fpadovani/swe-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
swe-latn-100mb-after-ppt-Dp…/checkpoint-2000/trainer_state.json
ModelHub XC 18dcce3b89 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/swe-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407
Source: Original Platform
2026-10-04 04:57:18 +08:00

4079 lines
110 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.6570008285004143,
"eval_steps": 500,
"global_step": 2000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 7.218191003799438,
"epoch": 0.004142502071251036,
"grad_norm": 1.265625,
"learning_rate": 2e-06,
"loss": 6.8898,
"mean_token_accuracy": 0.08805170804262161,
"num_tokens": 9149.0,
"step": 5
},
{
"entropy": 7.191673707962036,
"epoch": 0.008285004142502071,
"grad_norm": 1.5078125,
"learning_rate": 4.5e-06,
"loss": 6.7268,
"mean_token_accuracy": 0.09304728880524635,
"num_tokens": 16800.0,
"step": 10
},
{
"entropy": 7.134524202346801,
"epoch": 0.012427506213753107,
"grad_norm": 1.1640625,
"learning_rate": 7e-06,
"loss": 6.8905,
"mean_token_accuracy": 0.0921505056321621,
"num_tokens": 26617.0,
"step": 15
},
{
"entropy": 7.115550088882446,
"epoch": 0.016570008285004142,
"grad_norm": 1.25,
"learning_rate": 9.5e-06,
"loss": 6.8021,
"mean_token_accuracy": 0.09268744736909866,
"num_tokens": 35855.0,
"step": 20
},
{
"entropy": 7.12396879196167,
"epoch": 0.020712510356255178,
"grad_norm": 1.1328125,
"learning_rate": 1.2e-05,
"loss": 7.0094,
"mean_token_accuracy": 0.08752714470028877,
"num_tokens": 45644.0,
"step": 25
},
{
"entropy": 7.054446792602539,
"epoch": 0.024855012427506214,
"grad_norm": 1.1328125,
"learning_rate": 1.4500000000000002e-05,
"loss": 6.8922,
"mean_token_accuracy": 0.09398627206683159,
"num_tokens": 54641.0,
"step": 30
},
{
"entropy": 7.012255907058716,
"epoch": 0.02899751449875725,
"grad_norm": 1.0234375,
"learning_rate": 1.7000000000000003e-05,
"loss": 6.7519,
"mean_token_accuracy": 0.08958807587623596,
"num_tokens": 63924.0,
"step": 35
},
{
"entropy": 7.068031549453735,
"epoch": 0.033140016570008285,
"grad_norm": 1.1328125,
"learning_rate": 1.95e-05,
"loss": 6.89,
"mean_token_accuracy": 0.09437783882021904,
"num_tokens": 73452.0,
"step": 40
},
{
"entropy": 7.06072301864624,
"epoch": 0.037282518641259324,
"grad_norm": 1.0234375,
"learning_rate": 2.2e-05,
"loss": 6.9455,
"mean_token_accuracy": 0.0851291574537754,
"num_tokens": 82731.0,
"step": 45
},
{
"entropy": 7.090083169937134,
"epoch": 0.041425020712510356,
"grad_norm": 1.1015625,
"learning_rate": 2.4500000000000003e-05,
"loss": 6.7734,
"mean_token_accuracy": 0.09048294499516488,
"num_tokens": 92626.0,
"step": 50
},
{
"entropy": 7.149203157424926,
"epoch": 0.045567522783761395,
"grad_norm": 1.21875,
"learning_rate": 2.7e-05,
"loss": 6.9497,
"mean_token_accuracy": 0.08217904865741729,
"num_tokens": 101380.0,
"step": 55
},
{
"entropy": 7.119951963424683,
"epoch": 0.04971002485501243,
"grad_norm": 1.046875,
"learning_rate": 2.95e-05,
"loss": 6.8604,
"mean_token_accuracy": 0.09202293381094932,
"num_tokens": 109895.0,
"step": 60
},
{
"entropy": 7.2014998435974125,
"epoch": 0.053852526926263466,
"grad_norm": 1.1015625,
"learning_rate": 3.2e-05,
"loss": 6.8449,
"mean_token_accuracy": 0.09257669150829315,
"num_tokens": 119203.0,
"step": 65
},
{
"entropy": 7.1111664295196535,
"epoch": 0.0579950289975145,
"grad_norm": 1.1484375,
"learning_rate": 3.4500000000000005e-05,
"loss": 6.7903,
"mean_token_accuracy": 0.09657297283411026,
"num_tokens": 127572.0,
"step": 70
},
{
"entropy": 7.04930944442749,
"epoch": 0.06213753106876554,
"grad_norm": 1.03125,
"learning_rate": 3.7e-05,
"loss": 6.8304,
"mean_token_accuracy": 0.09063600301742554,
"num_tokens": 137161.0,
"step": 75
},
{
"entropy": 7.044711875915527,
"epoch": 0.06628003314001657,
"grad_norm": 1.1015625,
"learning_rate": 3.95e-05,
"loss": 6.9451,
"mean_token_accuracy": 0.08874713778495788,
"num_tokens": 145991.0,
"step": 80
},
{
"entropy": 7.130092525482178,
"epoch": 0.07042253521126761,
"grad_norm": 1.0546875,
"learning_rate": 4.2000000000000004e-05,
"loss": 6.8464,
"mean_token_accuracy": 0.08955071866512299,
"num_tokens": 154927.0,
"step": 85
},
{
"entropy": 7.090402746200562,
"epoch": 0.07456503728251865,
"grad_norm": 1.34375,
"learning_rate": 4.45e-05,
"loss": 6.7595,
"mean_token_accuracy": 0.09832768961787224,
"num_tokens": 163234.0,
"step": 90
},
{
"entropy": 7.08957576751709,
"epoch": 0.07870753935376967,
"grad_norm": 1.015625,
"learning_rate": 4.7000000000000004e-05,
"loss": 6.7163,
"mean_token_accuracy": 0.09379880875349045,
"num_tokens": 171554.0,
"step": 95
},
{
"entropy": 7.044517278671265,
"epoch": 0.08285004142502071,
"grad_norm": 1.1328125,
"learning_rate": 4.9500000000000004e-05,
"loss": 6.8173,
"mean_token_accuracy": 0.0955286517739296,
"num_tokens": 180171.0,
"step": 100
},
{
"entropy": 7.050950908660889,
"epoch": 0.08699254349627175,
"grad_norm": 1.0859375,
"learning_rate": 5.2e-05,
"loss": 6.761,
"mean_token_accuracy": 0.09741089940071106,
"num_tokens": 189216.0,
"step": 105
},
{
"entropy": 7.173562717437744,
"epoch": 0.09113504556752279,
"grad_norm": 1.015625,
"learning_rate": 5.45e-05,
"loss": 6.9519,
"mean_token_accuracy": 0.08675800338387489,
"num_tokens": 200022.0,
"step": 110
},
{
"entropy": 7.183744478225708,
"epoch": 0.09527754763877382,
"grad_norm": 1.2265625,
"learning_rate": 5.7e-05,
"loss": 6.8642,
"mean_token_accuracy": 0.09210581332445145,
"num_tokens": 208686.0,
"step": 115
},
{
"entropy": 7.121539115905762,
"epoch": 0.09942004971002485,
"grad_norm": 1.1875,
"learning_rate": 5.9499999999999996e-05,
"loss": 6.8829,
"mean_token_accuracy": 0.09462191313505172,
"num_tokens": 218266.0,
"step": 120
},
{
"entropy": 7.006736946105957,
"epoch": 0.1035625517812759,
"grad_norm": 1.046875,
"learning_rate": 6.2e-05,
"loss": 6.7523,
"mean_token_accuracy": 0.08771666437387467,
"num_tokens": 227658.0,
"step": 125
},
{
"entropy": 7.040027761459351,
"epoch": 0.10770505385252693,
"grad_norm": 1.1015625,
"learning_rate": 6.450000000000001e-05,
"loss": 6.7014,
"mean_token_accuracy": 0.1004502885043621,
"num_tokens": 236301.0,
"step": 130
},
{
"entropy": 7.106470441818237,
"epoch": 0.11184755592377796,
"grad_norm": 1.1171875,
"learning_rate": 6.7e-05,
"loss": 6.8424,
"mean_token_accuracy": 0.08812138065695763,
"num_tokens": 246355.0,
"step": 135
},
{
"entropy": 7.174679756164551,
"epoch": 0.115990057995029,
"grad_norm": 1.15625,
"learning_rate": 6.950000000000001e-05,
"loss": 6.8174,
"mean_token_accuracy": 0.0924198254942894,
"num_tokens": 256023.0,
"step": 140
},
{
"entropy": 7.10174069404602,
"epoch": 0.12013256006628004,
"grad_norm": 0.9921875,
"learning_rate": 7.2e-05,
"loss": 6.9036,
"mean_token_accuracy": 0.08638993799686431,
"num_tokens": 266893.0,
"step": 145
},
{
"entropy": 7.019921684265137,
"epoch": 0.12427506213753108,
"grad_norm": 1.1640625,
"learning_rate": 7.45e-05,
"loss": 6.6606,
"mean_token_accuracy": 0.08872481286525727,
"num_tokens": 276075.0,
"step": 150
},
{
"entropy": 7.042754888534546,
"epoch": 0.12841756420878211,
"grad_norm": 1.25,
"learning_rate": 7.7e-05,
"loss": 6.7532,
"mean_token_accuracy": 0.09296932220458984,
"num_tokens": 285529.0,
"step": 155
},
{
"entropy": 7.064399909973145,
"epoch": 0.13256006628003314,
"grad_norm": 1.15625,
"learning_rate": 7.950000000000001e-05,
"loss": 6.762,
"mean_token_accuracy": 0.09406789913773536,
"num_tokens": 294403.0,
"step": 160
},
{
"entropy": 7.152362823486328,
"epoch": 0.13670256835128416,
"grad_norm": 1.28125,
"learning_rate": 8.2e-05,
"loss": 6.8553,
"mean_token_accuracy": 0.09358441680669785,
"num_tokens": 303507.0,
"step": 165
},
{
"entropy": 7.071487474441528,
"epoch": 0.14084507042253522,
"grad_norm": 1.21875,
"learning_rate": 8.450000000000001e-05,
"loss": 6.7534,
"mean_token_accuracy": 0.09645206108689308,
"num_tokens": 312393.0,
"step": 170
},
{
"entropy": 6.956210470199585,
"epoch": 0.14498757249378624,
"grad_norm": 1.0859375,
"learning_rate": 8.7e-05,
"loss": 6.7226,
"mean_token_accuracy": 0.09953253641724587,
"num_tokens": 321309.0,
"step": 175
},
{
"entropy": 7.2200812816619875,
"epoch": 0.1491300745650373,
"grad_norm": 1.1328125,
"learning_rate": 8.95e-05,
"loss": 6.7879,
"mean_token_accuracy": 0.09582221284508705,
"num_tokens": 330217.0,
"step": 180
},
{
"entropy": 6.964824962615967,
"epoch": 0.15327257663628832,
"grad_norm": 1.2109375,
"learning_rate": 9.2e-05,
"loss": 6.8296,
"mean_token_accuracy": 0.09584516286849976,
"num_tokens": 339587.0,
"step": 185
},
{
"entropy": 7.104565525054932,
"epoch": 0.15741507870753935,
"grad_norm": 1.28125,
"learning_rate": 9.45e-05,
"loss": 6.7603,
"mean_token_accuracy": 0.09563353508710862,
"num_tokens": 349367.0,
"step": 190
},
{
"entropy": 6.90956711769104,
"epoch": 0.1615575807787904,
"grad_norm": 1.1953125,
"learning_rate": 9.7e-05,
"loss": 6.6535,
"mean_token_accuracy": 0.09533214569091797,
"num_tokens": 358272.0,
"step": 195
},
{
"entropy": 7.184430170059204,
"epoch": 0.16570008285004142,
"grad_norm": 1.1875,
"learning_rate": 9.95e-05,
"loss": 6.8847,
"mean_token_accuracy": 0.09342379868030548,
"num_tokens": 367709.0,
"step": 200
},
{
"entropy": 7.108258152008057,
"epoch": 0.16984258492129245,
"grad_norm": 1.171875,
"learning_rate": 0.000102,
"loss": 6.7363,
"mean_token_accuracy": 0.09321728497743606,
"num_tokens": 376180.0,
"step": 205
},
{
"entropy": 7.143853282928466,
"epoch": 0.1739850869925435,
"grad_norm": 1.1328125,
"learning_rate": 0.00010449999999999999,
"loss": 6.8927,
"mean_token_accuracy": 0.08639989122748375,
"num_tokens": 385636.0,
"step": 210
},
{
"entropy": 7.005528688430786,
"epoch": 0.17812758906379453,
"grad_norm": 1.2109375,
"learning_rate": 0.000107,
"loss": 6.7603,
"mean_token_accuracy": 0.10387283116579056,
"num_tokens": 394486.0,
"step": 215
},
{
"entropy": 7.066776275634766,
"epoch": 0.18227009113504558,
"grad_norm": 1.2890625,
"learning_rate": 0.0001095,
"loss": 6.7286,
"mean_token_accuracy": 0.09739693626761436,
"num_tokens": 403765.0,
"step": 220
},
{
"entropy": 6.9391695022583,
"epoch": 0.1864125932062966,
"grad_norm": 1.3359375,
"learning_rate": 0.000112,
"loss": 6.7198,
"mean_token_accuracy": 0.09883175939321517,
"num_tokens": 412988.0,
"step": 225
},
{
"entropy": 7.029045248031617,
"epoch": 0.19055509527754763,
"grad_norm": 1.1015625,
"learning_rate": 0.0001145,
"loss": 6.6219,
"mean_token_accuracy": 0.09850638806819915,
"num_tokens": 422096.0,
"step": 230
},
{
"entropy": 7.055552911758423,
"epoch": 0.19469759734879868,
"grad_norm": 1.3359375,
"learning_rate": 0.00011700000000000001,
"loss": 6.7907,
"mean_token_accuracy": 0.08912368193268776,
"num_tokens": 431333.0,
"step": 235
},
{
"entropy": 7.076433277130127,
"epoch": 0.1988400994200497,
"grad_norm": 1.2109375,
"learning_rate": 0.00011949999999999999,
"loss": 6.825,
"mean_token_accuracy": 0.0998022735118866,
"num_tokens": 440382.0,
"step": 240
},
{
"entropy": 7.098494958877564,
"epoch": 0.20298260149130073,
"grad_norm": 1.2421875,
"learning_rate": 0.000122,
"loss": 6.7184,
"mean_token_accuracy": 0.1035430908203125,
"num_tokens": 448469.0,
"step": 245
},
{
"entropy": 7.065437746047974,
"epoch": 0.2071251035625518,
"grad_norm": 1.4375,
"learning_rate": 0.0001245,
"loss": 6.7562,
"mean_token_accuracy": 0.0973759837448597,
"num_tokens": 457356.0,
"step": 250
},
{
