "learning_rate": 1.9351257496666442e-08,
      "loss": 0.0853,
      "num_input_tokens_seen": 74838368,
      "step": 39500
    },
    {
      "epoch": 6.444571335345461,
      "grad_norm": 0.2773425877094269,
      "learning_rate": 1.896698086447657e-08,
      "loss": 0.1231,
      "num_input_tokens_seen": 74849536,
      "step": 39505
    },
    {
      "epoch": 6.445387062566278,
      "grad_norm": 0.5228831171989441,
      "learning_rate": 1.8586556621505436e-08,
      "loss": 0.0694,
      "num_input_tokens_seen": 74857392,
      "step": 39510
    },
    {
      "epoch": 6.446202789787095,
      "grad_norm": 0.25231337547302246,
      "learning_rate": 1.820998482642833e-08,
      "loss": 0.302,
      "num_input_tokens_seen": 74867744,
      "step": 39515
    },
    {
      "epoch": 6.447018517007913,
      "grad_norm": 0.16965040564537048,
      "learning_rate": 1.7837265537309912e-08,
      "loss": 0.0826,
      "num_input_tokens_seen": 74876640,
      "step": 39520
    },
    {
      "epoch": 6.44783424422873,
      "grad_norm": 1.7230017185211182,
      "learning_rate": 1.7468398811629206e-08,
      "loss": 0.0704,
      "num_input_tokens_seen": 74886640,
      "step": 39525
    },
    {
      "epoch": 6.448649971449547,
      "grad_norm": 0.7992973327636719,
      "learning_rate": 1.710338470627404e-08,
      "loss": 0.1329,
      "num_input_tokens_seen": 74896256,
      "step": 39530
    },
    {
      "epoch": 6.449465698670364,
      "grad_norm": 0.4691260755062103,
      "learning_rate": 1.6742223277529945e-08,
      "loss": 0.0829,
      "num_input_tokens_seen": 74905696,
      "step": 39535
    },
    {
      "epoch": 6.450281425891182,
      "grad_norm": 0.3978077173233032,
      "learning_rate": 1.6384914581094036e-08,
      "loss": 0.0983,
      "num_input_tokens_seen": 74913088,
      "step": 39540
    },
    {
      "epoch": 6.451097153111999,
      "grad_norm": 1.1025954484939575,
      "learning_rate": 1.6031458672069455e-08,
      "loss": 0.1312,
      "num_input_tokens_seen": 74922432,
      "step": 39545
    },
    {
      "epoch": 6.4519128803328165,
      "grad_norm": 1.2665549516677856,
      "learning_rate": 1.5681855604962602e-08,
      "loss": 0.2371,
      "num_input_tokens_seen": 74931936,
      "step": 39550
    },
    {
      "epoch": 6.4527286075536345,
      "grad_norm": 1.4920340776443481,
      "learning_rate": 1.5336105433683135e-08,
      "loss": 0.1796,
      "num_input_tokens_seen": 74940048,
      "step": 39555
    },
    {
      "epoch": 6.453544334774452,
      "grad_norm": 2.383114814758301,
      "learning_rate": 1.499420821155506e-08,
      "loss": 0.1837,
      "num_input_tokens_seen": 74949104,
      "step": 39560
    },
    {
      "epoch": 6.454360061995269,
      "grad_norm": 0.5156840085983276,
      "learning_rate": 1.4656163991302874e-08,
      "loss": 0.1889,
      "num_input_tokens_seen": 74959056,
      "step": 39565
    },
    {
      "epoch": 6.455175789216086,
      "grad_norm": 0.2952776253223419,
      "learning_rate": 1.4321972825051544e-08,
      "loss": 0.1075,
      "num_input_tokens_seen": 74968816,
      "step": 39570
    },
    {
      "epoch": 6.455991516436903,
      "grad_norm": 1.0077159404754639,
      "learning_rate": 1.3991634764345951e-08,