"entropy": 7.103154182434082,
"epoch": 0.2112676056338028,
"grad_norm": 1.296875,
"learning_rate": 0.000127,
"loss": 6.7821,
"mean_token_accuracy": 0.09721289053559304,
"num_tokens": 465928.0,
"step": 255
},
{
"entropy": 6.99681601524353,
"epoch": 0.21541010770505387,
"grad_norm": 1.140625,
"learning_rate": 0.0001295,
"loss": 6.6931,
"mean_token_accuracy": 0.09931288808584213,
"num_tokens": 475269.0,
"step": 260
},
{
"entropy": 6.960800266265869,
"epoch": 0.2195526097763049,
"grad_norm": 1.0703125,
"learning_rate": 0.000132,
"loss": 6.679,
"mean_token_accuracy": 0.09512786641716957,
"num_tokens": 485450.0,
"step": 265
},
{
"entropy": 7.0514568328857425,
"epoch": 0.22369511184755592,
"grad_norm": 1.3125,
"learning_rate": 0.00013450000000000002,
"loss": 6.7347,
"mean_token_accuracy": 0.10228226557374001,
"num_tokens": 494029.0,
"step": 270
},
{
"entropy": 6.9516912460327145,
"epoch": 0.22783761391880697,
"grad_norm": 1.3203125,
"learning_rate": 0.00013700000000000002,
"loss": 6.676,
"mean_token_accuracy": 0.10202871784567832,
"num_tokens": 503253.0,
"step": 275
},
{
"entropy": 7.018160057067871,
"epoch": 0.231980115990058,
"grad_norm": 1.1484375,
"learning_rate": 0.0001395,
"loss": 6.6685,
"mean_token_accuracy": 0.09529435187578202,
"num_tokens": 512857.0,
"step": 280
},
{
"entropy": 6.9714985370635985,
"epoch": 0.23612261806130902,
"grad_norm": 1.1640625,
"learning_rate": 0.00014199999999999998,
"loss": 6.7719,
"mean_token_accuracy": 0.09288227260112762,
"num_tokens": 522319.0,
"step": 285
},
{
"entropy": 7.113205528259277,
"epoch": 0.24026512013256007,
"grad_norm": 1.203125,
"learning_rate": 0.0001445,
"loss": 6.7224,
"mean_token_accuracy": 0.09471780881285667,
"num_tokens": 531469.0,
"step": 290
},
{
"entropy": 6.976934480667114,
"epoch": 0.2444076222038111,
"grad_norm": 1.34375,
"learning_rate": 0.000147,
"loss": 6.756,
"mean_token_accuracy": 0.09480062946677208,
"num_tokens": 540470.0,
"step": 295
},
{
"entropy": 7.011937475204467,
"epoch": 0.24855012427506215,
"grad_norm": 1.171875,
"learning_rate": 0.0001495,
"loss": 6.7584,
"mean_token_accuracy": 0.09700122326612473,
"num_tokens": 550317.0,
"step": 300
},
{
"entropy": 7.0365160465240475,
"epoch": 0.2526926263463132,
"grad_norm": 1.2265625,
"learning_rate": 0.000152,
"loss": 6.7067,
"mean_token_accuracy": 0.09434132501482964,
"num_tokens": 559409.0,
"step": 305
},
{
"entropy": 6.928857326507568,
"epoch": 0.25683512841756423,
"grad_norm": 1.171875,
"learning_rate": 0.00015450000000000001,
"loss": 6.768,
"mean_token_accuracy": 0.10764710158109665,
"num_tokens": 568726.0,
"step": 310
},
{
"entropy": 7.103991413116455,
"epoch": 0.2609776304888152,
"grad_norm": 1.21875,
"learning_rate": 0.000157,
"loss": 6.7179,
"mean_token_accuracy": 0.0976541943848133,
"num_tokens": 577926.0,
"step": 315
},
{
"entropy": 7.047203779220581,
"epoch": 0.2651201325600663,
"grad_norm": 1.265625,
"learning_rate": 0.0001595,
"loss": 6.7448,
"mean_token_accuracy": 0.09809661954641342,
"num_tokens": 586695.0,
"step": 320
},
{
"entropy": 7.039609956741333,
"epoch": 0.26926263463131733,
"grad_norm": 1.171875,
"learning_rate": 0.000162,
"loss": 6.771,
"mean_token_accuracy": 0.0949723981320858,
"num_tokens": 596250.0,
"step": 325
},
{
"entropy": 7.063632202148438,
"epoch": 0.27340513670256833,
"grad_norm": 1.1875,
"learning_rate": 0.00016450000000000001,
"loss": 6.8221,
"mean_token_accuracy": 0.09051206335425377,
"num_tokens": 605761.0,
"step": 330
},
{
"entropy": 7.019326877593994,
"epoch": 0.2775476387738194,
"grad_norm": 1.2109375,
"learning_rate": 0.00016700000000000002,
"loss": 6.7365,
"mean_token_accuracy": 0.09872144013643265,
"num_tokens": 614313.0,
"step": 335
},
{
"entropy": 7.037422132492066,
"epoch": 0.28169014084507044,
"grad_norm": 1.15625,
"learning_rate": 0.00016950000000000003,
"loss": 6.7692,
"mean_token_accuracy": 0.09455474764108658,
"num_tokens": 624470.0,
"step": 340
},
{
"entropy": 7.039082098007202,
"epoch": 0.28583264291632143,
"grad_norm": 1.2109375,
"learning_rate": 0.00017199999999999998,
"loss": 6.6843,
"mean_token_accuracy": 0.09269843101501465,
"num_tokens": 633458.0,
"step": 345
},
{
"entropy": 6.942268037796021,
"epoch": 0.2899751449875725,
"grad_norm": 1.1953125,
"learning_rate": 0.00017449999999999999,
"loss": 6.6939,
"mean_token_accuracy": 0.09877684414386749,
"num_tokens": 643251.0,
"step": 350
},
{
"entropy": 6.99584732055664,
"epoch": 0.29411764705882354,
"grad_norm": 1.3359375,
"learning_rate": 0.000177,
"loss": 6.7635,
"mean_token_accuracy": 0.09713540598750114,
"num_tokens": 652005.0,
"step": 355
},
{
"entropy": 7.006586170196533,
"epoch": 0.2982601491300746,
"grad_norm": 1.3125,
"learning_rate": 0.0001795,
"loss": 6.6575,
"mean_token_accuracy": 0.10120807588100433,
"num_tokens": 661003.0,
"step": 360
},
{
"entropy": 6.874996137619019,
"epoch": 0.3024026512013256,
"grad_norm": 1.109375,
"learning_rate": 0.000182,
"loss": 6.6294,
"mean_token_accuracy": 0.09844254851341247,
"num_tokens": 670221.0,
"step": 365
},
{
"entropy": 7.155635738372803,
"epoch": 0.30654515327257664,
"grad_norm": 1.296875,
"learning_rate": 0.0001845,
"loss": 6.9236,
"mean_token_accuracy": 0.0882349319756031,
"num_tokens": 680756.0,
"step": 370
},
{
"entropy": 6.94540753364563,
"epoch": 0.3106876553438277,
"grad_norm": 1.4140625,
"learning_rate": 0.000187,
"loss": 6.6157,
"mean_token_accuracy": 0.10289054960012436,
"num_tokens": 689641.0,
"step": 375
},
{
"entropy": 6.87992377281189,
"epoch": 0.3148301574150787,
"grad_norm": 1.3203125,
"learning_rate": 0.0001895,
"loss": 6.5696,
"mean_token_accuracy": 0.09839320331811904,
"num_tokens": 698209.0,
"step": 380
},
{
"entropy": 6.823869276046753,
"epoch": 0.31897265948632975,
"grad_norm": 1.21875,
"learning_rate": 0.000192,
"loss": 6.5736,
"mean_token_accuracy": 0.10355777889490128,
"num_tokens": 706712.0,
"step": 385
},
{
"entropy": 7.037634468078613,
"epoch": 0.3231151615575808,
"grad_norm": 1.15625,
"learning_rate": 0.0001945,
"loss": 6.8017,
"mean_token_accuracy": 0.09728012606501579,
"num_tokens": 716087.0,
"step": 390
},
{
"entropy": 6.911249494552612,
"epoch": 0.3272576636288318,
"grad_norm": 1.125,
"learning_rate": 0.00019700000000000002,
"loss": 6.6465,
"mean_token_accuracy": 0.09347675815224647,
"num_tokens": 725913.0,
"step": 395
},
{
"entropy": 6.944440317153931,
"epoch": 0.33140016570008285,
"grad_norm": 1.3515625,
"learning_rate": 0.00019950000000000002,
"loss": 6.7201,
"mean_token_accuracy": 0.10362306237220764,
"num_tokens": 735020.0,
"step": 400
},
{
"entropy": 6.985636854171753,
"epoch": 0.3355426677713339,
"grad_norm": 1.171875,
"learning_rate": 0.000202,
"loss": 6.6216,
"mean_token_accuracy": 0.09469136893749237,
"num_tokens": 744299.0,
"step": 405
},
{
"entropy": 6.883630275726318,
"epoch": 0.3396851698425849,
"grad_norm": 1.25,
"learning_rate": 0.00020449999999999998,
"loss": 6.6099,
"mean_token_accuracy": 0.09694516137242318,
"num_tokens": 753231.0,
"step": 410
},
{
"entropy": 6.825357389450073,
"epoch": 0.34382767191383595,
"grad_norm": 1.34375,
"learning_rate": 0.000207,
"loss": 6.6717,
"mean_token_accuracy": 0.10230854973196983,
"num_tokens": 763043.0,
"step": 415
},
{
"entropy": 7.016851997375488,
"epoch": 0.347970173985087,
"grad_norm": 1.4921875,
"learning_rate": 0.0002095,
"loss": 6.6905,
"mean_token_accuracy": 0.09913048222661018,
"num_tokens": 771977.0,
"step": 420
},
{
"entropy": 6.820244407653808,
"epoch": 0.352112676056338,
"grad_norm": 1.2734375,
"learning_rate": 0.000212,
"loss": 6.6243,
"mean_token_accuracy": 0.1011092983186245,
"num_tokens": 780285.0,
"step": 425
},
{
"entropy": 6.956075525283813,
"epoch": 0.35625517812758906,
"grad_norm": 1.3046875,
"learning_rate": 0.0002145,
"loss": 6.6593,
"mean_token_accuracy": 0.09649045094847679,
"num_tokens": 789980.0,
"step": 430
},
{
"entropy": 6.984846019744873,
"epoch": 0.3603976801988401,
"grad_norm": 1.4765625,
"learning_rate": 0.00021700000000000002,
"loss": 6.6275,
"mean_token_accuracy": 0.10474459528923034,
"num_tokens": 798700.0,
"step": 435
},
{
"entropy": 6.767291498184204,
"epoch": 0.36454018227009116,
"grad_norm": 1.2734375,
"learning_rate": 0.0002195,
"loss": 6.7208,
"mean_token_accuracy": 0.09600067660212516,
"num_tokens": 808758.0,
"step": 440
},
{
"entropy": 7.046369791030884,
"epoch": 0.36868268434134216,
"grad_norm": 1.4453125,
"learning_rate": 0.000222,
"loss": 6.6666,
"mean_token_accuracy": 0.10458802133798599,
"num_tokens": 817692.0,
"step": 445
},
{
"entropy": 6.95425066947937,
"epoch": 0.3728251864125932,
"grad_norm": 1.203125,
"learning_rate": 0.0002245,
"loss": 6.7146,
"mean_token_accuracy": 0.10474570766091347,
"num_tokens": 827052.0,
"step": 450
},
{
"entropy": 6.895345544815063,
"epoch": 0.37696768848384427,
"grad_norm": 1.265625,
"learning_rate": 0.00022700000000000002,
"loss": 6.7139,
"mean_token_accuracy": 0.1091882586479187,
"num_tokens": 836442.0,
"step": 455
},
{
"entropy": 6.922871923446655,
"epoch": 0.38111019055509526,
"grad_norm": 1.1875,
"learning_rate": 0.00022950000000000002,
"loss": 6.5775,
"mean_token_accuracy": 0.10118989571928978,
"num_tokens": 846441.0,
"step": 460
},
{
"entropy": 6.8815388679504395,
"epoch": 0.3852526926263463,
"grad_norm": 1.5390625,
"learning_rate": 0.00023200000000000003,
"loss": 6.6215,
"mean_token_accuracy": 0.10475475937128068,
"num_tokens": 856108.0,
"step": 465
},
{
"entropy": 6.848587274551392,
"epoch": 0.38939519469759737,
"grad_norm": 1.21875,
"learning_rate": 0.00023449999999999998,
"loss": 6.5833,
"mean_token_accuracy": 0.10137399956583977,
"num_tokens": 865788.0,
"step": 470
},
{
"entropy": 6.886804008483887,
"epoch": 0.39353769676884837,
"grad_norm": 1.2109375,
"learning_rate": 0.000237,
"loss": 6.6302,
"mean_token_accuracy": 0.09514767676591873,
"num_tokens": 874833.0,
"step": 475
},
{
"entropy": 6.751422643661499,
"epoch": 0.3976801988400994,
"grad_norm": 1.28125,
"learning_rate": 0.0002395,
"loss": 6.5997,
"mean_token_accuracy": 0.10276172533631325,
"num_tokens": 883685.0,
"step": 480
},
{
"entropy": 6.881311464309692,
"epoch": 0.40182270091135047,
"grad_norm": 1.2890625,
"learning_rate": 0.000242,
"loss": 6.6001,
"mean_token_accuracy": 0.09796027019619942,
"num_tokens": 893908.0,
"step": 485
},
{
"entropy": 6.934821891784668,
"epoch": 0.40596520298260147,
"grad_norm": 1.4140625,
"learning_rate": 0.0002445,
"loss": 6.7142,
"mean_token_accuracy": 0.09827135354280472,
"num_tokens": 903081.0,
"step": 490
},
{
"entropy": 6.7868632793426515,
"epoch": 0.4101077050538525,
"grad_norm": 1.1953125,
"learning_rate": 0.000247,
"loss": 6.5522,
"mean_token_accuracy": 0.10336750745773315,
"num_tokens": 912515.0,
"step": 495
},
{
"entropy": 6.886235475540161,
"epoch": 0.4142502071251036,
"grad_norm": 1.375,
"learning_rate": 0.0002495,
"loss": 6.5797,
"mean_token_accuracy": 0.1070313923060894,
"num_tokens": 921056.0,
"step": 500
},
{
"epoch": 0.4142502071251036,
"eval_entropy": 6.836038604728194,
"eval_loss": 6.684774875640869,
"eval_mean_token_accuracy": 0.10272799217021034,
"eval_num_tokens": 921056.0,
"eval_runtime": 2.7891,
"eval_samples_per_second": 1343.421,
"eval_steps_per_second": 168.152,
"step": 500
},
{
"entropy": 6.7657403469085695,
"epoch": 0.4183927091963546,
"grad_norm": 1.2421875,
"learning_rate": 0.000252,
"loss": 6.5317,
"mean_token_accuracy": 0.10417452603578567,
"num_tokens": 931112.0,
"step": 505
},
{
"entropy": 6.821895503997803,
"epoch": 0.4225352112676056,
"grad_norm": 1.1640625,
"learning_rate": 0.0002545,
"loss": 6.6605,
"mean_token_accuracy": 0.0978618748486042,
"num_tokens": 941120.0,
"step": 510
},
{
"entropy": 6.886211729049682,
"epoch": 0.4266777133388567,
"grad_norm": 1.3046875,
"learning_rate": 0.000257,
"loss": 6.6044,
"mean_token_accuracy": 0.09737976416945457,
"num_tokens": 950041.0,
"step": 515
},
{
"entropy": 6.879851579666138,
"epoch": 0.43082021541010773,
"grad_norm": 1.296875,
"learning_rate": 0.0002595,
"loss": 6.5907,
"mean_token_accuracy": 0.10606569275259972,
"num_tokens": 958608.0,
"step": 520
},
{
"entropy": 6.886484861373901,
"epoch": 0.43496271748135873,
"grad_norm": 1.40625,
"learning_rate": 0.000262,
"loss": 6.7158,
"mean_token_accuracy": 0.09990910813212395,
"num_tokens": 966901.0,
"step": 525
},
{
"entropy": 6.767313003540039,
"epoch": 0.4391052195526098,
"grad_norm": 1.234375,
"learning_rate": 0.00026450000000000003,
"loss": 6.5504,
"mean_token_accuracy": 0.10768762230873108,
"num_tokens": 976198.0,
"step": 530
},
{
"entropy": 6.892781782150268,
"epoch": 0.44324772162386084,
"grad_norm": 1.3671875,
"learning_rate": 0.00026700000000000004,
"loss": 6.624,
"mean_token_accuracy": 0.09861580505967141,
"num_tokens": 985248.0,
"step": 535
},
{
"entropy": 6.772344255447388,
"epoch": 0.44739022369511183,
"grad_norm": 1.2109375,
"learning_rate": 0.00026950000000000005,
"loss": 6.5014,
"mean_token_accuracy": 0.1071978710591793,
"num_tokens": 993289.0,
"step": 540
},
{
"entropy": 6.820696449279785,
"epoch": 0.4515327257663629,
"grad_norm": 1.3046875,
"learning_rate": 0.00027200000000000005,
"loss": 6.5909,
"mean_token_accuracy": 0.10533165782690049,
"num_tokens": 1001784.0,
"step": 545
},
{
"entropy": 6.935409688949585,
"epoch": 0.45567522783761394,
"grad_norm": 1.3203125,
"learning_rate": 0.0002745,
"loss": 6.5627,
"mean_token_accuracy": 0.10804210156202317,
"num_tokens": 1010445.0,
"step": 550
},
{
"entropy": 6.745790100097656,
"epoch": 0.45981772990886494,
"grad_norm": 1.4453125,
"learning_rate": 0.000277,
"loss": 6.6184,
"mean_token_accuracy": 0.10709234476089477,
"num_tokens": 1019084.0,
"step": 555
},
{
"entropy": 6.7490733623504635,
"epoch": 0.463960231980116,
"grad_norm": 1.1953125,
"learning_rate": 0.0002795,
"loss": 6.5249,
"mean_token_accuracy": 0.10461713373661041,
"num_tokens": 1027263.0,
"step": 560
},
{
"entropy": 6.876884937286377,
"epoch": 0.46810273405136704,
"grad_norm": 1.2109375,
"learning_rate": 0.00028199999999999997,
"loss": 6.6233,
"mean_token_accuracy": 0.09752989709377288,
"num_tokens": 1036275.0,
"step": 565
},
{
"entropy": 6.765736055374146,
"epoch": 0.47224523612261804,
"grad_norm": 1.375,
"learning_rate": 0.0002845,
"loss": 6.7015,
"mean_token_accuracy": 0.09858408868312836,
"num_tokens": 1045208.0,
"step": 570
},
{
"entropy": 6.9024957656860355,
"epoch": 0.4763877381938691,
"grad_norm": 1.34375,
"learning_rate": 0.000287,
"loss": 6.6287,
"mean_token_accuracy": 0.1026283286511898,
"num_tokens": 1054703.0,
"step": 575
},
{
"entropy": 6.647345924377442,
"epoch": 0.48053024026512015,
"grad_norm": 1.5,
"learning_rate": 0.0002895,
"loss": 6.4841,
"mean_token_accuracy": 0.10212503597140313,
"num_tokens": 1063652.0,
"step": 580
},
{
"entropy": 6.717900657653809,
"epoch": 0.48467274233637114,
"grad_norm": 1.203125,
"learning_rate": 0.000292,
"loss": 6.5873,
"mean_token_accuracy": 0.10145951434969902,
"num_tokens": 1072070.0,
"step": 585
},
{
"entropy": 6.997959756851197,
"epoch": 0.4888152444076222,
"grad_norm": 1.25,
"learning_rate": 0.0002945,
"loss": 6.6143,
"mean_token_accuracy": 0.10064447894692422,
"num_tokens": 1079992.0,
"step": 590
},
{
"entropy": 6.657160854339599,
"epoch": 0.49295774647887325,
"grad_norm": 1.2890625,
"learning_rate": 0.000297,
"loss": 6.628,
"mean_token_accuracy": 0.10435748845338821,
"num_tokens": 1088862.0,
"step": 595
},
{
"entropy": 6.773271942138672,
"epoch": 0.4971002485501243,
"grad_norm": 1.2890625,
"learning_rate": 0.0002995,
"loss": 6.4317,
"mean_token_accuracy": 0.10795770138502121,
"num_tokens": 1097614.0,
"step": 600
},
{
"entropy": 6.7102379322052,
"epoch": 0.5012427506213754,
"grad_norm": 1.1796875,
"learning_rate": 0.000302,
"loss": 6.6288,
"mean_token_accuracy": 0.10706980749964715,
"num_tokens": 1107347.0,
"step": 605
},
{
"entropy": 6.8672473430633545,
"epoch": 0.5053852526926264,
"grad_norm": 1.25,
"learning_rate": 0.0003045,
"loss": 6.6445,
"mean_token_accuracy": 0.10073774978518486,
"num_tokens": 1116707.0,
"step": 610
},
{
"entropy": 6.764383840560913,
"epoch": 0.5095277547638773,
"grad_norm": 1.296875,
"learning_rate": 0.000307,
"loss": 6.6549,
"mean_token_accuracy": 0.10234303548932075,
"num_tokens": 1126356.0,
"step": 615
},
{
"entropy": 6.787193298339844,
"epoch": 0.5136702568351285,
"grad_norm": 1.28125,
"learning_rate": 0.0003095,
"loss": 6.5847,
"mean_token_accuracy": 0.10079671740531922,
"num_tokens": 1136281.0,
"step": 620
},
{
"entropy": 6.713733196258545,
"epoch": 0.5178127589063795,
"grad_norm": 1.0859375,
"learning_rate": 0.000312,
"loss": 6.5355,
"mean_token_accuracy": 0.10429632142186165,
"num_tokens": 1145466.0,
"step": 625
},
{
"entropy": 6.7503923892974855,
"epoch": 0.5219552609776305,
"grad_norm": 1.2890625,
"learning_rate": 0.0003145,
"loss": 6.5409,
"mean_token_accuracy": 0.10918259546160698,
"num_tokens": 1154225.0,
"step": 630
},
{
"entropy": 6.761834907531738,
"epoch": 0.5260977630488816,
"grad_norm": 1.203125,
"learning_rate": 0.000317,
"loss": 6.5885,
"mean_token_accuracy": 0.09673135951161385,
"num_tokens": 1163872.0,
"step": 635
},
{
"entropy": 6.69431324005127,
"epoch": 0.5302402651201326,
"grad_norm": 1.1953125,
"learning_rate": 0.0003195,
"loss": 6.5244,
"mean_token_accuracy": 0.10705070868134499,
"num_tokens": 1172695.0,
"step": 640
},
{
"entropy": 6.651590061187744,
"epoch": 0.5343827671913836,
"grad_norm": 1.2109375,
"learning_rate": 0.000322,
"loss": 6.3962,
"mean_token_accuracy": 0.11006554961204529,
"num_tokens": 1181291.0,
"step": 645
},
{
"entropy": 6.782713270187378,
"epoch": 0.5385252692626347,
"grad_norm": 1.3515625,
"learning_rate": 0.00032450000000000003,
"loss": 6.5715,
"mean_token_accuracy": 0.10394523739814758,
"num_tokens": 1189927.0,
"step": 650
},
{
"entropy": 6.705166435241699,
"epoch": 0.5426677713338857,
"grad_norm": 1.2734375,
"learning_rate": 0.00032700000000000003,
"loss": 6.5043,
"mean_token_accuracy": 0.10940467119216919,
"num_tokens": 1197913.0,
"step": 655
},
{
"entropy": 6.670699262619019,
"epoch": 0.5468102734051367,
"grad_norm": 1.296875,
"learning_rate": 0.00032950000000000004,
"loss": 6.4309,
"mean_token_accuracy": 0.11147961542010307,
"num_tokens": 1207032.0,
"step": 660
},
{
"entropy": 6.640204334259034,
"epoch": 0.5509527754763878,
"grad_norm": 1.2109375,
"learning_rate": 0.00033200000000000005,
"loss": 6.5397,
"mean_token_accuracy": 0.10571447387337685,
"num_tokens": 1215710.0,
"step": 665
},
{
"entropy": 6.694819068908691,
"epoch": 0.5550952775476388,
"grad_norm": 1.21875,
"learning_rate": 0.00033450000000000005,
"loss": 6.52,
"mean_token_accuracy": 0.10049431845545768,
"num_tokens": 1225619.0,
"step": 670
},
{
"entropy": 6.649594163894653,
"epoch": 0.5592377796188898,
"grad_norm": 1.2421875,
"learning_rate": 0.000337,
"loss": 6.4407,
"mean_token_accuracy": 0.11157044768333435,
"num_tokens": 1235000.0,
"step": 675
},
{
"entropy": 6.612611961364746,
"epoch": 0.5633802816901409,
"grad_norm": 1.3359375,
"learning_rate": 0.0003395,
"loss": 6.4462,
"mean_token_accuracy": 0.108580531924963,
"num_tokens": 1243594.0,
"step": 680
},
{
"entropy": 6.843221950531006,
"epoch": 0.5675227837613919,
"grad_norm": 1.2578125,
"learning_rate": 0.000342,
"loss": 6.6445,
"mean_token_accuracy": 0.10460855960845947,
"num_tokens": 1252279.0,
"step": 685
},
{
"entropy": 6.760235929489136,
"epoch": 0.5716652858326429,
"grad_norm": 1.3359375,
"learning_rate": 0.00034449999999999997,
"loss": 6.559,
"mean_token_accuracy": 0.10665302500128745,
"num_tokens": 1261314.0,
"step": 690
},
{
"entropy": 6.55326566696167,
"epoch": 0.575807787903894,
"grad_norm": 1.1015625,
"learning_rate": 0.000347,
"loss": 6.4337,
"mean_token_accuracy": 0.11188110187649727,
"num_tokens": 1270854.0,
"step": 695
},
{
"entropy": 6.618375587463379,
"epoch": 0.579950289975145,
"grad_norm": 1.2734375,
"learning_rate": 0.0003495,
"loss": 6.3929,
"mean_token_accuracy": 0.11392810270190239,
"num_tokens": 1279183.0,
"step": 700
},
{
"entropy": 6.626208257675171,
"epoch": 0.584092792046396,
"grad_norm": 1.1640625,
"learning_rate": 0.000352,
"loss": 6.5015,
"mean_token_accuracy": 0.10761247500777245,
"num_tokens": 1288347.0,
"step": 705
},
{
"entropy": 6.6855669021606445,
"epoch": 0.5882352941176471,
"grad_norm": 1.25,
"learning_rate": 0.0003545,
"loss": 6.5036,
"mean_token_accuracy": 0.10823310390114785,
"num_tokens": 1297449.0,
"step": 710
},
{
"entropy": 6.710147094726563,
"epoch": 0.5923777961888981,
"grad_norm": 1.2578125,
"learning_rate": 0.000357,
"loss": 6.4688,
"mean_token_accuracy": 0.10904998034238815,
"num_tokens": 1306232.0,
"step": 715
},
{
"entropy": 6.6553794860839846,
"epoch": 0.5965202982601492,
"grad_norm": 1.2890625,
"learning_rate": 0.0003595,
"loss": 6.4857,
"mean_token_accuracy": 0.11109372302889824,
"num_tokens": 1315073.0,
"step": 720
},
{
"entropy": 6.674065732955933,
"epoch": 0.6006628003314002,
"grad_norm": 1.140625,
"learning_rate": 0.000362,
"loss": 6.5301,
"mean_token_accuracy": 0.10648825243115426,
"num_tokens": 1324629.0,
"step": 725
},
{
"entropy": 6.700368881225586,
"epoch": 0.6048053024026512,
"grad_norm": 1.1640625,
"learning_rate": 0.0003645,
"loss": 6.6444,
"mean_token_accuracy": 0.10443340390920638,
"num_tokens": 1334152.0,
"step": 730
},
{
"entropy": 6.779498624801636,
"epoch": 0.6089478044739023,
"grad_norm": 1.1484375,
"learning_rate": 0.000367,
"loss": 6.5946,
"mean_token_accuracy": 0.10130336955189705,
"num_tokens": 1343063.0,
"step": 735
},
{
"entropy": 6.684298419952393,
"epoch": 0.6130903065451533,
"grad_norm": 1.09375,
"learning_rate": 0.0003695,
"loss": 6.5241,
"mean_token_accuracy": 0.10516620799899101,
"num_tokens": 1352632.0,
"step": 740
},
{
"entropy": 6.582866573333741,
"epoch": 0.6172328086164043,
"grad_norm": 1.2578125,
"learning_rate": 0.000372,
"loss": 6.4712,
"mean_token_accuracy": 0.10765027329325676,
"num_tokens": 1361103.0,
"step": 745
},
{
"entropy": 6.643687963485718,
"epoch": 0.6213753106876554,
"grad_norm": 1.3359375,
"learning_rate": 0.0003745,
"loss": 6.4642,
"mean_token_accuracy": 0.10832677409052849,
"num_tokens": 1370658.0,
"step": 750
},
{
"entropy": 6.64506516456604,
"epoch": 0.6255178127589064,
"grad_norm": 1.15625,
"learning_rate": 0.000377,
"loss": 6.4878,
"mean_token_accuracy": 0.10808288529515267,
"num_tokens": 1379817.0,
"step": 755
},
{
"entropy": 6.640752458572388,
"epoch": 0.6296603148301574,
"grad_norm": 1.375,
"learning_rate": 0.0003795,
"loss": 6.4418,
"mean_token_accuracy": 0.10874675288796425,
"num_tokens": 1389329.0,
"step": 760
},
{
"entropy": 6.4627049446105955,
"epoch": 0.6338028169014085,
"grad_norm": 1.1953125,
"learning_rate": 0.000382,
"loss": 6.4902,
"mean_token_accuracy": 0.10606684386730195,
"num_tokens": 1399279.0,
"step": 765
},
{