      "loss": 0.11,
      "num_input_tokens_seen": 74978864,
      "step": 39575
    },
    {
      "epoch": 6.456807243657721,
      "grad_norm": 0.8759254217147827,
      "learning_rate": 1.3665149860120352e-08,
      "loss": 0.1096,
      "num_input_tokens_seen": 74988432,
      "step": 39580
    },
    {
      "epoch": 6.457622970878538,
      "grad_norm": 0.7527428269386292,
      "learning_rate": 1.3342518162728912e-08,
      "loss": 0.1253,
      "num_input_tokens_seen": 74997120,
      "step": 39585
    },
    {
      "epoch": 6.458438698099355,
      "grad_norm": 0.08259136974811554,
      "learning_rate": 1.30237397219235e-08,
      "loss": 0.0717,
      "num_input_tokens_seen": 75006896,
      "step": 39590
    },
    {
      "epoch": 6.459254425320173,
      "grad_norm": 0.4227370619773865,
      "learning_rate": 1.2708814586862016e-08,
      "loss": 0.2183,
      "num_input_tokens_seen": 75015984,
      "step": 39595
    },
    {
      "epoch": 6.4600701525409905,
      "grad_norm": 0.11849796026945114,
      "learning_rate": 1.2397742806111168e-08,
      "loss": 0.0652,
      "num_input_tokens_seen": 75025808,
      "step": 39600
    },
    {
      "epoch": 6.4600701525409905,
      "eval_loss": 0.14245256781578064,
      "eval_runtime": 133.2853,
      "eval_samples_per_second": 20.445,
      "eval_steps_per_second": 10.226,
      "num_input_tokens_seen": 75025808,
      "step": 39600
    },
    {
      "epoch": 6.460885879761808,
      "grad_norm": 0.09575550258159637,
      "learning_rate": 1.209052442764369e-08,
      "loss": 0.0592,
      "num_input_tokens_seen": 75036560,
      "step": 39605
    },
    {
      "epoch": 6.461701606982625,
      "grad_norm": 0.24186062812805176,
      "learning_rate": 1.17871594988328e-08,
      "loss": 0.1112,
      "num_input_tokens_seen": 75045664,
      "step": 39610
    },
    {
      "epoch": 6.462517334203443,
      "grad_norm": 3.0662481784820557,
      "learning_rate": 1.1487648066466072e-08,
      "loss": 0.2804,
      "num_input_tokens_seen": 75056032,
      "step": 39615
    },
    {
      "epoch": 6.46333306142426,
      "grad_norm": 0.9036136865615845,
      "learning_rate": 1.1191990176728784e-08,
      "loss": 0.2032,
      "num_input_tokens_seen": 75065760,
      "step": 39620
    },
    {
      "epoch": 6.464148788645077,
      "grad_norm": 1.019392967224121,
      "learning_rate": 1.0900185875215018e-08,
      "loss": 0.2032,
      "num_input_tokens_seen": 75076128,
      "step": 39625
    },
    {
      "epoch": 6.464964515865894,
      "grad_norm": 0.2994774878025055,
      "learning_rate": 1.0612235206924891e-08,
      "loss": 0.0529,
      "num_input_tokens_seen": 75084416,
      "step": 39630
    },
    {
      "epoch": 6.465780243086712,
      "grad_norm": 0.22003641724586487,
      "learning_rate": 1.0328138216264549e-08,
      "loss": 0.0879,
      "num_input_tokens_seen": 75093552,
      "step": 39635
    },
    {
      "epoch": 6.466595970307529,
      "grad_norm": 1.471720814704895,
      "learning_rate": 1.004789494704339e-08,
      "loss": 0.15,
      "num_input_tokens_seen": 75102032,
      "step": 39640
    },
    {
      "epoch": 6.467411697528346,
      "grad_norm": 0.37328433990478516,
      "learning_rate": 9.771505442482397e-09,
      "loss": 0.2492,
      "num_input_tokens_seen": 75111584,
      "step": 39645
    },
    {
      "epoch": 6.468227424749164,
      "grad_norm": 0.17636419832706451,
      "learning_rate": 9.498969745200259e-09,
      "loss": 0.1001,
      "num_input_tokens_seen": 75121104,
      "step": 39650
    },
    {
      "epoch": 6.469043151969982,
      "grad_norm": 0.04080367833375931,
      "learning_rate": 9.230287897230017e-09,
      "loss": 0.0493,
      "num_input_tokens_seen": 75130048,
      "step": 39655
    },
    {
      "epoch": 6.469858879190799,
      "grad_norm": 0.8869524002075195,
      "learning_rate": 8.965459940002419e-09,
      "loss": 0.1431,
      "num_input_tokens_seen": 75139312,
      "step": 39660
    },
    {
      "epoch": 6.470674606411616,
      "grad_norm": 2.249614953994751,
      "learning_rate": 8.704485914357019e-09,
      "loss": 0.0898,
      "num_input_tokens_seen": 75148080,
      "step": 39665
    },
    {
      "epoch": 6.471490333632433,
      "grad_norm": 0.13253869116306305,
      "learning_rate": 8.447365860539402e-09,
      "loss": 0.0754,
      "num_input_tokens_seen": 75156992,
      "step": 39670
    },
    {
      "epoch": 6.472306060853251,
      "grad_norm": 0.6750577688217163,
      "learning_rate": 8.194099818201184e-09,
      "loss": 0.0583,
      "num_input_tokens_seen": 75166768,
      "step": 39675
    },
    {
      "epoch": 6.473121788074068,
      "grad_norm": 0.13084876537322998,
      "learning_rate": 7.944687826400011e-09,
      "loss": 0.1457,
      "num_input_tokens_seen": 75176944,
      "step": 39680
    },
    {
      "epoch": 6.473937515294885,
      "grad_norm": 0.18848566710948944,
      "learning_rate": 7.699129923599557e-09,
      "loss": 0.0675,
      "num_input_tokens_seen": 75185520,
      "step": 39685
    },
    {
      "epoch": 6.474753242515702,
      "grad_norm": 1.853149652481079,
      "learning_rate": 7.457426147663982e-09,
      "loss": 0.101,
      "num_input_tokens_seen": 75195232,
      "step": 39690
    },
    {
      "epoch": 6.47556896973652,
      "grad_norm": 0.02968081459403038,
      "learning_rate": 7.219576535871797e-09,
      "loss": 0.0983,
      "num_input_tokens_seen": 75203904,
      "step": 39695
    },
    {
      "epoch": 6.4763846969573375,
      "grad_norm": 0.5586473941802979,
      "learning_rate": 6.985581124896445e-09,
      "loss": 0.0449,
      "num_input_tokens_seen": 75213472,
      "step": 39700
    },
    {
      "epoch": 6.477200424178155,
      "grad_norm": 1.5430290699005127,
      "learning_rate": 6.755439950828501e-09,
      "loss": 0.1406,
      "num_input_tokens_seen": 75222928,
      "step": 39705
    },
    {
      "epoch": 6.478016151398972,
      "grad_norm": 1.02631676197052,
      "learning_rate": 6.5291530491562444e-09,
      "loss": 0.1168,
      "num_input_tokens_seen": 75232512,
      "step": 39710
    },
    {
      "epoch": 6.47883187861979,
      "grad_norm": 0.5743056535720825,
      "learning_rate": 6.3067204547739845e-09,
      "loss": 0.2117,
      "num_input_tokens_seen": 75240800,
      "step": 39715
    },
    {
      "epoch": 6.479647605840607,
      "grad_norm": 0.8191107511520386,
      "learning_rate": 6.088142201987612e-09,
      "loss": 0.0443,
      "num_input_tokens_seen": 75251024,
      "step": 39720