"entropy": 6.621781063079834,
"epoch": 0.6379453189726595,
"grad_norm": 1.1484375,
"learning_rate": 0.0003845,
"loss": 6.4146,
"mean_token_accuracy": 0.11267002299427986,
"num_tokens": 1408765.0,
"step": 770
},
{
"entropy": 6.736465692520142,
"epoch": 0.6420878210439105,
"grad_norm": 1.2890625,
"learning_rate": 0.00038700000000000003,
"loss": 6.5632,
"mean_token_accuracy": 0.10607504546642303,
"num_tokens": 1418615.0,
"step": 775
},
{
"entropy": 6.616378879547119,
"epoch": 0.6462303231151616,
"grad_norm": 1.2890625,
"learning_rate": 0.00038950000000000003,
"loss": 6.4664,
"mean_token_accuracy": 0.10207068845629692,
"num_tokens": 1427359.0,
"step": 780
},
{
"entropy": 6.70065336227417,
"epoch": 0.6503728251864126,
"grad_norm": 1.203125,
"learning_rate": 0.00039200000000000004,
"loss": 6.622,
"mean_token_accuracy": 0.1116428442299366,
"num_tokens": 1435516.0,
"step": 785
},
{
"entropy": 6.709064102172851,
"epoch": 0.6545153272576636,
"grad_norm": 1.2265625,
"learning_rate": 0.00039450000000000005,
"loss": 6.5704,
"mean_token_accuracy": 0.10583743155002594,
"num_tokens": 1444438.0,
"step": 790
},
{
"entropy": 6.671351766586303,
"epoch": 0.6586578293289147,
"grad_norm": 1.171875,
"learning_rate": 0.00039700000000000005,
"loss": 6.4996,
"mean_token_accuracy": 0.11326526403427124,
"num_tokens": 1453740.0,
"step": 795
},
{
"entropy": 6.614573907852173,
"epoch": 0.6628003314001657,
"grad_norm": 1.2890625,
"learning_rate": 0.0003995,
"loss": 6.4874,
"mean_token_accuracy": 0.11111863479018211,
"num_tokens": 1462083.0,
"step": 800
},
{
"entropy": 6.484545516967773,
"epoch": 0.6669428334714167,
"grad_norm": 1.109375,
"learning_rate": 0.000402,
"loss": 6.4127,
"mean_token_accuracy": 0.1100592203438282,
"num_tokens": 1471459.0,
"step": 805
},
{
"entropy": 6.621589279174804,
"epoch": 0.6710853355426678,
"grad_norm": 1.1640625,
"learning_rate": 0.0004045,
"loss": 6.3972,
"mean_token_accuracy": 0.11681128293275833,
"num_tokens": 1480516.0,
"step": 810
},
{
"entropy": 6.421737241744995,
"epoch": 0.6752278376139188,
"grad_norm": 1.1796875,
"learning_rate": 0.00040699999999999997,
"loss": 6.3,
"mean_token_accuracy": 0.11318568140268326,
"num_tokens": 1489268.0,
"step": 815
},
{
"entropy": 6.700246906280517,
"epoch": 0.6793703396851698,
"grad_norm": 1.203125,
"learning_rate": 0.0004095,
"loss": 6.5174,
"mean_token_accuracy": 0.11110908538103104,
"num_tokens": 1498819.0,
"step": 820
},
{
"entropy": 6.622490644454956,
"epoch": 0.6835128417564209,
"grad_norm": 1.15625,
"learning_rate": 0.000412,
"loss": 6.501,
"mean_token_accuracy": 0.11033953949809075,
"num_tokens": 1507852.0,
"step": 825
},
{
"entropy": 6.59730544090271,
"epoch": 0.6876553438276719,
"grad_norm": 1.171875,
"learning_rate": 0.0004145,
"loss": 6.5302,
"mean_token_accuracy": 0.11122860461473465,
"num_tokens": 1517538.0,
"step": 830
},
{
"entropy": 6.458763217926025,
"epoch": 0.6917978458989229,
"grad_norm": 1.3046875,
"learning_rate": 0.000417,
"loss": 6.4061,
"mean_token_accuracy": 0.11324542313814163,
"num_tokens": 1525389.0,
"step": 835
},
{
"entropy": 6.57511887550354,
"epoch": 0.695940347970174,
"grad_norm": 1.2734375,
"learning_rate": 0.0004195,
"loss": 6.3741,
"mean_token_accuracy": 0.11116622686386109,
"num_tokens": 1534117.0,
"step": 840
},
{
"entropy": 6.522304344177246,
"epoch": 0.700082850041425,
"grad_norm": 1.2265625,
"learning_rate": 0.000422,
"loss": 6.4029,
"mean_token_accuracy": 0.11315450444817543,
"num_tokens": 1542610.0,
"step": 845
},
{
"entropy": 6.5175107479095455,
"epoch": 0.704225352112676,
"grad_norm": 1.3046875,
"learning_rate": 0.0004245,
"loss": 6.4206,
"mean_token_accuracy": 0.1143906556069851,
"num_tokens": 1551771.0,
"step": 850
},
{
"entropy": 6.620167636871338,
"epoch": 0.7083678541839271,
"grad_norm": 1.3984375,
"learning_rate": 0.000427,
"loss": 6.5779,
"mean_token_accuracy": 0.10512799099087715,
"num_tokens": 1560231.0,
"step": 855
},
{
"entropy": 6.589151382446289,
"epoch": 0.7125103562551781,
"grad_norm": 1.0859375,
"learning_rate": 0.0004295,
"loss": 6.5346,
"mean_token_accuracy": 0.10532297939062119,
"num_tokens": 1569578.0,
"step": 860
},
{
"entropy": 6.563750839233398,
"epoch": 0.7166528583264291,
"grad_norm": 1.171875,
"learning_rate": 0.000432,
"loss": 6.4187,
"mean_token_accuracy": 0.11381370946764946,
"num_tokens": 1579308.0,
"step": 865
},
{
"entropy": 6.597287034988403,
"epoch": 0.7207953603976802,
"grad_norm": 1.140625,
"learning_rate": 0.0004345,
"loss": 6.6081,
"mean_token_accuracy": 0.10267190784215927,
"num_tokens": 1589506.0,
"step": 870
},
{
"entropy": 6.482245397567749,
"epoch": 0.7249378624689312,
"grad_norm": 1.2890625,
"learning_rate": 0.000437,
"loss": 6.3671,
"mean_token_accuracy": 0.11368186473846435,
"num_tokens": 1598127.0,
"step": 875
},
{
"entropy": 6.548954200744629,
"epoch": 0.7290803645401823,
"grad_norm": 1.2109375,
"learning_rate": 0.0004395,
"loss": 6.4125,
"mean_token_accuracy": 0.11406067088246345,
"num_tokens": 1608073.0,
"step": 880
},
{
"entropy": 6.551317119598389,
"epoch": 0.7332228666114333,
"grad_norm": 1.1640625,
"learning_rate": 0.000442,
"loss": 6.4304,
"mean_token_accuracy": 0.11178040653467178,
"num_tokens": 1616272.0,
"step": 885
},
{
"entropy": 6.496135663986206,
"epoch": 0.7373653686826843,
"grad_norm": 1.125,
"learning_rate": 0.0004445,
"loss": 6.447,
"mean_token_accuracy": 0.10845504105091094,
"num_tokens": 1625577.0,
"step": 890
},
{
"entropy": 6.685812950134277,
"epoch": 0.7415078707539354,
"grad_norm": 1.234375,
"learning_rate": 0.000447,
"loss": 6.533,
"mean_token_accuracy": 0.10158233270049095,
"num_tokens": 1634902.0,
"step": 895
},
{
"entropy": 6.520436191558838,
"epoch": 0.7456503728251864,
"grad_norm": 1.1953125,
"learning_rate": 0.00044950000000000003,
"loss": 6.4674,
"mean_token_accuracy": 0.10865833014249801,
"num_tokens": 1644131.0,
"step": 900
},
{
"entropy": 6.461379718780518,
"epoch": 0.7497928748964374,
"grad_norm": 1.3671875,
"learning_rate": 0.00045200000000000004,
"loss": 6.3694,
"mean_token_accuracy": 0.12158194407820702,
"num_tokens": 1652272.0,
"step": 905
},
{
"entropy": 6.533070278167725,
"epoch": 0.7539353769676885,
"grad_norm": 1.2421875,
"learning_rate": 0.00045450000000000004,
"loss": 6.4632,
"mean_token_accuracy": 0.11447574123740197,
"num_tokens": 1660962.0,
"step": 910
},
{
"entropy": 6.475569725036621,
"epoch": 0.7580778790389395,
"grad_norm": 1.203125,
"learning_rate": 0.00045700000000000005,
"loss": 6.4006,
"mean_token_accuracy": 0.11205699220299721,
"num_tokens": 1670570.0,
"step": 915
},
{
"entropy": 6.526782608032226,
"epoch": 0.7622203811101905,
"grad_norm": 1.0859375,
"learning_rate": 0.00045950000000000006,
"loss": 6.4497,
"mean_token_accuracy": 0.1096197433769703,
"num_tokens": 1679674.0,
"step": 920
},
{
"entropy": 6.530770015716553,
"epoch": 0.7663628831814416,
"grad_norm": 1.21875,
"learning_rate": 0.000462,
"loss": 6.4324,
"mean_token_accuracy": 0.11466658413410187,
"num_tokens": 1688849.0,
"step": 925
},
{
"entropy": 6.4163538932800295,
"epoch": 0.7705053852526926,
"grad_norm": 1.25,
"learning_rate": 0.0004645,
"loss": 6.3128,
"mean_token_accuracy": 0.1098814107477665,
"num_tokens": 1698565.0,
"step": 930
},
{
"entropy": 6.5036052703857425,
"epoch": 0.7746478873239436,
"grad_norm": 1.2265625,
"learning_rate": 0.000467,
"loss": 6.3986,
"mean_token_accuracy": 0.11683695837855339,
"num_tokens": 1708009.0,
"step": 935
},
{
"entropy": 6.518198871612549,
"epoch": 0.7787903893951947,
"grad_norm": 1.2734375,
"learning_rate": 0.0004695,
"loss": 6.3878,
"mean_token_accuracy": 0.1126109592616558,
"num_tokens": 1716497.0,
"step": 940
},
{
"entropy": 6.415347719192505,
"epoch": 0.7829328914664457,
"grad_norm": 1.1640625,
"learning_rate": 0.000472,
"loss": 6.4082,
"mean_token_accuracy": 0.12070046439766884,
"num_tokens": 1725609.0,
"step": 945
},
{
"entropy": 6.477306175231933,
"epoch": 0.7870753935376967,
"grad_norm": 1.2734375,
"learning_rate": 0.0004745,
"loss": 6.3456,
"mean_token_accuracy": 0.11454906165599824,
"num_tokens": 1734165.0,
"step": 950
},
{
"entropy": 6.454520893096924,
"epoch": 0.7912178956089478,
"grad_norm": 1.109375,
"learning_rate": 0.000477,
"loss": 6.4546,
"mean_token_accuracy": 0.10918105840682983,
"num_tokens": 1744856.0,
"step": 955
},
{
"entropy": 6.4711943626403805,
"epoch": 0.7953603976801988,
"grad_norm": 1.1953125,
"learning_rate": 0.0004795,
"loss": 6.4509,
"mean_token_accuracy": 0.10578904524445534,
"num_tokens": 1753386.0,
"step": 960
},
{
"entropy": 6.4536961078643795,
"epoch": 0.7995028997514498,
"grad_norm": 1.1875,
"learning_rate": 0.000482,
"loss": 6.3053,
"mean_token_accuracy": 0.11872726455330848,
"num_tokens": 1762521.0,
"step": 965
},
{
"entropy": 6.34936490058899,
"epoch": 0.8036454018227009,
"grad_norm": 1.1640625,
"learning_rate": 0.0004845,
"loss": 6.3765,
"mean_token_accuracy": 0.11444728896021843,
"num_tokens": 1772403.0,
"step": 970
},
{
"entropy": 6.375087881088257,
"epoch": 0.8077879038939519,
"grad_norm": 1.2421875,
"learning_rate": 0.000487,
"loss": 6.2884,
"mean_token_accuracy": 0.11179185584187508,
"num_tokens": 1780924.0,
"step": 975
},
{
"entropy": 6.547906732559204,
"epoch": 0.8119304059652029,
"grad_norm": 1.203125,
"learning_rate": 0.0004895,
"loss": 6.4367,
"mean_token_accuracy": 0.11225800514221192,
"num_tokens": 1789992.0,
"step": 980
},
{
"entropy": 6.421040678024292,
"epoch": 0.816072908036454,
"grad_norm": 1.3046875,
"learning_rate": 0.000492,
"loss": 6.3919,
"mean_token_accuracy": 0.11761699616909027,
"num_tokens": 1799513.0,
"step": 985
},
{
"entropy": 6.332476711273193,
"epoch": 0.820215410107705,
"grad_norm": 1.375,
"learning_rate": 0.0004945,
"loss": 6.2287,
"mean_token_accuracy": 0.11279567033052444,
"num_tokens": 1807393.0,
"step": 990
},
{
"entropy": 6.438618326187134,
"epoch": 0.824357912178956,
"grad_norm": 1.1875,
"learning_rate": 0.000497,
"loss": 6.5079,
"mean_token_accuracy": 0.10579989776015282,
"num_tokens": 1817284.0,
"step": 995
},
{
"entropy": 6.449818515777588,
"epoch": 0.8285004142502072,
"grad_norm": 1.328125,
"learning_rate": 0.0004995,
"loss": 6.2996,
"mean_token_accuracy": 0.11799877136945724,
"num_tokens": 1825591.0,
"step": 1000
},
{
"epoch": 0.8285004142502072,
"eval_entropy": 6.222835092402216,
"eval_loss": 6.475013256072998,
"eval_mean_token_accuracy": 0.11269166863866961,
"eval_num_tokens": 1825591.0,
"eval_runtime": 2.7611,
"eval_samples_per_second": 1357.068,
"eval_steps_per_second": 169.86,
"step": 1000
},
{
"entropy": 6.457831478118896,
"epoch": 0.8326429163214581,
"grad_norm": 1.359375,
"learning_rate": 0.0004999998547681232,
"loss": 6.3959,
"mean_token_accuracy": 0.11798396334052086,
"num_tokens": 1834655.0,
"step": 1005
},
{
"entropy": 6.26599349975586,
"epoch": 0.8367854183927091,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999992647639447,
"loss": 6.2578,
"mean_token_accuracy": 0.11858200430870056,
"num_tokens": 1844187.0,
"step": 1010
},
{
"entropy": 6.415271520614624,
"epoch": 0.8409279204639603,
"grad_norm": 1.15625,
"learning_rate": 0.0004999982209116618,
"loss": 6.3345,
"mean_token_accuracy": 0.11406200677156449,
"num_tokens": 1852768.0,
"step": 1015
},
{
"entropy": 6.355557584762574,
"epoch": 0.8450704225352113,
"grad_norm": 1.2421875,
"learning_rate": 0.0004999967232133796,
"loss": 6.3913,