    },
    {
      "epoch": 6.480463333061424,
      "grad_norm": 0.4279698431491852,
      "learning_rate": 5.873418324503499e-09,
      "loss": 0.0496,
      "num_input_tokens_seen": 75261456,
      "step": 39725
    },
    {
      "epoch": 6.481279060282241,
      "grad_norm": 0.42166557908058167,
      "learning_rate": 5.6625488554340465e-09,
      "loss": 0.078,
      "num_input_tokens_seen": 75271072,
      "step": 39730
    },
    {
      "epoch": 6.482094787503059,
      "grad_norm": 0.14464104175567627,
      "learning_rate": 5.455533827297688e-09,
      "loss": 0.0445,
      "num_input_tokens_seen": 75281104,
      "step": 39735
    },
    {
      "epoch": 6.482910514723876,
      "grad_norm": 1.135565161705017,
      "learning_rate": 5.252373272018885e-09,
      "loss": 0.2343,
      "num_input_tokens_seen": 75290608,
      "step": 39740
    },
    {
      "epoch": 6.4837262419446935,
      "grad_norm": 0.16016501188278198,
      "learning_rate": 5.053067220925356e-09,
      "loss": 0.1945,
      "num_input_tokens_seen": 75301520,
      "step": 39745
    },
    {
      "epoch": 6.484541969165511,
      "grad_norm": 0.08365603536367416,
      "learning_rate": 4.857615704759177e-09,
      "loss": 0.0392,
      "num_input_tokens_seen": 75309568,
      "step": 39750
    },
    {
      "epoch": 6.485357696386329,
      "grad_norm": 0.35971397161483765,
      "learning_rate": 4.666018753654577e-09,
      "loss": 0.0533,
      "num_input_tokens_seen": 75319136,
      "step": 39755
    },
    {
      "epoch": 6.486173423607146,
      "grad_norm": 0.15975484251976013,
      "learning_rate": 4.478276397162917e-09,
      "loss": 0.34,
      "num_input_tokens_seen": 75328000,
      "step": 39760
    },
    {
      "epoch": 6.486989150827963,
      "grad_norm": 0.09720250219106674,
      "learning_rate": 4.294388664233262e-09,
      "loss": 0.2616,
      "num_input_tokens_seen": 75336752,
      "step": 39765
    },
    {
      "epoch": 6.487804878048781,
      "grad_norm": 0.8578381538391113,
      "learning_rate": 4.114355583223484e-09,
      "loss": 0.158,
      "num_input_tokens_seen": 75345488,
      "step": 39770
    },
    {
      "epoch": 6.488620605269598,
      "grad_norm": 1.7517222166061401,
      "learning_rate": 3.9381771818974845e-09,
      "loss": 0.1599,
      "num_input_tokens_seen": 75355680,
      "step": 39775
    },
    {
      "epoch": 6.489436332490415,
      "grad_norm": 0.8019508123397827,
      "learning_rate": 3.765853487427973e-09,
      "loss": 0.0351,
      "num_input_tokens_seen": 75365712,
      "step": 39780
    },
    {
      "epoch": 6.490252059711232,
      "grad_norm": 0.42680326104164124,
      "learning_rate": 3.5973845263825857e-09,
      "loss": 0.1506,
      "num_input_tokens_seen": 75374320,
      "step": 39785
    },
    {
      "epoch": 6.49106778693205,
      "grad_norm": 0.07068247348070145,
      "learning_rate": 3.4327703247488684e-09,
      "loss": 0.0746,
      "num_input_tokens_seen": 75382832,
      "step": 39790
    },
    {
      "epoch": 6.4918835141528675,
      "grad_norm": 1.0653772354125977,
      "learning_rate": 3.2720109079037443e-09,
      "loss": 0.0999,
      "num_input_tokens_seen": 75392208,
      "step": 39795
    },
    {
      "epoch": 6.492699241373685,