"mean_token_accuracy": 0.12348653748631477,
"num_tokens": 1862095.0,
"step": 1020
},
{
"entropy": 6.427043008804321,
"epoch": 0.8492129246064622,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999947716721196,
"loss": 6.4206,
"mean_token_accuracy": 0.10961534604430198,
"num_tokens": 1871239.0,
"step": 1025
},
{
"entropy": 6.509608554840088,
"epoch": 0.8533554266777134,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999923662918179,
"loss": 6.4323,
"mean_token_accuracy": 0.11334251016378402,
"num_tokens": 1880065.0,
"step": 1030
},
{
"entropy": 6.417975521087646,
"epoch": 0.8574979287489644,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999895070773266,
"loss": 6.4608,
"mean_token_accuracy": 0.10985229611396789,
"num_tokens": 1889065.0,
"step": 1035
},
{
"entropy": 6.429206514358521,
"epoch": 0.8616404308202155,
"grad_norm": 1.140625,
"learning_rate": 0.0004999861940344129,
"loss": 6.4721,
"mean_token_accuracy": 0.11190373301506043,
"num_tokens": 1898238.0,
"step": 1040
},
{
"entropy": 6.462369585037232,
"epoch": 0.8657829328914665,
"grad_norm": 1.296875,
"learning_rate": 0.0004999824271697597,
"loss": 6.3929,
"mean_token_accuracy": 0.12136606723070145,
"num_tokens": 1906780.0,
"step": 1045
},
{
"entropy": 6.506088638305664,
"epoch": 0.8699254349627175,
"grad_norm": 1.1484375,
"learning_rate": 0.000499978206490965,
"loss": 6.4745,
"mean_token_accuracy": 0.11197563111782075,
"num_tokens": 1916183.0,
"step": 1050
},
{
"entropy": 6.427097368240356,
"epoch": 0.8740679370339686,
"grad_norm": 1.140625,
"learning_rate": 0.0004999735320065427,
"loss": 6.3371,
"mean_token_accuracy": 0.11927392706274986,
"num_tokens": 1925382.0,
"step": 1055
},
{
"entropy": 6.292429256439209,
"epoch": 0.8782104391052196,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999684037259216,
"loss": 6.1827,
"mean_token_accuracy": 0.1270924136042595,
"num_tokens": 1933639.0,
"step": 1060
},
{
"entropy": 6.3600146770477295,
"epoch": 0.8823529411764706,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999628216594456,
"loss": 6.464,
"mean_token_accuracy": 0.11409285366535186,
"num_tokens": 1943044.0,
"step": 1065
},
{
"entropy": 6.460871028900146,
"epoch": 0.8864954432477217,
"grad_norm": 1.1875,
"learning_rate": 0.000499956785818375,
"loss": 6.303,
"mean_token_accuracy": 0.11871922463178634,
"num_tokens": 1951777.0,
"step": 1070
},
{
"entropy": 6.376981210708618,
"epoch": 0.8906379453189727,
"grad_norm": 1.09375,
"learning_rate": 0.0004999502962148843,
"loss": 6.3666,
"mean_token_accuracy": 0.1170460693538189,
"num_tokens": 1961356.0,
"step": 1075
},
{
"entropy": 6.439280414581299,
"epoch": 0.8947804473902237,
"grad_norm": 1.15625,
"learning_rate": 0.000499943352862064,
"loss": 6.4248,
"mean_token_accuracy": 0.11459407210350037,
"num_tokens": 1970341.0,
"step": 1080
},
{
"entropy": 6.394289827346801,
"epoch": 0.8989229494614748,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999359557739194,
"loss": 6.2796,
"mean_token_accuracy": 0.12198276743292809,
"num_tokens": 1978576.0,
"step": 1085
},
{
"entropy": 6.319045782089233,
"epoch": 0.9030654515327258,
"grad_norm": 1.1875,
"learning_rate": 0.0004999281049653713,
"loss": 6.3227,
"mean_token_accuracy": 0.11601013913750649,
"num_tokens": 1987229.0,
"step": 1090
},
{
"entropy": 6.376017141342163,
"epoch": 0.9072079536039768,
"grad_norm": 1.15625,
"learning_rate": 0.0004999198004522557,
"loss": 6.261,
"mean_token_accuracy": 0.12269015237689018,
"num_tokens": 1996473.0,
"step": 1095
},
{
"entropy": 6.2405853271484375,
"epoch": 0.9113504556752279,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999110422513237,
"loss": 6.286,
"mean_token_accuracy": 0.12056776732206345,
"num_tokens": 2005551.0,
"step": 1100
},
{
"entropy": 6.398191165924072,
"epoch": 0.9154929577464789,
"grad_norm": 1.140625,
"learning_rate": 0.0004999018303802414,
"loss": 6.3869,
"mean_token_accuracy": 0.11947718560695648,
"num_tokens": 2015012.0,
"step": 1105
},
{
"entropy": 6.374700450897217,
"epoch": 0.9196354598177299,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998921648575904,
"loss": 6.3999,
"mean_token_accuracy": 0.1203969657421112,
"num_tokens": 2024435.0,
"step": 1110
},
{
"entropy": 6.380674362182617,
"epoch": 0.923777961888981,
"grad_norm": 1.1484375,
"learning_rate": 0.0004998820457028669,
"loss": 6.3005,
"mean_token_accuracy": 0.12089794874191284,
"num_tokens": 2033991.0,
"step": 1115
},
{
"entropy": 6.409522771835327,
"epoch": 0.927920463960232,
"grad_norm": 1.15625,
"learning_rate": 0.0004998714729364824,
"loss": 6.3017,
"mean_token_accuracy": 0.1228084571659565,
"num_tokens": 2043117.0,
"step": 1120
},
{
"entropy": 6.326260423660278,
"epoch": 0.932062966031483,
"grad_norm": 1.265625,
"learning_rate": 0.0004998604465797634,
"loss": 6.3305,
"mean_token_accuracy": 0.11992078274488449,
"num_tokens": 2052304.0,
"step": 1125
},
{
"entropy": 6.251798295974732,
"epoch": 0.9362054681027341,
"grad_norm": 1.15625,
"learning_rate": 0.0004998489666549512,
"loss": 6.2779,
"mean_token_accuracy": 0.12114684730768203,
"num_tokens": 2061217.0,
"step": 1130
},
{
"entropy": 6.319644594192505,
"epoch": 0.9403479701739851,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998370331852021,
"loss": 6.2517,
"mean_token_accuracy": 0.11879112273454666,
"num_tokens": 2070796.0,
"step": 1135
},
{
"entropy": 6.465651321411133,
"epoch": 0.9444904722452361,
"grad_norm": 1.125,
"learning_rate": 0.0004998246461945872,
"loss": 6.4037,
"mean_token_accuracy": 0.12191760018467904,
"num_tokens": 2080202.0,
"step": 1140
},
{
"entropy": 6.320812797546386,
"epoch": 0.9486329743164872,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998118057080923,
"loss": 6.3045,
"mean_token_accuracy": 0.11891168132424354,
"num_tokens": 2089788.0,
"step": 1145
},
{
"entropy": 6.301202344894409,
"epoch": 0.9527754763877382,
"grad_norm": 1.1875,
"learning_rate": 0.0004997985117516183,
"loss": 6.3462,
"mean_token_accuracy": 0.11142431199550629,
"num_tokens": 2098727.0,
"step": 1150
},
{
"entropy": 6.344118309020996,
"epoch": 0.9569179784589892,
"grad_norm": 1.2109375,
"learning_rate": 0.0004997847643519803,
"loss": 6.226,
"mean_token_accuracy": 0.11924557387828827,
"num_tokens": 2107645.0,
"step": 1155
},
{
"entropy": 6.1790977954864506,
"epoch": 0.9610604805302403,
"grad_norm": 1.2109375,
"learning_rate": 0.0004997705635369085,
"loss": 6.1991,
"mean_token_accuracy": 0.11988648250699044,
"num_tokens": 2116613.0,
"step": 1160
},
{
"entropy": 6.284418439865112,
"epoch": 0.9652029826014913,
"grad_norm": 1.203125,
"learning_rate": 0.0004997559093350474,
"loss": 6.1881,
"mean_token_accuracy": 0.12116428092122078,
"num_tokens": 2125585.0,
"step": 1165
},
{
"entropy": 6.396596622467041,
"epoch": 0.9693454846727423,
"grad_norm": 1.1796875,
"learning_rate": 0.0004997408017759562,
"loss": 6.3359,
"mean_token_accuracy": 0.11936862170696258,
"num_tokens": 2134795.0,
"step": 1170
},
{
"entropy": 6.273751878738404,
"epoch": 0.9734879867439934,
"grad_norm": 1.2265625,
"learning_rate": 0.0004997252408901084,
"loss": 6.3271,
"mean_token_accuracy": 0.11899949684739113,
"num_tokens": 2143699.0,
"step": 1175
},
{
"entropy": 6.386678028106689,
"epoch": 0.9776304888152444,
"grad_norm": 1.09375,
"learning_rate": 0.0004997092267088922,
"loss": 6.2707,
"mean_token_accuracy": 0.11861606612801552,
"num_tokens": 2152187.0,
"step": 1180
},
{
"entropy": 6.275117111206055,
"epoch": 0.9817729908864954,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996927592646097,
"loss": 6.3412,
"mean_token_accuracy": 0.11767763197422028,
"num_tokens": 2160941.0,
"step": 1185
},
{
"entropy": 6.339710283279419,
"epoch": 0.9859154929577465,
"grad_norm": 1.2421875,
"learning_rate": 0.0004996758385904776,
"loss": 6.3119,
"mean_token_accuracy": 0.11789651140570641,
"num_tokens": 2168946.0,
"step": 1190
},
{
"entropy": 6.181374216079712,
"epoch": 0.9900579950289975,
"grad_norm": 1.1875,
"learning_rate": 0.0004996584647206269,
"loss": 6.2237,
"mean_token_accuracy": 0.12055367529392243,
"num_tokens": 2177662.0,
"step": 1195
},
{
"entropy": 6.316773986816406,
"epoch": 0.9942004971002486,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996406376901026,
"loss": 6.3268,
"mean_token_accuracy": 0.12757316455245019,
"num_tokens": 2187461.0,
"step": 1200
},
{
"entropy": 6.283385133743286,
"epoch": 0.9983429991714996,
"grad_norm": 1.109375,
"learning_rate": 0.0004996223575348635,
"loss": 6.2207,
"mean_token_accuracy": 0.12578138560056687,
"num_tokens": 2196878.0,
"step": 1205
},
{
"entropy": 6.311019659042358,
"epoch": 1.0024855012427507,
"grad_norm": 1.125,
"learning_rate": 0.0004996036242917828,
"loss": 6.2021,
"mean_token_accuracy": 0.1151522897183895,
"num_tokens": 2206178.0,
"step": 1210
},
{
"entropy": 6.25421667098999,
"epoch": 1.0066280033140016,
"grad_norm": 1.09375,
"learning_rate": 0.0004995844379986476,
"loss": 6.0526,
"mean_token_accuracy": 0.12476176023483276,
"num_tokens": 2215330.0,
"step": 1215
},
{
"entropy": 6.309921216964722,
"epoch": 1.0107705053852527,
"grad_norm": 1.078125,
"learning_rate": 0.0004995647986941587,
"loss": 6.0354,
"mean_token_accuracy": 0.13005029112100602,
"num_tokens": 2224146.0,
"step": 1220
},
{
"entropy": 6.241363286972046,
"epoch": 1.0149130074565038,
"grad_norm": 1.0625,
"learning_rate": 0.0004995447064179308,
"loss": 6.0171,
"mean_token_accuracy": 0.13421598002314566,
"num_tokens": 2233646.0,
"step": 1225
},
{
"entropy": 6.058859920501709,
"epoch": 1.0190555095277547,
"grad_norm": 1.0234375,
"learning_rate": 0.000499524161210492,
"loss": 5.9379,
"mean_token_accuracy": 0.12560308948159218,
"num_tokens": 2243663.0,
"step": 1230
},
{
"entropy": 6.270004224777222,
"epoch": 1.0231980115990058,
"grad_norm": 1.140625,
"learning_rate": 0.0004995031631132844,
"loss": 6.1035,
"mean_token_accuracy": 0.13170127123594283,
"num_tokens": 2252564.0,
"step": 1235
},
{
"entropy": 6.2710247993469235,
"epoch": 1.027340513670257,
"grad_norm": 1.1328125,
"learning_rate": 0.0004994817121686633,
"loss": 6.099,
"mean_token_accuracy": 0.12465815395116805,
"num_tokens": 2262504.0,
"step": 1240
},
{
"entropy": 6.193396997451782,
"epoch": 1.0314830157415078,
"grad_norm": 1.203125,
"learning_rate": 0.0004994598084198979,
"loss": 5.9843,
"mean_token_accuracy": 0.1374910719692707,
"num_tokens": 2270553.0,
"step": 1245
},
{
"entropy": 6.136552810668945,
"epoch": 1.035625517812759,
"grad_norm": 1.21875,
"learning_rate": 0.0004994374519111701,
"loss": 5.977,
"mean_token_accuracy": 0.1272726461291313,
"num_tokens": 2279319.0,
"step": 1250
},
{
"entropy": 6.217378282546997,
"epoch": 1.03976801988401,
"grad_norm": 1.109375,
"learning_rate": 0.0004994146426875757,
"loss": 6.123,
"mean_token_accuracy": 0.126935476064682,
"num_tokens": 2288749.0,
"step": 1255
},
{
"entropy": 6.218695068359375,
"epoch": 1.043910521955261,
"grad_norm": 1.078125,
"learning_rate": 0.0004993913807951232,
"loss": 6.0156,
"mean_token_accuracy": 0.12562426924705505,
"num_tokens": 2298272.0,
"step": 1260
},
{
"entropy": 6.221994352340698,
"epoch": 1.048053024026512,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993676662807345,
"loss": 6.047,
"mean_token_accuracy": 0.12884460240602494,
"num_tokens": 2307230.0,
"step": 1265
},
{
"entropy": 6.183428144454956,
"epoch": 1.0521955260977631,
"grad_norm": 1.171875,