      "grad_norm": 0.23158521950244904,
      "learning_rate": 3.1151063006468193e-09,
      "loss": 0.0902,
      "num_input_tokens_seen": 75402576,
      "step": 39800
    },
    {
      "epoch": 6.492699241373685,
      "eval_loss": 0.14245256781578064,
      "eval_runtime": 133.3312,
      "eval_samples_per_second": 20.438,
      "eval_steps_per_second": 10.223,
      "num_input_tokens_seen": 75402576,
      "step": 39800
    },
    {
      "epoch": 6.493514968594502,
      "grad_norm": 1.4412978887557983,
      "learning_rate": 2.962056527169854e-09,
      "loss": 0.2669,
      "num_input_tokens_seen": 75412160,
      "step": 39805
    },
    {
      "epoch": 6.49433069581532,
      "grad_norm": 0.4506610333919525,
      "learning_rate": 2.8128616110761898e-09,
      "loss": 0.0811,
      "num_input_tokens_seen": 75421648,
      "step": 39810
    },
    {
      "epoch": 6.495146423036137,
      "grad_norm": 0.398832231760025,
      "learning_rate": 2.6675215753724223e-09,
      "loss": 0.0605,
      "num_input_tokens_seen": 75432032,
      "step": 39815
    },
    {
      "epoch": 6.495962150256954,
      "grad_norm": 0.2763766944408417,
      "learning_rate": 2.5260364424739557e-09,
      "loss": 0.0521,
      "num_input_tokens_seen": 75441760,
      "step": 39820
    },
    {
      "epoch": 6.496777877477771,
      "grad_norm": 0.7520205974578857,
      "learning_rate": 2.3884062341994475e-09,
      "loss": 0.1185,
      "num_input_tokens_seen": 75451104,
      "step": 39825
    },
    {
      "epoch": 6.497593604698589,
      "grad_norm": 0.06056993454694748,
      "learning_rate": 2.25463097177081e-09,
      "loss": 0.2317,
      "num_input_tokens_seen": 75458992,
      "step": 39830
    },
    {
      "epoch": 6.498409331919406,
      "grad_norm": 0.48759526014328003,
      "learning_rate": 2.1247106758215397e-09,
      "loss": 0.1584,
      "num_input_tokens_seen": 75466768,
      "step": 39835
    },
    {
      "epoch": 6.499225059140223,
      "grad_norm": 0.2678789794445038,
      "learning_rate": 1.998645366382834e-09,
      "loss": 0.1486,
      "num_input_tokens_seen": 75477344,
      "step": 39840
    },
    {
      "epoch": 6.5000407863610405,
      "grad_norm": 0.29218265414237976,
      "learning_rate": 1.876435062897475e-09,
      "loss": 0.0896,
      "num_input_tokens_seen": 75488112,
      "step": 39845
    },
    {
      "epoch": 6.500856513581859,
      "grad_norm": 1.0181243419647217,
      "learning_rate": 1.758079784211497e-09,
      "loss": 0.2212,
      "num_input_tokens_seen": 75497344,
      "step": 39850
    },
    {
      "epoch": 6.501672240802676,
      "grad_norm": 0.5171177983283997,
      "learning_rate": 1.6435795485797434e-09,
      "loss": 0.0507,
      "num_input_tokens_seen": 75506928,
      "step": 39855
    },
    {
      "epoch": 6.502487968023493,
      "grad_norm": 0.8645734190940857,
      "learning_rate": 1.5329343736547596e-09,
      "loss": 0.1987,
      "num_input_tokens_seen": 75516624,
      "step": 39860
    },
    {
      "epoch": 6.50330369524431,
      "grad_norm": 0.8671128749847412,
      "learning_rate": 1.4261442765006739e-09,
      "loss": 0.0461,
      "num_input_tokens_seen": 75525952,
      "step": 39865
    },
    {
      "epoch": 6.504119422465128,
      "grad_norm": 0.07069169729948044,
      "learning_rate": 1.3232092735876445e-09,
      "loss": 0.1,
      "num_input_tokens_seen": 75535680,
      "step": 39870
    },
    {
      "epoch": 6.504935149685945,
      "grad_norm": 0.18107649683952332,
      "learning_rate": 1.2241293807918607e-09,
      "loss": 0.0572,
      "num_input_tokens_seen": 75544816,
      "step": 39875
    },
    {
      "epoch": 6.505750876906762,
      "grad_norm": 0.1588710993528366,
      "learning_rate": 1.128904613387216e-09,
      "loss": 0.1361,
      "num_input_tokens_seen": 75555120,
      "step": 39880
    },
    {
      "epoch": 6.506566604127579,
      "grad_norm": 0.050861768424510956,
      "learning_rate": 1.0375349860591853e-09,
      "loss": 0.1418,
      "num_input_tokens_seen": 75565568,
      "step": 39885
    },
    {
      "epoch": 6.507382331348397,
      "grad_norm": 0.7511735558509827,
      "learning_rate": 9.5002051290205e-10,
      "loss": 0.0828,
      "num_input_tokens_seen": 75573360,
      "step": 39890
    },
    {
      "epoch": 6.5081980585692145,
      "grad_norm": 0.02038341574370861,
      "learning_rate": 8.663612074077954e-10,
      "loss": 0.2526,
      "num_input_tokens_seen": 75582720,
      "step": 39895
    },
    {
      "epoch": 6.509013785790032,
      "grad_norm": 1.931482195854187,
      "learning_rate": 7.865570824799884e-10,
      "loss": 0.1144,
      "num_input_tokens_seen": 75592496,
      "step": 39900
    },
    {
      "epoch": 6.50982951301085,
      "grad_norm": 1.6246843338012695,
      "learning_rate": 7.106081504254514e-10,
      "loss": 0.2278,
      "num_input_tokens_seen": 75602064,
      "step": 39905
    },
    {
      "epoch": 6.510645240231667,
      "grad_norm": 0.2420821636915207,
      "learning_rate": 6.385144229570372e-10,
      "loss": 0.0666,
      "num_input_tokens_seen": 75611456,
      "step": 39910
    },
    {
      "epoch": 6.511460967452484,
      "grad_norm": 0.24482804536819458,
      "learning_rate": 5.70275911190854e-10,
      "loss": 0.069,
      "num_input_tokens_seen": 75619840,
      "step": 39915
    },
    {
      "epoch": 6.512276694673301,
      "grad_norm": 1.1324602365493774,
      "learning_rate": 5.058926256490403e-10,
      "loss": 0.1353,
      "num_input_tokens_seen": 75628864,
      "step": 39920
    },
    {
      "epoch": 6.513092421894118,
      "grad_norm": 0.20809394121170044,
      "learning_rate": 4.4536457626254134e-10,
      "loss": 0.0889,
      "num_input_tokens_seen": 75637728,
      "step": 39925
    },
    {
      "epoch": 6.513908149114936,
      "grad_norm": 0.7123565673828125,
      "learning_rate": 3.88691772365557e-10,
      "loss": 0.0948,
      "num_input_tokens_seen": 75647120,
      "step": 39930
    },
    {
      "epoch": 6.514723876335753,
      "grad_norm": 0.10274849086999893,
      "learning_rate": 3.358742226955425e-10,
      "loss": 0.0463,
      "num_input_tokens_seen": 75656224,
      "step": 39935
    },
    {
      "epoch": 6.5155396035565705,
      "grad_norm": 0.41245147585868835,
      "learning_rate": 2.8691193539875925e-10,
      "loss": 0.1995,
      "num_input_tokens_seen": 75665648,
      "step": 39940
    },
    {
      "epoch": 6.5163553307773885,
      "grad_norm": 0.8409091234207153,