"learning_rate": 0.0004993434991922443,
"loss": 6.1317,
"mean_token_accuracy": 0.12460145726799965,
"num_tokens": 2316306.0,
"step": 1270
},
{
"entropy": 6.22090425491333,
"epoch": 1.056338028169014,
"grad_norm": 1.03125,
"learning_rate": 0.0004993188795784002,
"loss": 6.0531,
"mean_token_accuracy": 0.1241947129368782,
"num_tokens": 2325672.0,
"step": 1275
},
{
"entropy": 6.1453857421875,
"epoch": 1.0604805302402651,
"grad_norm": 1.21875,
"learning_rate": 0.0004992938074888628,
"loss": 6.066,
"mean_token_accuracy": 0.1274007849395275,
"num_tokens": 2335251.0,
"step": 1280
},
{
"entropy": 6.180039167404175,
"epoch": 1.0646230323115162,
"grad_norm": 1.2109375,
"learning_rate": 0.0004992682829742052,
"loss": 6.0216,
"mean_token_accuracy": 0.1308726042509079,
"num_tokens": 2344054.0,
"step": 1285
},
{
"entropy": 6.086296463012696,
"epoch": 1.0687655343827671,
"grad_norm": 1.2421875,
"learning_rate": 0.0004992423060859132,
"loss": 6.0236,
"mean_token_accuracy": 0.127706591039896,
"num_tokens": 2353493.0,
"step": 1290
},
{
"entropy": 6.223158740997315,
"epoch": 1.0729080364540182,
"grad_norm": 1.140625,
"learning_rate": 0.0004992158768763849,
"loss": 6.1321,
"mean_token_accuracy": 0.12850543335080147,
"num_tokens": 2362687.0,
"step": 1295
},
{
"entropy": 6.2506636619567875,
"epoch": 1.0770505385252693,
"grad_norm": 1.078125,
"learning_rate": 0.0004991889953989308,
"loss": 6.0965,
"mean_token_accuracy": 0.12563316524028778,
"num_tokens": 2372325.0,
"step": 1300
},
{
"entropy": 6.173602199554443,
"epoch": 1.0811930405965202,
"grad_norm": 1.1171875,
"learning_rate": 0.0004991616617077741,
"loss": 6.12,
"mean_token_accuracy": 0.1287991665303707,
"num_tokens": 2381987.0,
"step": 1305
},
{
"entropy": 6.307486963272095,
"epoch": 1.0853355426677713,
"grad_norm": 1.2421875,
"learning_rate": 0.0004991338758580496,
"loss": 6.0234,
"mean_token_accuracy": 0.13178783878684044,
"num_tokens": 2390901.0,
"step": 1310
},
{
"entropy": 6.034789133071899,
"epoch": 1.0894780447390224,
"grad_norm": 1.0390625,
"learning_rate": 0.0004991056379058045,
"loss": 6.0813,
"mean_token_accuracy": 0.12450421676039695,
"num_tokens": 2400756.0,
"step": 1315
},
{
"entropy": 6.188017988204956,
"epoch": 1.0936205468102733,
"grad_norm": 1.1171875,
"learning_rate": 0.0004990769479079976,
"loss": 5.9232,
"mean_token_accuracy": 0.12984196096658707,
"num_tokens": 2409526.0,
"step": 1320
},
{
"entropy": 6.086554956436157,
"epoch": 1.0977630488815244,
"grad_norm": 0.984375,
"learning_rate": 0.0004990478059225003,
"loss": 6.0931,
"mean_token_accuracy": 0.12972966134548186,
"num_tokens": 2419583.0,
"step": 1325
},
{
"entropy": 6.2457873821258545,
"epoch": 1.1019055509527755,
"grad_norm": 1.0390625,
"learning_rate": 0.0004990182120080945,
"loss": 6.0338,
"mean_token_accuracy": 0.12990806251764297,
"num_tokens": 2429439.0,
"step": 1330
},
{
"entropy": 6.078778219223023,
"epoch": 1.1060480530240264,
"grad_norm": 1.0859375,
"learning_rate": 0.0004989881662244748,
"loss": 6.0416,
"mean_token_accuracy": 0.12380676046013832,
"num_tokens": 2438467.0,
"step": 1335
},
{
"entropy": 6.138932418823242,
"epoch": 1.1101905550952775,
"grad_norm": 1.1796875,
"learning_rate": 0.0004989576686322468,
"loss": 5.9919,
"mean_token_accuracy": 0.1332716926932335,
"num_tokens": 2446619.0,
"step": 1340
},
{
"entropy": 6.1281578540802,
"epoch": 1.1143330571665286,
"grad_norm": 1.1875,
"learning_rate": 0.0004989267192929272,
"loss": 6.0379,
"mean_token_accuracy": 0.12955363765358924,
"num_tokens": 2454998.0,
"step": 1345
},
{
"entropy": 6.09554443359375,
"epoch": 1.1184755592377795,
"grad_norm": 1.1640625,
"learning_rate": 0.0004988953182689445,
"loss": 6.0072,
"mean_token_accuracy": 0.13532012328505516,
"num_tokens": 2463497.0,
"step": 1350
},
{
"entropy": 6.134703922271728,
"epoch": 1.1226180613090306,
"grad_norm": 1.03125,
"learning_rate": 0.0004988634656236378,
"loss": 5.9216,
"mean_token_accuracy": 0.13609320148825646,
"num_tokens": 2471763.0,
"step": 1355
},
{
"entropy": 6.145527791976929,
"epoch": 1.1267605633802817,
"grad_norm": 1.21875,
"learning_rate": 0.0004988311614212576,
"loss": 6.0016,
"mean_token_accuracy": 0.1402701161801815,
"num_tokens": 2479949.0,
"step": 1360
},
{
"entropy": 6.116470623016357,
"epoch": 1.1309030654515326,
"grad_norm": 1.1484375,
"learning_rate": 0.0004987984057269648,
"loss": 6.0539,
"mean_token_accuracy": 0.13444141745567323,
"num_tokens": 2489097.0,
"step": 1365
},
{
"entropy": 5.972180604934692,
"epoch": 1.1350455675227837,
"grad_norm": 1.0859375,
"learning_rate": 0.0004987651986068311,
"loss": 5.8781,
"mean_token_accuracy": 0.13601028621196748,
"num_tokens": 2497625.0,
"step": 1370
},
{
"entropy": 6.143597984313965,
"epoch": 1.1391880695940348,
"grad_norm": 1.1015625,
"learning_rate": 0.0004987315401278394,
"loss": 6.1252,
"mean_token_accuracy": 0.1272761717438698,
"num_tokens": 2507107.0,
"step": 1375
},
{
"entropy": 6.286170959472656,
"epoch": 1.143330571665286,
"grad_norm": 1.265625,
"learning_rate": 0.0004986974303578819,
"loss": 6.1408,
"mean_token_accuracy": 0.12348597794771195,
"num_tokens": 2516629.0,
"step": 1380
},
{
"entropy": 6.077923250198364,
"epoch": 1.1474730737365368,
"grad_norm": 1.125,
"learning_rate": 0.0004986628693657621,
"loss": 6.021,
"mean_token_accuracy": 0.1280151717364788,
"num_tokens": 2525924.0,
"step": 1385
},
{
"entropy": 6.229060649871826,
"epoch": 1.151615575807788,
"grad_norm": 1.3125,
"learning_rate": 0.0004986278572211933,
"loss": 6.1723,
"mean_token_accuracy": 0.1280426375567913,
"num_tokens": 2535586.0,
"step": 1390
},
{
"entropy": 6.189875984191895,
"epoch": 1.1557580778790388,
"grad_norm": 1.015625,
"learning_rate": 0.0004985923939947987,
"loss": 6.2595,
"mean_token_accuracy": 0.1247698113322258,
"num_tokens": 2546685.0,
"step": 1395
},
{
"entropy": 6.148470640182495,
"epoch": 1.15990057995029,
"grad_norm": 1.1484375,
"learning_rate": 0.0004985564797581115,
"loss": 5.9681,
"mean_token_accuracy": 0.12971177995204924,
"num_tokens": 2556873.0,
"step": 1400
},
{
"entropy": 6.143459701538086,
"epoch": 1.164043082021541,
"grad_norm": 1.125,
"learning_rate": 0.0004985201145835747,
"loss": 5.9836,
"mean_token_accuracy": 0.13127154856920242,
"num_tokens": 2565222.0,
"step": 1405
},
{
"entropy": 6.031957960128784,
"epoch": 1.1681855840927922,
"grad_norm": 1.046875,
"learning_rate": 0.000498483298544541,
"loss": 5.9875,
"mean_token_accuracy": 0.13211632743477822,
"num_tokens": 2575467.0,
"step": 1410
},
{
"entropy": 6.136318731307983,
"epoch": 1.172328086164043,
"grad_norm": 1.2109375,
"learning_rate": 0.000498446031715272,
"loss": 6.0125,
"mean_token_accuracy": 0.13508019223809242,
"num_tokens": 2583963.0,
"step": 1415
},
{
"entropy": 6.133963680267334,
"epoch": 1.1764705882352942,
"grad_norm": 1.109375,
"learning_rate": 0.0004984083141709391,
"loss": 6.042,
"mean_token_accuracy": 0.13524071872234344,
"num_tokens": 2593350.0,
"step": 1420
},
{
"entropy": 5.975436115264893,
"epoch": 1.180613090306545,
"grad_norm": 1.0859375,
"learning_rate": 0.000498370145987623,
"loss": 5.9652,
"mean_token_accuracy": 0.13682079091668128,
"num_tokens": 2602191.0,
"step": 1425
},
{
"entropy": 6.170718050003051,
"epoch": 1.1847555923777962,
"grad_norm": 0.984375,
"learning_rate": 0.0004983315272423127,
"loss": 6.0999,
"mean_token_accuracy": 0.1291058838367462,
"num_tokens": 2613590.0,
"step": 1430
},
{
"entropy": 6.049332237243652,
"epoch": 1.1888980944490473,
"grad_norm": 1.1640625,
"learning_rate": 0.0004982924580129066,
"loss": 5.8895,
"mean_token_accuracy": 0.1340464949607849,
"num_tokens": 2622072.0,
"step": 1435
},
{
"entropy": 6.093293142318726,
"epoch": 1.1930405965202984,
"grad_norm": 1.1796875,
"learning_rate": 0.0004982529383782116,
"loss": 6.1012,
"mean_token_accuracy": 0.13022035285830497,
"num_tokens": 2630904.0,
"step": 1440
},
{
"entropy": 6.189996242523193,
"epoch": 1.1971830985915493,
"grad_norm": 1.1171875,
"learning_rate": 0.0004982129684179432,
"loss": 6.0594,
"mean_token_accuracy": 0.13088259026408194,
"num_tokens": 2639925.0,
"step": 1445
},
{
"entropy": 6.1187131881713865,
"epoch": 1.2013256006628004,
"grad_norm": 1.21875,
"learning_rate": 0.0004981725482127251,
"loss": 5.977,
"mean_token_accuracy": 0.13439981341362,
"num_tokens": 2649397.0,
"step": 1450
},
{
"entropy": 6.16315655708313,
"epoch": 1.2054681027340513,
"grad_norm": 1.078125,
"learning_rate": 0.0004981316778440892,
"loss": 6.081,
"mean_token_accuracy": 0.13515593931078912,
"num_tokens": 2658877.0,
"step": 1455
},
{
"entropy": 6.063838720321655,
"epoch": 1.2096106048053024,
"grad_norm": 1.1640625,
"learning_rate": 0.0004980903573944756,
"loss": 6.0171,
"mean_token_accuracy": 0.13121431767940522,
"num_tokens": 2668518.0,
"step": 1460
},
{
"entropy": 6.059180212020874,
"epoch": 1.2137531068765535,
"grad_norm": 1.0078125,
"learning_rate": 0.0004980485869472322,
"loss": 6.1148,
"mean_token_accuracy": 0.12996160089969636,
"num_tokens": 2677459.0,
"step": 1465
},
{
"entropy": 6.223527908325195,
"epoch": 1.2178956089478046,
"grad_norm": 1.1484375,
"learning_rate": 0.0004980063665866143,
"loss": 6.0758,
"mean_token_accuracy": 0.12482839822769165,
"num_tokens": 2686668.0,
"step": 1470
},
{
"entropy": 5.935272264480591,
"epoch": 1.2220381110190555,
"grad_norm": 1.1640625,
"learning_rate": 0.0004979636963977853,
"loss": 5.8825,
"mean_token_accuracy": 0.1352998949587345,
"num_tokens": 2695545.0,
"step": 1475
},
{
"entropy": 6.066101598739624,
"epoch": 1.2261806130903066,
"grad_norm": 1.1328125,
"learning_rate": 0.0004979205764668156,
"loss": 6.0569,
"mean_token_accuracy": 0.13228241056203843,
"num_tokens": 2705514.0,
"step": 1480
},
{
"entropy": 6.138315343856812,
"epoch": 1.2303231151615575,
"grad_norm": 1.1875,
"learning_rate": 0.0004978770068806825,
"loss": 5.9599,
"mean_token_accuracy": 0.13136665150523186,
"num_tokens": 2714373.0,
"step": 1485
},
{
"entropy": 6.061770391464234,
"epoch": 1.2344656172328086,
"grad_norm": 1.09375,
"learning_rate": 0.000497832987727271,
"loss": 6.0715,
"mean_token_accuracy": 0.1264137916266918,
"num_tokens": 2723887.0,
"step": 1490
},
{
"entropy": 6.061756038665772,
"epoch": 1.2386081193040597,
"grad_norm": 1.125,
"learning_rate": 0.0004977885190953724,
"loss": 5.956,
"mean_token_accuracy": 0.13353569880127908,
"num_tokens": 2733591.0,
"step": 1495
},
{
"entropy": 6.104292392730713,
"epoch": 1.2427506213753108,
"grad_norm": 1.21875,
"learning_rate": 0.0004977436010746846,
"loss": 6.066,
"mean_token_accuracy": 0.13720986992120743,
"num_tokens": 2741638.0,
"step": 1500
},
{
"epoch": 1.2427506213753108,
"eval_entropy": 6.029743814773397,
"eval_loss": 6.2611613273620605,
"eval_mean_token_accuracy": 0.12890694399219332,
"eval_num_tokens": 2741638.0,
"eval_runtime": 2.9518,
"eval_samples_per_second": 1269.399,
"eval_steps_per_second": 158.887,
"step": 1500
},
{
"entropy": 6.078821802139283,
"epoch": 1.2468931234465617,
"grad_norm": 1.296875,
"learning_rate": 0.0004976982337558124,
"loss": 5.9412,
"mean_token_accuracy": 0.13084649220108985,
"num_tokens": 2749711.0,
"step": 1505
},
{
"entropy": 6.094208097457885,
"epoch": 1.2510356255178128,
"grad_norm": 1.1015625,
"learning_rate": 0.0004976524172302665,
"loss": 6.121,
"mean_token_accuracy": 0.13192041367292404,
"num_tokens": 2758121.0,
"step": 1510
},
{