      "learning_rate": 2.418049180274995e-10,
      "loss": 0.1197,
      "num_input_tokens_seen": 75675488,
      "step": 39945
    },
    {
      "epoch": 6.517171057998206,
      "grad_norm": 1.228999376296997,
      "learning_rate": 2.005531775373104e-10,
      "loss": 0.2009,
      "num_input_tokens_seen": 75684480,
      "step": 39950
    },
    {
      "epoch": 6.517986785219023,
      "grad_norm": 0.8481007218360901,
      "learning_rate": 1.6315672028699435e-10,
      "loss": 0.1635,
      "num_input_tokens_seen": 75693952,
      "step": 39955
    },
    {
      "epoch": 6.51880251243984,
      "grad_norm": 0.5860039591789246,
      "learning_rate": 1.2961555204693555e-10,
      "loss": 0.1509,
      "num_input_tokens_seen": 75703808,
      "step": 39960
    },
    {
      "epoch": 6.519618239660657,
      "grad_norm": 0.8073468208312988,
      "learning_rate": 9.992967798799768e-11,
      "loss": 0.2701,
      "num_input_tokens_seen": 75712960,
      "step": 39965
    },
    {
      "epoch": 6.520433966881475,
      "grad_norm": 0.6856575012207031,
      "learning_rate": 7.409910268707521e-11,
      "loss": 0.0591,
      "num_input_tokens_seen": 75722928,
      "step": 39970
    },
    {
      "epoch": 6.521249694102292,
      "grad_norm": 0.7067521810531616,
      "learning_rate": 5.212383012986877e-11,
      "loss": 0.0467,
      "num_input_tokens_seen": 75730560,
      "step": 39975
    },
    {
      "epoch": 6.522065421323109,
      "grad_norm": 1.072304368019104,
      "learning_rate": 3.400386370533415e-11,
      "loss": 0.0595,
      "num_input_tokens_seen": 75739568,
      "step": 39980
    },
    {
      "epoch": 6.522881148543927,
      "grad_norm": 0.11282755434513092,
      "learning_rate": 1.9739206205682258e-11,
      "loss": 0.1092,
      "num_input_tokens_seen": 75749696,
      "step": 39985
    },
    {
      "epoch": 6.523696875764744,
      "grad_norm": 1.2605178356170654,
      "learning_rate": 9.329859829154685e-12,
      "loss": 0.1889,
      "num_input_tokens_seen": 75759696,
      "step": 39990
    },
    {
      "epoch": 6.524512602985562,
      "grad_norm": 0.05451243370771408,
      "learning_rate": 2.7758261855748148e-12,
      "loss": 0.2616,
      "num_input_tokens_seen": 75769312,
      "step": 39995
    },
    {
      "epoch": 6.525328330206379,
      "grad_norm": 0.3582322299480438,
      "learning_rate": 7.710628524559838e-14,
      "loss": 0.0873,
      "num_input_tokens_seen": 75778784,
      "step": 40000
    },
    {
      "epoch": 6.525328330206379,
      "eval_loss": 0.14245256781578064,
      "eval_runtime": 133.2616,
      "eval_samples_per_second": 20.448,
      "eval_steps_per_second": 10.228,
      "num_input_tokens_seen": 75778784,
      "step": 40000
    },
    {
      "epoch": 6.525328330206379,
      "num_input_tokens_seen": 75778784,
      "step": 40000,
      "total_flos": 3.412373786225541e+18,
      "train_loss": 0.14572211567377671,
      "train_runtime": 51601.2388,
      "train_samples_per_second": 3.101,
      "train_steps_per_second": 0.775
    }
  ],
  "logging_steps": 5,
  "max_steps": 40000,
  "num_input_tokens_seen": 75778784,
  "num_train_epochs": 7,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.412373786225541e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}