"entropy": 6.159745693206787,
"epoch": 1.2551781275890637,
"grad_norm": 1.0,
"learning_rate": 0.0004976061515904641,
"loss": 6.1181,
"mean_token_accuracy": 0.12667254582047463,
"num_tokens": 2766764.0,
"step": 1515
},
{
"entropy": 6.116680526733399,
"epoch": 1.2593206296603148,
"grad_norm": 1.1953125,
"learning_rate": 0.0004975594369297277,
"loss": 6.0557,
"mean_token_accuracy": 0.1348518520593643,
"num_tokens": 2775561.0,
"step": 1520
},
{
"entropy": 6.08063235282898,
"epoch": 1.2634631317315659,
"grad_norm": 1.109375,
"learning_rate": 0.0004975122733422861,
"loss": 6.0201,
"mean_token_accuracy": 0.13085112273693084,
"num_tokens": 2784691.0,
"step": 1525
},
{
"entropy": 6.129113435745239,
"epoch": 1.267605633802817,
"grad_norm": 1.1875,
"learning_rate": 0.0004974646609232732,
"loss": 5.9816,
"mean_token_accuracy": 0.13641836866736412,
"num_tokens": 2793893.0,
"step": 1530
},
{
"entropy": 6.108260202407837,
"epoch": 1.2717481358740679,
"grad_norm": 1.046875,
"learning_rate": 0.0004974165997687288,
"loss": 6.0751,
"mean_token_accuracy": 0.13353196606040002,
"num_tokens": 2804479.0,
"step": 1535
},
{
"entropy": 6.16022138595581,
"epoch": 1.275890637945319,
"grad_norm": 1.1484375,
"learning_rate": 0.0004973680899755972,
"loss": 6.107,
"mean_token_accuracy": 0.1326654002070427,
"num_tokens": 2813966.0,
"step": 1540
},
{
"entropy": 6.007886743545532,
"epoch": 1.2800331400165699,
"grad_norm": 1.1953125,
"learning_rate": 0.000497319131641728,
"loss": 5.9326,
"mean_token_accuracy": 0.13846119418740271,
"num_tokens": 2822328.0,
"step": 1545
},
{
"entropy": 6.003260755538941,
"epoch": 1.284175642087821,
"grad_norm": 1.2578125,
"learning_rate": 0.0004972697248658755,
"loss": 5.8899,
"mean_token_accuracy": 0.14066707715392113,
"num_tokens": 2829845.0,
"step": 1550
},
{
"entropy": 5.948832178115845,
"epoch": 1.288318144159072,
"grad_norm": 1.2265625,
"learning_rate": 0.0004972198697476987,
"loss": 5.8783,
"mean_token_accuracy": 0.13914238885045052,
"num_tokens": 2839040.0,
"step": 1555
},
{
"entropy": 5.961882209777832,
"epoch": 1.2924606462303232,
"grad_norm": 1.03125,
"learning_rate": 0.0004971695663877607,
"loss": 5.9665,
"mean_token_accuracy": 0.13561231270432472,
"num_tokens": 2848574.0,
"step": 1560
},
{
"entropy": 6.077086877822876,
"epoch": 1.296603148301574,
"grad_norm": 1.2265625,
"learning_rate": 0.0004971188148875289,
"loss": 5.9421,
"mean_token_accuracy": 0.14444561377167703,
"num_tokens": 2856812.0,
"step": 1565
},
{
"entropy": 6.051652336120606,
"epoch": 1.3007456503728252,
"grad_norm": 1.140625,
"learning_rate": 0.0004970676153493748,
"loss": 6.0283,
"mean_token_accuracy": 0.13189572617411613,
"num_tokens": 2866001.0,
"step": 1570
},
{
"entropy": 6.1068840503692625,
"epoch": 1.3048881524440763,
"grad_norm": 1.0859375,
"learning_rate": 0.0004970159678765731,
"loss": 6.0164,
"mean_token_accuracy": 0.13239546120166779,
"num_tokens": 2875181.0,
"step": 1575
},
{
"entropy": 6.049808597564697,
"epoch": 1.3090306545153272,
"grad_norm": 1.15625,
"learning_rate": 0.0004969638725733028,
"loss": 5.9587,
"mean_token_accuracy": 0.13397156223654746,
"num_tokens": 2885103.0,
"step": 1580
},
{
"entropy": 5.92131872177124,
"epoch": 1.3131731565865783,
"grad_norm": 1.0859375,
"learning_rate": 0.0004969113295446455,
"loss": 5.9116,
"mean_token_accuracy": 0.1383065924048424,
"num_tokens": 2894055.0,
"step": 1585
},
{
"entropy": 6.116864538192749,
"epoch": 1.3173156586578294,
"grad_norm": 1.0859375,
"learning_rate": 0.0004968583388965863,
"loss": 5.9846,
"mean_token_accuracy": 0.133075200766325,
"num_tokens": 2903391.0,
"step": 1590
},
{
"entropy": 6.006745624542236,
"epoch": 1.3214581607290803,
"grad_norm": 1.1875,
"learning_rate": 0.0004968049007360132,
"loss": 5.9854,
"mean_token_accuracy": 0.13297903537750244,
"num_tokens": 2911760.0,
"step": 1595
},
{
"entropy": 5.926331329345703,
"epoch": 1.3256006628003314,
"grad_norm": 1.125,
"learning_rate": 0.0004967510151707169,
"loss": 5.9369,
"mean_token_accuracy": 0.13469143435359002,
"num_tokens": 2920589.0,
"step": 1600
},
{
"entropy": 6.15295524597168,
"epoch": 1.3297431648715825,
"grad_norm": 1.1640625,
"learning_rate": 0.0004966966823093903,
"loss": 6.0746,
"mean_token_accuracy": 0.1302391268312931,
"num_tokens": 2929635.0,
"step": 1605
},
{
"entropy": 5.986531114578247,
"epoch": 1.3338856669428334,
"grad_norm": 1.140625,
"learning_rate": 0.0004966419022616287,
"loss": 5.9349,
"mean_token_accuracy": 0.13593596145510672,
"num_tokens": 2938778.0,
"step": 1610
},
{
"entropy": 6.039755535125733,
"epoch": 1.3380281690140845,
"grad_norm": 1.3125,
"learning_rate": 0.0004965866751379296,
"loss": 6.0503,
"mean_token_accuracy": 0.13109695687890052,
"num_tokens": 2947836.0,
"step": 1615
},
{
"entropy": 6.133180904388428,
"epoch": 1.3421706710853356,
"grad_norm": 1.125,
"learning_rate": 0.0004965310010496922,
"loss": 6.0032,
"mean_token_accuracy": 0.1280514419078827,
"num_tokens": 2957028.0,
"step": 1620
},
{
"entropy": 6.026159048080444,
"epoch": 1.3463131731565865,
"grad_norm": 1.109375,
"learning_rate": 0.0004964748801092168,
"loss": 5.9571,
"mean_token_accuracy": 0.1338861681520939,
"num_tokens": 2966088.0,
"step": 1625
},
{
"entropy": 6.078959226608276,
"epoch": 1.3504556752278376,
"grad_norm": 1.0703125,
"learning_rate": 0.0004964183124297059,
"loss": 6.0438,
"mean_token_accuracy": 0.13178688064217567,
"num_tokens": 2975271.0,
"step": 1630
},
{
"entropy": 5.992210865020752,
"epoch": 1.3545981772990887,
"grad_norm": 1.1015625,
"learning_rate": 0.0004963612981252625,
"loss": 6.0524,
"mean_token_accuracy": 0.1321226365864277,
"num_tokens": 2983799.0,
"step": 1635
},
{
"entropy": 6.056794929504394,
"epoch": 1.3587406793703396,
"grad_norm": 1.0859375,
"learning_rate": 0.0004963038373108908,
"loss": 6.0179,
"mean_token_accuracy": 0.13609340488910676,
"num_tokens": 2992675.0,
"step": 1640
},
{
"entropy": 6.155933427810669,
"epoch": 1.3628831814415907,
"grad_norm": 1.0703125,
"learning_rate": 0.0004962459301024954,
"loss": 6.0838,
"mean_token_accuracy": 0.13211635649204254,
"num_tokens": 3002483.0,
"step": 1645
},
{
"entropy": 6.027872896194458,
"epoch": 1.3670256835128418,
"grad_norm": 1.109375,
"learning_rate": 0.0004961875766168816,
"loss": 5.985,
"mean_token_accuracy": 0.13883458971977233,
"num_tokens": 3010611.0,
"step": 1650
},
{
"entropy": 6.057289028167725,
"epoch": 1.371168185584093,
"grad_norm": 1.046875,
"learning_rate": 0.0004961287769717547,
"loss": 5.9397,
"mean_token_accuracy": 0.13197456523776055,
"num_tokens": 3019928.0,
"step": 1655
},
{
"entropy": 5.9325989246368405,
"epoch": 1.3753106876553438,
"grad_norm": 1.1015625,
"learning_rate": 0.0004960695312857199,
"loss": 5.9279,
"mean_token_accuracy": 0.13408055454492568,
"num_tokens": 3029003.0,
"step": 1660
},
{
"entropy": 6.0389604568481445,
"epoch": 1.379453189726595,
"grad_norm": 1.109375,
"learning_rate": 0.0004960098396782824,
"loss": 5.9464,
"mean_token_accuracy": 0.1375861033797264,
"num_tokens": 3038049.0,
"step": 1665
},
{
"entropy": 5.969662141799927,
"epoch": 1.3835956917978458,
"grad_norm": 1.03125,
"learning_rate": 0.0004959497022698466,
"loss": 5.9515,
"mean_token_accuracy": 0.14408120438456534,
"num_tokens": 3046674.0,
"step": 1670
},
{
"entropy": 6.0561059474945065,
"epoch": 1.387738193869097,
"grad_norm": 1.1484375,
"learning_rate": 0.0004958891191817163,
"loss": 5.9927,
"mean_token_accuracy": 0.12801324650645257,
"num_tokens": 3056459.0,
"step": 1675
},
{
"entropy": 6.032327270507812,
"epoch": 1.391880695940348,
"grad_norm": 1.1796875,
"learning_rate": 0.000495828090536094,
"loss": 5.9845,
"mean_token_accuracy": 0.13036760538816453,
"num_tokens": 3066294.0,
"step": 1680
},
{
"entropy": 6.0512574195861815,
"epoch": 1.3960231980115991,
"grad_norm": 1.1796875,
"learning_rate": 0.0004957666164560815,
"loss": 5.9412,
"mean_token_accuracy": 0.13139644563198088,
"num_tokens": 3075420.0,
"step": 1685
},
{
"entropy": 6.063203239440918,
"epoch": 1.40016570008285,
"grad_norm": 1.140625,
"learning_rate": 0.0004957046970656784,
"loss": 6.0385,
"mean_token_accuracy": 0.13581736162304878,
"num_tokens": 3084373.0,
"step": 1690
},
{
"entropy": 6.00097541809082,
"epoch": 1.4043082021541011,
"grad_norm": 1.09375,
"learning_rate": 0.0004956423324897831,
"loss": 6.0257,
"mean_token_accuracy": 0.13937310948967935,
"num_tokens": 3094212.0,
"step": 1695
},
{
"entropy": 6.043115425109863,
"epoch": 1.408450704225352,
"grad_norm": 1.09375,
"learning_rate": 0.0004955795228541917,
"loss": 5.9589,
"mean_token_accuracy": 0.14452596232295037,
"num_tokens": 3102769.0,
"step": 1700
},
{
"entropy": 6.040110731124878,
"epoch": 1.4125932062966031,
"grad_norm": 1.171875,
"learning_rate": 0.000495516268285598,
"loss": 5.9805,
"mean_token_accuracy": 0.13440805301070213,
"num_tokens": 3111465.0,
"step": 1705
},
{
"entropy": 5.978239822387695,
"epoch": 1.4167357083678542,
"grad_norm": 1.078125,
"learning_rate": 0.0004954525689115935,
"loss": 5.9177,
"mean_token_accuracy": 0.13326097056269645,
"num_tokens": 3121106.0,
"step": 1710
},
{
"entropy": 6.007881164550781,
"epoch": 1.4208782104391053,
"grad_norm": 1.1796875,
"learning_rate": 0.0004953884248606669,
"loss": 5.945,
"mean_token_accuracy": 0.14831372052431108,
"num_tokens": 3130662.0,
"step": 1715
},
{
"entropy": 5.97642674446106,
"epoch": 1.4250207125103562,
"grad_norm": 1.140625,
"learning_rate": 0.0004953238362622034,
"loss": 5.9812,
"mean_token_accuracy": 0.13328263983130456,
"num_tokens": 3139538.0,
"step": 1720
},
{
"entropy": 6.0256367206573485,
"epoch": 1.4291632145816073,
"grad_norm": 1.1171875,
"learning_rate": 0.0004952588032464855,
"loss": 6.0298,
"mean_token_accuracy": 0.12707214802503586,
"num_tokens": 3148552.0,
"step": 1725
},
{
"entropy": 5.996851205825806,
"epoch": 1.4333057166528582,
"grad_norm": 1.15625,
"learning_rate": 0.0004951933259446919,
"loss": 5.9365,
"mean_token_accuracy": 0.14197203740477563,
"num_tokens": 3156549.0,
"step": 1730
},
{
"entropy": 6.114988136291504,
"epoch": 1.4374482187241093,
"grad_norm": 1.0859375,
"learning_rate": 0.0004951274044888975,
"loss": 6.1006,
"mean_token_accuracy": 0.12147112190723419,
"num_tokens": 3166821.0,
"step": 1735
},
{
"entropy": 5.960144329071045,
"epoch": 1.4415907207953604,
"grad_norm": 1.203125,
"learning_rate": 0.0004950610390120731,
"loss": 5.9848,
"mean_token_accuracy": 0.135729306191206,
"num_tokens": 3174962.0,
"step": 1740
},
{
"entropy": 6.087535095214844,
"epoch": 1.4457332228666115,
"grad_norm": 1.0703125,
"learning_rate": 0.000494994229648085,
"loss": 5.8903,
"mean_token_accuracy": 0.13537728488445283,
"num_tokens": 3184027.0,
"step": 1745
},
{
"entropy": 5.957475805282593,
"epoch": 1.4498757249378624,
"grad_norm": 1.1171875,
"learning_rate": 0.0004949269765316948,
"loss": 6.0321,
"mean_token_accuracy": 0.13071285635232927,
"num_tokens": 3193186.0,
"step": 1750
},
{
"entropy": 6.171153354644775,
"epoch": 1.4540182270091135,
"grad_norm": 2.140625,
"learning_rate": 0.0004948592797985599,
"loss": 6.0525,
"mean_token_accuracy": 0.13555259853601456,
"num_tokens": 3202676.0,
"step": 1755
},
{
"entropy": 6.123908805847168,
"epoch": 1.4581607290803644,
"grad_norm": 1.1484375,
"learning_rate": 0.0004947911395852317,
"loss": 6.1524,
"mean_token_accuracy": 0.1352410912513733,
"num_tokens": 3211950.0,
"step": 1760
},
{
"entropy": 6.120599699020386,
"epoch": 1.4623032311516155,
"grad_norm": 1.1484375,
"learning_rate": 0.0004947225560291565,
"loss": 6.0758,
"mean_token_accuracy": 0.12961760386824608,
"num_tokens": 3221135.0,
"step": 1765
},
{
"entropy": 5.962483358383179,
"epoch": 1.4664457332228666,
"grad_norm": 1.1484375,
"learning_rate": 0.0004946535292686746,
"loss": 5.8975,
"mean_token_accuracy": 0.13825839310884475,
"num_tokens": 3230174.0,
"step": 1770
},
{
"entropy": 5.979943370819091,
"epoch": 1.4705882352941178,
"grad_norm": 1.2109375,
"learning_rate": 0.0004945840594430209,
"loss": 5.8871,
"mean_token_accuracy": 0.1381668671965599,
"num_tokens": 3238898.0,
"step": 1775
},
{
"entropy": 5.973515844345092,
"epoch": 1.4747307373653686,
"grad_norm": 1.1015625,
"learning_rate": 0.0004945141466923235,
"loss": 5.9292,
"mean_token_accuracy": 0.14479121565818787,
"num_tokens": 3247550.0,
"step": 1780
},
{
"entropy": 6.113377714157105,
"epoch": 1.4788732394366197,
"grad_norm": 1.125,
"learning_rate": 0.0004944437911576041,
"loss": 6.0779,
"mean_token_accuracy": 0.12900342419743538,
"num_tokens": 3256810.0,
"step": 1785
},
{
"entropy": 5.9020613670349125,
"epoch": 1.4830157415078706,
"grad_norm": 1.109375,
"learning_rate": 0.0004943729929807772,
"loss": 5.877,
"mean_token_accuracy": 0.139550269395113,
"num_tokens": 3265886.0,
"step": 1790
},
{
"entropy": 6.06021089553833,
"epoch": 1.4871582435791217,
"grad_norm": 1.1328125,
"learning_rate": 0.0004943017523046507,
"loss": 5.9273,
"mean_token_accuracy": 0.137468421459198,
"num_tokens": 3274522.0,
"step": 1795
},
{
"entropy": 6.027210426330567,
"epoch": 1.4913007456503728,
"grad_norm": 1.25,
"learning_rate": 0.000494230069272925,
"loss": 6.0486,
"mean_token_accuracy": 0.13558472618460654,
"num_tokens": 3283388.0,
"step": 1800
},
{
"entropy": 5.989331531524658,
"epoch": 1.495443247721624,
"grad_norm": 1.1640625,
"learning_rate": 0.0004941579440301923,
"loss": 5.9224,
"mean_token_accuracy": 0.1305856943130493,
"num_tokens": 3291887.0,
"step": 1805
},
{
"entropy": 5.96136965751648,
"epoch": 1.4995857497928748,
"grad_norm": 1.0625,
"learning_rate": 0.0004940853767219372,
"loss": 5.9685,
"mean_token_accuracy": 0.13564062863588333,
"num_tokens": 3300674.0,
"step": 1810
},
{
"entropy": 5.9364429950714115,
"epoch": 1.503728251864126,
"grad_norm": 1.0625,
"learning_rate": 0.0004940123674945359,
"loss": 5.8311,
"mean_token_accuracy": 0.13808463513851166,
"num_tokens": 3309745.0,
"step": 1815
},
{
"entropy": 6.076629638671875,
"epoch": 1.5078707539353768,
"grad_norm": 1.234375,
"learning_rate": 0.0004939389164952561,
"loss": 5.9722,
"mean_token_accuracy": 0.1378110609948635,
"num_tokens": 3318888.0,
"step": 1820
},
{
"entropy": 5.8970863819122314,
"epoch": 1.512013256006628,
"grad_norm": 1.234375,
"learning_rate": 0.0004938650238722564,
"loss": 5.9296,
"mean_token_accuracy": 0.13823849260807036,
"num_tokens": 3328274.0,
"step": 1825
},
{
"entropy": 6.003948831558228,
"epoch": 1.516155758077879,
"grad_norm": 1.1875,
"learning_rate": 0.0004937906897745861,
"loss": 5.9387,
"mean_token_accuracy": 0.1468382939696312,
"num_tokens": 3337579.0,
"step": 1830
},
{
"entropy": 6.045965003967285,
"epoch": 1.5202982601491302,
"grad_norm": 1.2734375,
"learning_rate": 0.0004937159143521856,
"loss": 5.9081,
"mean_token_accuracy": 0.14555162489414214,
"num_tokens": 3346262.0,
"step": 1835
},
{
"entropy": 5.972390079498291,
"epoch": 1.5244407622203813,
"grad_norm": 1.125,
"learning_rate": 0.0004936406977558847,
"loss": 6.0156,
"mean_token_accuracy": 0.13030771315097808,
"num_tokens": 3354833.0,
"step": 1840
},
{
"entropy": 6.051387977600098,
"epoch": 1.5285832642916322,
"grad_norm": 1.0,
"learning_rate": 0.0004935650401374039,
"loss": 6.0906,
"mean_token_accuracy": 0.14119647368788718,
"num_tokens": 3364314.0,
"step": 1845
},
{
"entropy": 6.046140813827515,
"epoch": 1.532725766362883,
"grad_norm": 1.1328125,
"learning_rate": 0.0004934889416493524,
"loss": 5.9157,
"mean_token_accuracy": 0.13524662256240844,
"num_tokens": 3373648.0,
"step": 1850
},
{
"entropy": 5.863892507553101,
"epoch": 1.5368682684341342,
"grad_norm": 1.1796875,
"learning_rate": 0.0004934124024452295,
"loss": 5.7869,
"mean_token_accuracy": 0.1471114367246628,
"num_tokens": 3382919.0,
"step": 1855
},
{
"entropy": 5.888700532913208,
"epoch": 1.5410107705053853,
"grad_norm": 1.171875,
"learning_rate": 0.0004933354226794232,
"loss": 5.9383,
"mean_token_accuracy": 0.1432707317173481,
"num_tokens": 3391814.0,
"step": 1860
},
{
"entropy": 6.080925798416137,
"epoch": 1.5451532725766364,
"grad_norm": 1.15625,
"learning_rate": 0.0004932580025072098,
"loss": 5.9919,
"mean_token_accuracy": 0.13580586463212968,
"num_tokens": 3400392.0,
"step": 1865
},
{
"entropy": 5.996384525299073,
"epoch": 1.5492957746478875,
"grad_norm": 1.046875,
"learning_rate": 0.0004931801420847545,
"loss": 6.036,
"mean_token_accuracy": 0.13089759424328803,
"num_tokens": 3410430.0,
"step": 1870
},
{
"entropy": 5.932257795333863,
"epoch": 1.5534382767191384,
"grad_norm": 0.9609375,
"learning_rate": 0.0004931018415691102,
"loss": 5.877,
"mean_token_accuracy": 0.13694921880960464,
"num_tokens": 3420654.0,
"step": 1875
},
{
"entropy": 5.951343011856079,
"epoch": 1.5575807787903893,
"grad_norm": 1.09375,
"learning_rate": 0.0004930231011182176,
"loss": 5.9986,
"mean_token_accuracy": 0.13937177062034606,
"num_tokens": 3430514.0,
"step": 1880
},
{
"entropy": 6.071512508392334,
"epoch": 1.5617232808616404,
"grad_norm": 1.15625,
"learning_rate": 0.0004929439208909048,
"loss": 5.9331,
"mean_token_accuracy": 0.1375361390411854,
"num_tokens": 3439551.0,
"step": 1885
},
{
"entropy": 5.902947330474854,
"epoch": 1.5658657829328915,
"grad_norm": 1.1796875,
"learning_rate": 0.0004928643010468869,
"loss": 5.8911,
"mean_token_accuracy": 0.13622887432575226,
"num_tokens": 3448160.0,
"step": 1890
},
{
"entropy": 5.993059539794922,
"epoch": 1.5700082850041426,
"grad_norm": 1.03125,
"learning_rate": 0.0004927842417467659,
"loss": 5.9517,
"mean_token_accuracy": 0.14548188596963882,
"num_tokens": 3457296.0,
"step": 1895
},
{
"entropy": 6.027637386322022,
"epoch": 1.5741507870753937,
"grad_norm": 1.0859375,
"learning_rate": 0.0004927037431520302,
"loss": 5.9073,
"mean_token_accuracy": 0.14102519080042838,
"num_tokens": 3466108.0,
"step": 1900
},
{
"entropy": 5.975209665298462,
"epoch": 1.5782932891466446,
"grad_norm": 1.1328125,
"learning_rate": 0.0004926228054250541,
"loss": 5.9941,
"mean_token_accuracy": 0.14009249955415726,
"num_tokens": 3475336.0,
"step": 1905
},
{
"entropy": 6.073916673660278,
"epoch": 1.5824357912178955,
"grad_norm": 1.125,
"learning_rate": 0.0004925414287290979,
"loss": 5.9668,
"mean_token_accuracy": 0.13723743334412575,
"num_tokens": 3483963.0,
"step": 1910
},
{
"entropy": 5.848618030548096,
"epoch": 1.5865782932891466,
"grad_norm": 1.125,
"learning_rate": 0.0004924596132283075,
"loss": 5.8765,
"mean_token_accuracy": 0.14485997408628465,
"num_tokens": 3492947.0,
"step": 1915
},
{
"entropy": 5.966853284835816,
"epoch": 1.5907207953603977,
"grad_norm": 1.125,
"learning_rate": 0.0004923773590877135,
"loss": 5.9651,
"mean_token_accuracy": 0.13334692791104316,
"num_tokens": 3501877.0,
"step": 1920
},
{
"entropy": 6.042904996871949,
"epoch": 1.5948632974316488,
"grad_norm": 1.1484375,
"learning_rate": 0.0004922946664732314,
"loss": 6.0304,
"mean_token_accuracy": 0.13204361274838447,
"num_tokens": 3510610.0,
"step": 1925
},
{
"entropy": 6.0583656311035154,
"epoch": 1.5990057995029,
"grad_norm": 1.171875,
"learning_rate": 0.0004922115355516615,
"loss": 5.9841,
"mean_token_accuracy": 0.13304307833313941,
"num_tokens": 3519829.0,
"step": 1930
},
{
"entropy": 6.015839862823486,
"epoch": 1.6031483015741508,
"grad_norm": 1.1328125,
"learning_rate": 0.0004921279664906877,
"loss": 5.9661,
"mean_token_accuracy": 0.13071200698614122,
"num_tokens": 3529408.0,
"step": 1935
},
{
"entropy": 5.924501991271972,
"epoch": 1.6072908036454017,
"grad_norm": 1.171875,
"learning_rate": 0.0004920439594588781,
"loss": 5.8955,
"mean_token_accuracy": 0.14675913602113724,
"num_tokens": 3538068.0,
"step": 1940
},
{
"entropy": 6.051935148239136,
"epoch": 1.6114333057166528,
"grad_norm": 1.1015625,
"learning_rate": 0.000491959514625684,
"loss": 6.0177,
"mean_token_accuracy": 0.13845598921179772,
"num_tokens": 3547302.0,
"step": 1945
},
{
"entropy": 5.956210565567017,
"epoch": 1.6155758077879039,
"grad_norm": 1.1796875,
"learning_rate": 0.00049187463216144,
"loss": 5.9024,
"mean_token_accuracy": 0.145584287494421,
"num_tokens": 3556087.0,
"step": 1950
},
{
"entropy": 6.046857452392578,
"epoch": 1.619718309859155,
"grad_norm": 0.984375,
"learning_rate": 0.000491789312237363,
"loss": 5.9857,
"mean_token_accuracy": 0.13579782098531723,
"num_tokens": 3566235.0,
"step": 1955
},
{
"entropy": 5.983897924423218,
"epoch": 1.623860811930406,
"grad_norm": 1.1875,
"learning_rate": 0.0004917035550255528,
"loss": 5.899,
"mean_token_accuracy": 0.14092317298054696,
"num_tokens": 3574762.0,
"step": 1960
},
{
"entropy": 5.904513597488403,
"epoch": 1.628003314001657,
"grad_norm": 1.140625,
"learning_rate": 0.000491617360698991,
"loss": 5.8837,
"mean_token_accuracy": 0.139695380628109,
"num_tokens": 3583359.0,
"step": 1965
},
{
"entropy": 5.920687866210938,
"epoch": 1.6321458160729079,
"grad_norm": 1.125,
"learning_rate": 0.0004915307294315412,
"loss": 5.8505,
"mean_token_accuracy": 0.14011457338929176,
"num_tokens": 3591436.0,
"step": 1970
},
{
"entropy": 5.926318550109864,
"epoch": 1.636288318144159,
"grad_norm": 1.03125,
"learning_rate": 0.000491443661397948,
"loss": 5.9355,
"mean_token_accuracy": 0.13468224480748175,
"num_tokens": 3601054.0,
"step": 1975
},
{
"entropy": 5.974443817138672,
"epoch": 1.64043082021541,
"grad_norm": 1.0390625,
"learning_rate": 0.0004913561567738368,
"loss": 5.9014,
"mean_token_accuracy": 0.1383673779666424,
"num_tokens": 3610641.0,
"step": 1980
},
{
"entropy": 5.912512636184692,
"epoch": 1.6445733222866612,
"grad_norm": 1.171875,
"learning_rate": 0.0004912682157357146,
"loss": 5.8447,
"mean_token_accuracy": 0.13975749984383584,
"num_tokens": 3619351.0,
"step": 1985
},
{
"entropy": 5.853101539611816,
"epoch": 1.6487158243579123,
"grad_norm": 1.125,
"learning_rate": 0.0004911798384609675,
"loss": 5.8306,
"mean_token_accuracy": 0.1472959376871586,
"num_tokens": 3627341.0,
"step": 1990
},
{
"entropy": 5.942045879364014,
"epoch": 1.6528583264291632,
"grad_norm": 1.2109375,
"learning_rate": 0.0004910910251278626,
"loss": 5.8918,
"mean_token_accuracy": 0.14801610112190247,
"num_tokens": 3637132.0,
"step": 1995
},
{
"entropy": 5.919465351104736,
"epoch": 1.6570008285004143,
"grad_norm": 1.09375,
"learning_rate": 0.0004910017759155458,
"loss": 5.9479,
"mean_token_accuracy": 0.1400071859359741,
"num_tokens": 3646196.0,
"step": 2000
},
{
"epoch": 1.6570008285004143,
"eval_entropy": 5.874394122979789,
"eval_loss": 6.1148576736450195,
"eval_mean_token_accuracy": 0.1363023624046525,
"eval_num_tokens": 3646196.0,
"eval_runtime": 2.7574,
"eval_samples_per_second": 1358.871,
"eval_steps_per_second": 170.085,
"step": 2000
}
],
"logging_steps": 5,
"max_steps": 12060,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 5420923210368000.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink