fpadovani/ita-latn-10mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
ita-latn-10mb-after-ppt-shu…/checkpoint-6000/trainer_state.json
ModelHub XC 0e207e3c54 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/ita-latn-10mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
Source: Original Platform
2026-09-23 23:44:24 +08:00

12057 lines
330 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.5954743946010321,
"eval_steps": 3000,
"global_step": 6000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 4.898969125747681,
"epoch": 0.0004962286621675268,
"grad_norm": 13.5625,
"learning_rate": 2e-06,
"loss": 14.3924,
"mean_token_accuracy": 0.0,
"num_tokens": 11448.0,
"step": 5
},
{
"entropy": 4.908018112182617,
"epoch": 0.0009924573243350536,
"grad_norm": 13.875,
"learning_rate": 4.5e-06,
"loss": 14.4058,
"mean_token_accuracy": 0.0,
"num_tokens": 22646.0,
"step": 10
},
{
"entropy": 4.919006061553955,
"epoch": 0.0014886859865025804,
"grad_norm": 13.875,
"learning_rate": 7e-06,
"loss": 14.3319,
"mean_token_accuracy": 6.66666659526527e-05,
"num_tokens": 34011.0,
"step": 15
},
{
"entropy": 4.953374242782592,
"epoch": 0.001984914648670107,
"grad_norm": 17.0,
"learning_rate": 9.5e-06,
"loss": 14.1459,
"mean_token_accuracy": 0.0,
"num_tokens": 44566.0,
"step": 20
},
{
"entropy": 5.0529521942138675,
"epoch": 0.002481143310837634,
"grad_norm": 20.375,
"learning_rate": 1.2e-05,
"loss": 13.7995,
"mean_token_accuracy": 8.56164377182722e-05,
"num_tokens": 56274.0,
"step": 25
},
{
"entropy": 5.29259762763977,
"epoch": 0.0029773719730051607,
"grad_norm": 25.75,
"learning_rate": 1.4500000000000002e-05,
"loss": 13.2456,
"mean_token_accuracy": 0.0,
"num_tokens": 66584.0,
"step": 30
},
{
"entropy": 6.583654308319092,
"epoch": 0.0034736006351726877,
"grad_norm": 25.625,
"learning_rate": 1.7000000000000003e-05,
"loss": 12.143,
"mean_token_accuracy": 0.0,
"num_tokens": 76948.0,
"step": 35
},
{
"entropy": 9.861779689788818,
"epoch": 0.003969829297340214,
"grad_norm": 3.375,
"learning_rate": 1.95e-05,
"loss": 10.9562,
"mean_token_accuracy": 0.00021723387762904168,
"num_tokens": 87253.0,
"step": 40
},
{
"entropy": 10.708572006225586,
"epoch": 0.004466057959507741,
"grad_norm": 3.4375,
"learning_rate": 2.2e-05,
"loss": 10.7237,
"mean_token_accuracy": 0.010100901639088989,
"num_tokens": 98273.0,
"step": 45
},
{
"entropy": 10.741462326049804,
"epoch": 0.004962286621675268,
"grad_norm": 3.015625,
"learning_rate": 2.4500000000000003e-05,
"loss": 10.5536,
"mean_token_accuracy": 0.017115887999534608,
"num_tokens": 107788.0,
"step": 50
},
{
"entropy": 10.71127405166626,
"epoch": 0.0054585152838427945,
"grad_norm": 2.125,
"learning_rate": 2.7e-05,
"loss": 10.3864,
"mean_token_accuracy": 0.015602491609752179,
"num_tokens": 118632.0,
"step": 55
},
{
"entropy": 10.660900974273682,
"epoch": 0.0059547439460103215,
"grad_norm": 2.3125,
"learning_rate": 2.95e-05,
"loss": 10.1743,
"mean_token_accuracy": 0.016560734901577234,
"num_tokens": 128649.0,
"step": 60
},
{
"entropy": 10.705959510803222,
"epoch": 0.0064509726081778485,
"grad_norm": 1.703125,
"learning_rate": 3.2e-05,
"loss": 10.0457,
"mean_token_accuracy": 0.01719390321522951,
"num_tokens": 138958.0,
"step": 65
},
{
"entropy": 10.725889587402344,
"epoch": 0.0069472012703453755,
"grad_norm": 1.7734375,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.9768,
"mean_token_accuracy": 0.04213028624653816,
"num_tokens": 151445.0,
"step": 70
},
{
"entropy": 10.721480083465575,
"epoch": 0.007443429932512902,
"grad_norm": 1.6328125,
"learning_rate": 3.7e-05,
"loss": 9.8683,
"mean_token_accuracy": 0.04831866100430489,
"num_tokens": 162239.0,
"step": 75
},
{
"entropy": 10.70771484375,
"epoch": 0.007939658594680429,
"grad_norm": 1.7265625,
"learning_rate": 3.95e-05,
"loss": 9.8133,
"mean_token_accuracy": 0.042871393263339996,
"num_tokens": 173261.0,
"step": 80
},
{
"entropy": 10.706668090820312,
"epoch": 0.008435887256847956,
"grad_norm": 1.71875,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.6933,
"mean_token_accuracy": 0.042860785871744154,
"num_tokens": 183857.0,
"step": 85
},
{
"entropy": 10.696372127532959,
"epoch": 0.008932115919015483,
"grad_norm": 1.6015625,
"learning_rate": 4.45e-05,
"loss": 9.6957,
"mean_token_accuracy": 0.04004955124109984,
"num_tokens": 196635.0,
"step": 90
},
{
"entropy": 10.685241222381592,
"epoch": 0.00942834458118301,
"grad_norm": 1.6796875,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.5188,
"mean_token_accuracy": 0.04844313133507967,
"num_tokens": 208700.0,
"step": 95
},
{
"entropy": 10.666670799255371,
"epoch": 0.009924573243350537,
"grad_norm": 1.7421875,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.3761,
"mean_token_accuracy": 0.046723760291934015,
"num_tokens": 218915.0,
"step": 100
},
{
"entropy": 10.638776302337646,
"epoch": 0.010420801905518062,
"grad_norm": 1.640625,
"learning_rate": 5.2e-05,
"loss": 9.2697,
"mean_token_accuracy": 0.052684008702635766,
"num_tokens": 230211.0,
"step": 105
},
{
"entropy": 10.60411787033081,
"epoch": 0.010917030567685589,
"grad_norm": 1.6875,
"learning_rate": 5.45e-05,
"loss": 9.1633,
"mean_token_accuracy": 0.05152133964002133,
"num_tokens": 241174.0,
"step": 110
},
{
"entropy": 10.56174831390381,
"epoch": 0.011413259229853116,
"grad_norm": 1.546875,
"learning_rate": 5.7e-05,
"loss": 9.0895,
"mean_token_accuracy": 0.047252026945352556,
"num_tokens": 251708.0,
"step": 115
},
{
"entropy": 10.521473789215088,
"epoch": 0.011909487892020643,
"grad_norm": 1.4140625,
"learning_rate": 5.9499999999999996e-05,
"loss": 9.0937,
"mean_token_accuracy": 0.045639040693640706,
"num_tokens": 263287.0,
"step": 120
},
{
"entropy": 10.479307842254638,
"epoch": 0.01240571655418817,
"grad_norm": 1.3515625,
"learning_rate": 6.2e-05,
"loss": 8.924,
"mean_token_accuracy": 0.043540829047560695,
"num_tokens": 275179.0,
"step": 125
},
{
"entropy": 10.418986606597901,
"epoch": 0.012901945216355697,
"grad_norm": 1.578125,
"learning_rate": 6.450000000000001e-05,
"loss": 8.7387,
"mean_token_accuracy": 0.04855534341186285,
"num_tokens": 287223.0,
"step": 130
},
{
"entropy": 10.346403312683105,
"epoch": 0.013398173878523224,
"grad_norm": 1.3984375,
"learning_rate": 6.7e-05,
"loss": 8.6159,
"mean_token_accuracy": 0.04886538349092007,
"num_tokens": 297868.0,
"step": 135
},
{
"entropy": 10.254747295379639,
"epoch": 0.013894402540690751,
"grad_norm": 1.328125,
"learning_rate": 6.950000000000001e-05,
"loss": 8.5351,
"mean_token_accuracy": 0.050269151106476784,
"num_tokens": 309702.0,
"step": 140
},
{
"entropy": 10.153191566467285,
"epoch": 0.014390631202858276,
"grad_norm": 1.2109375,
"learning_rate": 7.2e-05,
"loss": 8.5012,
"mean_token_accuracy": 0.045820210874080655,
"num_tokens": 321613.0,
"step": 145
},
{
"entropy": 10.02096643447876,
"epoch": 0.014886859865025803,
"grad_norm": 1.171875,
"learning_rate": 7.45e-05,
"loss": 8.3155,
"mean_token_accuracy": 0.050194774940609935,
"num_tokens": 332668.0,
"step": 150
},
{
"entropy": 9.884018039703369,
"epoch": 0.01538308852719333,
"grad_norm": 1.0703125,
"learning_rate": 7.7e-05,
"loss": 8.2768,
"mean_token_accuracy": 0.050016530230641364,
"num_tokens": 342579.0,
"step": 155
},
{
"entropy": 9.711176681518555,
"epoch": 0.015879317189360857,
"grad_norm": 1.1171875,
"learning_rate": 7.950000000000001e-05,
"loss": 8.126,
"mean_token_accuracy": 0.04717333130538463,
"num_tokens": 352271.0,
"step": 160
},
{
"entropy": 9.53788137435913,
"epoch": 0.016375545851528384,
"grad_norm": 1.0078125,
"learning_rate": 8.2e-05,
"loss": 8.0862,
"mean_token_accuracy": 0.05482652448117733,
"num_tokens": 364033.0,
"step": 165
},
{
"entropy": 9.262330722808837,
"epoch": 0.01687177451369591,
"grad_norm": 0.84375,
"learning_rate": 8.450000000000001e-05,
"loss": 7.9892,
"mean_token_accuracy": 0.04694691449403763,
"num_tokens": 373788.0,
"step": 170
},
{
"entropy": 9.129754638671875,
"epoch": 0.01736800317586344,
"grad_norm": 0.8125,
"learning_rate": 8.7e-05,
"loss": 7.9617,
"mean_token_accuracy": 0.044896982982754706,
"num_tokens": 384883.0,
"step": 175
},
{
"entropy": 8.975637817382813,
"epoch": 0.017864231838030965,
"grad_norm": 0.93359375,
"learning_rate": 8.95e-05,
"loss": 7.8555,
"mean_token_accuracy": 0.04881042614579201,
"num_tokens": 396044.0,
"step": 180
},
{
"entropy": 8.740224266052246,
"epoch": 0.018360460500198492,
"grad_norm": 0.76953125,
"learning_rate": 9.2e-05,
"loss": 7.9152,
"mean_token_accuracy": 0.047916683554649356,
"num_tokens": 406995.0,
"step": 185
},
{
"entropy": 8.553805732727051,
"epoch": 0.01885668916236602,
"grad_norm": 0.78515625,
"learning_rate": 9.45e-05,
"loss": 7.7613,
"mean_token_accuracy": 0.05165142118930817,
"num_tokens": 417238.0,
"step": 190
},
{
"entropy": 8.464924621582032,
"epoch": 0.019352917824533546,
"grad_norm": 0.734375,
"learning_rate": 9.7e-05,
"loss": 7.8352,
"mean_token_accuracy": 0.044884103909134866,
"num_tokens": 427263.0,
"step": 195
},
{
"entropy": 8.39473524093628,
"epoch": 0.019849146486701073,
"grad_norm": 0.83984375,
"learning_rate": 9.95e-05,
"loss": 7.7506,
"mean_token_accuracy": 0.04875692799687385,
"num_tokens": 438357.0,
"step": 200
},
{
"entropy": 8.318683052062989,
"epoch": 0.0203453751488686,
"grad_norm": 0.67578125,
"learning_rate": 0.000102,
"loss": 7.7541,
"mean_token_accuracy": 0.04936891198158264,
"num_tokens": 448738.0,
"step": 205
},
{
"entropy": 8.268324375152588,
"epoch": 0.020841603811036124,
"grad_norm": 0.71875,
"learning_rate": 0.00010449999999999999,
"loss": 7.76,
"mean_token_accuracy": 0.056964365392923356,
"num_tokens": 457873.0,
"step": 210
},
{
"entropy": 8.257455635070801,
"epoch": 0.02133783247320365,
"grad_norm": 0.71875,
"learning_rate": 0.000107,
"loss": 7.7046,
"mean_token_accuracy": 0.05989706926047802,
"num_tokens": 468640.0,
"step": 215
},
{
"entropy": 8.317792892456055,
"epoch": 0.021834061135371178,
"grad_norm": 1.2421875,
"learning_rate": 0.0001095,
"loss": 7.7039,
"mean_token_accuracy": 0.06188451088964939,
"num_tokens": 479922.0,
"step": 220
},
{
"entropy": 8.245875453948974,
"epoch": 0.022330289797538705,
"grad_norm": 0.75390625,
"learning_rate": 0.000112,
"loss": 7.663,
"mean_token_accuracy": 0.053246379271149634,
"num_tokens": 491056.0,
"step": 225
},
{
"entropy": 8.234527969360352,
"epoch": 0.022826518459706232,
"grad_norm": 0.875,
"learning_rate": 0.0001145,
"loss": 7.7427,
"mean_token_accuracy": 0.05878411866724491,
"num_tokens": 504365.0,
"step": 230
},
{
"entropy": 8.225464916229248,
"epoch": 0.02332274712187376,
"grad_norm": 1.0,
"learning_rate": 0.00011700000000000001,
"loss": 7.638,
"mean_token_accuracy": 0.057496064156293866,
"num_tokens": 514347.0,
"step": 235
},
{
"entropy": 8.184865188598632,
"epoch": 0.023818975784041286,
"grad_norm": 1.15625,
"learning_rate": 0.00011949999999999999,
"loss": 7.6521,
"mean_token_accuracy": 0.062484626099467276,
"num_tokens": 525912.0,
"step": 240
},
{
"entropy": 8.202389526367188,
"epoch": 0.024315204446208813,
"grad_norm": 1.0,
"learning_rate": 0.000122,
"loss": 7.5455,
"mean_token_accuracy": 0.06251657865941525,
"num_tokens": 535566.0,
"step": 245
},
{
"entropy": 8.209773063659668,
"epoch": 0.02481143310837634,
"grad_norm": 0.9765625,
"learning_rate": 0.0001245,
"loss": 7.593,
"mean_token_accuracy": 0.0647435687482357,
"num_tokens": 546474.0,
"step": 250
},
{
"entropy": 8.14191026687622,
"epoch": 0.025307661770543867,
"grad_norm": 0.84375,
"learning_rate": 0.000127,
"loss": 7.6509,
"mean_token_accuracy": 0.061424101889133456,
"num_tokens": 558132.0,
"step": 255
},
{
"entropy": 8.203132820129394,
"epoch": 0.025803890432711394,
"grad_norm": 1.1640625,
"learning_rate": 0.0001295,
"loss": 7.6428,
"mean_token_accuracy": 0.06428770869970321,
"num_tokens": 568309.0,
"step": 260
},
{
"entropy": 8.081572723388671,
"epoch": 0.02630011909487892,
"grad_norm": 1.3046875,
"learning_rate": 0.000132,
"loss": 7.5249,
"mean_token_accuracy": 0.06533985063433648,
"num_tokens": 579701.0,
"step": 265
},
{
"entropy": 8.113927555084228,
"epoch": 0.026796347757046448,
"grad_norm": 0.9921875,
"learning_rate": 0.00013450000000000002,
"loss": 7.4971,
"mean_token_accuracy": 0.06705491952598094,
"num_tokens": 590832.0,
"step": 270
},
{
"entropy": 8.06861686706543,
"epoch": 0.027292576419213975,
"grad_norm": 1.0078125,
"learning_rate": 0.00013700000000000002,
"loss": 7.4523,
"mean_token_accuracy": 0.06724155806005001,
"num_tokens": 601094.0,
"step": 275
},
{
"entropy": 8.04738645553589,
"epoch": 0.027788805081381502,
"grad_norm": 0.98828125,
"learning_rate": 0.0001395,
"loss": 7.4503,
"mean_token_accuracy": 0.0721239697188139,
"num_tokens": 612027.0,
"step": 280
},
{
"entropy": 7.953780841827393,
"epoch": 0.02828503374354903,
"grad_norm": 0.84375,
"learning_rate": 0.00014199999999999998,
"loss": 7.4716,
"mean_token_accuracy": 0.0679593924432993,
"num_tokens": 623698.0,
"step": 285
},
{
"entropy": 7.966112470626831,
"epoch": 0.028781262405716553,
"grad_norm": 0.91015625,
"learning_rate": 0.0001445,
"loss": 7.4753,
"mean_token_accuracy": 0.06937897726893424,
"num_tokens": 635631.0,
"step": 290
},
{
"entropy": 7.981750106811523,
"epoch": 0.02927749106788408,
"grad_norm": 0.828125,
"learning_rate": 0.000147,
"loss": 7.4567,
"mean_token_accuracy": 0.06901448890566826,
"num_tokens": 647663.0,
"step": 295
},
{
"entropy": 7.9344415187835695,
"epoch": 0.029773719730051607,
"grad_norm": 0.96875,
"learning_rate": 0.0001495,
"loss": 7.4202,
"mean_token_accuracy": 0.06708400063216687,
"num_tokens": 659352.0,
"step": 300
},
{
"entropy": 7.962864351272583,
"epoch": 0.030269948392219134,
"grad_norm": 0.93359375,
"learning_rate": 0.000152,
"loss": 7.3883,
"mean_token_accuracy": 0.0732420600950718,
"num_tokens": 669379.0,
"step": 305
},
{
"entropy": 7.942562055587769,
"epoch": 0.03076617705438666,
"grad_norm": 1.0703125,
"learning_rate": 0.00015450000000000001,
"loss": 7.3844,
"mean_token_accuracy": 0.07024326287209988,
"num_tokens": 680144.0,
"step": 310
},
{
"entropy": 7.996751976013184,
"epoch": 0.03126240571655419,
"grad_norm": 1.15625,
"learning_rate": 0.000157,
"loss": 7.3426,
"mean_token_accuracy": 0.06623242497444153,
"num_tokens": 691165.0,
"step": 315
},
{
"entropy": 7.823547601699829,
"epoch": 0.031758634378721715,
"grad_norm": 1.1640625,
"learning_rate": 0.0001595,
"loss": 7.3996,
"mean_token_accuracy": 0.06626546904444694,
"num_tokens": 703767.0,
"step": 320
},
{
"entropy": 7.798118686676025,
"epoch": 0.032254863040889245,
"grad_norm": 1.03125,
"learning_rate": 0.000162,
"loss": 7.389,
"mean_token_accuracy": 0.07417060770094394,
"num_tokens": 714298.0,
"step": 325
},
{
"entropy": 7.857796287536621,
"epoch": 0.03275109170305677,
"grad_norm": 1.0234375,
"learning_rate": 0.00016450000000000001,
"loss": 7.3782,
"mean_token_accuracy": 0.07758076079189777,
"num_tokens": 724135.0,
"step": 330
},
{
"entropy": 7.84428334236145,
"epoch": 0.03324732036522429,
"grad_norm": 1.1015625,
"learning_rate": 0.00016700000000000002,
"loss": 7.4209,
"mean_token_accuracy": 0.06709365807473659,
"num_tokens": 735050.0,
"step": 335
},
{
"entropy": 7.832436752319336,
"epoch": 0.03374354902739182,
"grad_norm": 0.7265625,
"learning_rate": 0.00016950000000000003,
"loss": 7.3356,
"mean_token_accuracy": 0.07104014568030834,
"num_tokens": 746073.0,
"step": 340
},
{
"entropy": 7.795631265640258,
"epoch": 0.034239777689559346,
"grad_norm": 0.828125,
"learning_rate": 0.00017199999999999998,
"loss": 7.3294,
"mean_token_accuracy": 0.06888764202594758,
"num_tokens": 758998.0,
"step": 345
},
{
"entropy": 7.798632431030273,
"epoch": 0.03473600635172688,
"grad_norm": 1.0703125,
"learning_rate": 0.00017449999999999999,
"loss": 7.2774,
"mean_token_accuracy": 0.06999869644641876,
"num_tokens": 771608.0,
"step": 350
},
{
"entropy": 7.839082527160644,
"epoch": 0.0352322350138944,
"grad_norm": 0.99609375,
"learning_rate": 0.000177,
"loss": 7.3336,
"mean_token_accuracy": 0.07213714122772216,
"num_tokens": 782394.0,
"step": 355
},
{
"entropy": 7.719696950912476,
"epoch": 0.03572846367606193,
"grad_norm": 1.1796875,
"learning_rate": 0.0001795,
"loss": 7.2421,
"mean_token_accuracy": 0.07258376255631446,
"num_tokens": 792389.0,
"step": 360
},
{
"entropy": 7.709195804595947,
"epoch": 0.036224692338229454,
"grad_norm": 1.2734375,
"learning_rate": 0.000182,
"loss": 7.2874,
"mean_token_accuracy": 0.07547833137214184,
"num_tokens": 803219.0,
"step": 365
},
{
"entropy": 7.799539947509766,
"epoch": 0.036720921000396985,
"grad_norm": 1.140625,
"learning_rate": 0.0001845,
"loss": 7.3472,
"mean_token_accuracy": 0.07126162201166153,
"num_tokens": 814249.0,
"step": 370
},
{
"entropy": 7.732659769058228,
"epoch": 0.03721714966256451,
"grad_norm": 0.87109375,
"learning_rate": 0.000187,
"loss": 7.2464,
"mean_token_accuracy": 0.07738562412559986,
"num_tokens": 825138.0,
"step": 375
},
{
"entropy": 7.711156368255615,
"epoch": 0.03771337832473204,
"grad_norm": 1.0625,
"learning_rate": 0.0001895,
"loss": 7.2878,
"mean_token_accuracy": 0.07714832685887814,
"num_tokens": 836024.0,
"step": 380
},
{
"entropy": 7.6839886665344235,
"epoch": 0.03820960698689956,
"grad_norm": 0.890625,
"learning_rate": 0.000192,
"loss": 7.28,
"mean_token_accuracy": 0.0712902493774891,
"num_tokens": 847765.0,
"step": 385
},
{
"entropy": 7.674788284301758,
"epoch": 0.03870583564906709,
"grad_norm": 1.0390625,
"learning_rate": 0.0001945,
"loss": 7.274,
"mean_token_accuracy": 0.07328340820968152,
"num_tokens": 859980.0,
"step": 390
},
{
"entropy": 7.707912158966065,
"epoch": 0.039202064311234616,
"grad_norm": 1.2890625,
"learning_rate": 0.00019700000000000002,
"loss": 7.2697,
"mean_token_accuracy": 0.07509268596768379,
"num_tokens": 871514.0,
"step": 395
},
{
"entropy": 7.807592153549194,
"epoch": 0.03969829297340215,
"grad_norm": 1.2421875,
"learning_rate": 0.00019950000000000002,
"loss": 7.2594,
"mean_token_accuracy": 0.06952392719686032,
"num_tokens": 881984.0,
"step": 400
},
{
"entropy": 7.621029710769653,
"epoch": 0.04019452163556967,
"grad_norm": 1.2265625,
"learning_rate": 0.000202,
"loss": 7.2334,
"mean_token_accuracy": 0.07512535229325294,
"num_tokens": 894198.0,
"step": 405
},
{
"entropy": 7.6571179866790775,
"epoch": 0.0406907502977372,
"grad_norm": 0.91796875,
"learning_rate": 0.00020449999999999998,
"loss": 7.2387,
"mean_token_accuracy": 0.07779023945331573,
"num_tokens": 905394.0,
"step": 410
},
{
"entropy": 7.729351425170899,
"epoch": 0.041186978959904724,
"grad_norm": 0.9609375,
"learning_rate": 0.000207,
"loss": 7.2892,
"mean_token_accuracy": 0.0726883053779602,
"num_tokens": 917010.0,
"step": 415
},
{
"entropy": 7.617552757263184,
"epoch": 0.04168320762207225,
"grad_norm": 1.0078125,
"learning_rate": 0.0002095,
"loss": 7.1201,
"mean_token_accuracy": 0.07954838424921036,
"num_tokens": 926761.0,
"step": 420
},
{
"entropy": 7.515660905838013,
"epoch": 0.04217943628423978,
"grad_norm": 0.85546875,
"learning_rate": 0.000212,
"loss": 7.1273,
"mean_token_accuracy": 0.07839518934488296,
"num_tokens": 938385.0,
"step": 425
},
{
"entropy": 7.5848071575164795,
"epoch": 0.0426756649464073,
"grad_norm": 0.90625,
"learning_rate": 0.0002145,
"loss": 7.1617,
"mean_token_accuracy": 0.07512866109609603,
"num_tokens": 949709.0,
"step": 430
},
{
"entropy": 7.638444280624389,
"epoch": 0.04317189360857483,
"grad_norm": 1.0,
"learning_rate": 0.00021700000000000002,
"loss": 7.2004,
"mean_token_accuracy": 0.07354060970246792,
"num_tokens": 960837.0,
"step": 435
},
{
"entropy": 7.593792295455932,
"epoch": 0.043668122270742356,
"grad_norm": 0.99609375,
"learning_rate": 0.0002195,
"loss": 7.0385,
"mean_token_accuracy": 0.08071191161870957,
"num_tokens": 971036.0,
"step": 440
},
{
"entropy": 7.5954118251800535,
"epoch": 0.044164350932909886,
"grad_norm": 1.1015625,
"learning_rate": 0.000222,
"loss": 7.1804,
"mean_token_accuracy": 0.0766368854790926,
"num_tokens": 983057.0,
"step": 445
},
{
"entropy": 7.580991935729981,
"epoch": 0.04466057959507741,
"grad_norm": 0.94140625,
"learning_rate": 0.0002245,
"loss": 7.1365,
"mean_token_accuracy": 0.08031548969447613,
"num_tokens": 994303.0,
"step": 450
},
{
"entropy": 7.499250650405884,
"epoch": 0.04515680825724494,
"grad_norm": 1.0390625,
"learning_rate": 0.00022700000000000002,
"loss": 7.135,
"mean_token_accuracy": 0.07660448849201203,
"num_tokens": 1005033.0,
"step": 455
},
{
"entropy": 7.613832330703735,
"epoch": 0.045653036919412464,
"grad_norm": 0.91796875,
"learning_rate": 0.00022950000000000002,
"loss": 7.193,
"mean_token_accuracy": 0.08052344620227814,
"num_tokens": 1015940.0,
"step": 460
},
{
"entropy": 7.549146223068237,
"epoch": 0.046149265581579994,
"grad_norm": 1.109375,
"learning_rate": 0.00023200000000000003,
"loss": 7.1365,
"mean_token_accuracy": 0.08448440171778202,
"num_tokens": 1027582.0,
"step": 465
},
{
"entropy": 7.550874137878418,
"epoch": 0.04664549424374752,
"grad_norm": 1.234375,
"learning_rate": 0.00023449999999999998,
"loss": 7.1147,
"mean_token_accuracy": 0.07626789584755897,
"num_tokens": 1040046.0,
"step": 470
},
{
"entropy": 7.5545329570770265,
"epoch": 0.04714172290591505,
"grad_norm": 1.40625,
"learning_rate": 0.000237,
"loss": 7.1922,
"mean_token_accuracy": 0.0756704431027174,
"num_tokens": 1050636.0,
"step": 475
},
{
"entropy": 7.618306350708008,
"epoch": 0.04763795156808257,
"grad_norm": 1.4296875,
"learning_rate": 0.0002395,
"loss": 7.0724,
"mean_token_accuracy": 0.08306895196437836,
"num_tokens": 1060748.0,
"step": 480
},
{
"entropy": 7.600764751434326,
"epoch": 0.0481341802302501,
"grad_norm": 0.9140625,
"learning_rate": 0.000242,
"loss": 7.1277,
"mean_token_accuracy": 0.07725501246750355,
"num_tokens": 1071940.0,
"step": 485
},
{
"entropy": 7.485190486907959,
"epoch": 0.048630408892417626,
"grad_norm": 1.046875,
"learning_rate": 0.0002445,
"loss": 7.1243,
"mean_token_accuracy": 0.08531232550740242,
"num_tokens": 1082962.0,
"step": 490
},
{
"entropy": 7.577714586257935,
"epoch": 0.04912663755458515,
"grad_norm": 1.078125,
"learning_rate": 0.000247,
"loss": 7.0988,
"mean_token_accuracy": 0.08504232913255691,
"num_tokens": 1093082.0,
"step": 495
},
{
"entropy": 7.54087610244751,
"epoch": 0.04962286621675268,
"grad_norm": 0.97265625,
"learning_rate": 0.0002495,
"loss": 7.161,
"mean_token_accuracy": 0.07727786526083946,
"num_tokens": 1103528.0,
"step": 500
},
{
"entropy": 7.444737100601197,
"epoch": 0.0501190948789202,
"grad_norm": 1.359375,
"learning_rate": 0.000252,
"loss": 7.0602,
"mean_token_accuracy": 0.07846606895327568,
"num_tokens": 1114813.0,
"step": 505
},
{
"entropy": 7.511781406402588,
"epoch": 0.050615323541087734,
"grad_norm": 1.1875,
"learning_rate": 0.0002545,
"loss": 7.0536,
"mean_token_accuracy": 0.07985799759626389,
"num_tokens": 1125345.0,
"step": 510
},
{
"entropy": 7.452861642837524,
"epoch": 0.05111155220325526,
"grad_norm": 0.9765625,
"learning_rate": 0.000257,
"loss": 7.0618,
"mean_token_accuracy": 0.08063343465328217,
"num_tokens": 1134985.0,
"step": 515
},
{
"entropy": 7.497338771820068,
"epoch": 0.05160778086542279,
"grad_norm": 0.91796875,
"learning_rate": 0.0002595,
"loss": 7.0742,
"mean_token_accuracy": 0.08803893476724625,
"num_tokens": 1145796.0,
"step": 520
},
{
"entropy": 7.637927579879761,
"epoch": 0.05210400952759031,
"grad_norm": 1.1328125,
"learning_rate": 0.000262,
"loss": 7.207,
"mean_token_accuracy": 0.07291420958936215,
"num_tokens": 1156936.0,
"step": 525
},
{
"entropy": 7.462926721572876,
"epoch": 0.05260023818975784,
"grad_norm": 1.1171875,
"learning_rate": 0.00026450000000000003,
"loss": 7.057,
"mean_token_accuracy": 0.08704347461462021,
"num_tokens": 1167526.0,
"step": 530
},
{
"entropy": 7.486857652664185,
"epoch": 0.053096466851925365,
"grad_norm": 1.109375,
"learning_rate": 0.00026700000000000004,
"loss": 7.1034,
"mean_token_accuracy": 0.0767381489276886,
"num_tokens": 1177647.0,
"step": 535
},
{
"entropy": 7.492286443710327,
"epoch": 0.053592695514092896,
"grad_norm": 1.2890625,
"learning_rate": 0.00026950000000000005,
"loss": 7.0725,
"mean_token_accuracy": 0.08671703785657883,
"num_tokens": 1188255.0,
"step": 540
},
{
"entropy": 7.378701829910279,
"epoch": 0.05408892417626042,
"grad_norm": 0.796875,
"learning_rate": 0.00027200000000000005,
"loss": 6.9983,
"mean_token_accuracy": 0.09135269373655319,
"num_tokens": 1199433.0,
"step": 545
},
{
"entropy": 7.424235725402832,
"epoch": 0.05458515283842795,
"grad_norm": 1.015625,
"learning_rate": 0.0002745,
"loss": 7.0643,
"mean_token_accuracy": 0.08066366724669934,
"num_tokens": 1210067.0,
"step": 550
},
{
"entropy": 7.368731451034546,
"epoch": 0.055081381500595473,
"grad_norm": 1.171875,
"learning_rate": 0.000277,
"loss": 6.956,
"mean_token_accuracy": 0.0897083193063736,
"num_tokens": 1220639.0,
"step": 555
},
{
"entropy": 7.4522312641143795,
"epoch": 0.055577610162763004,
"grad_norm": 1.2578125,
"learning_rate": 0.0002795,
"loss": 7.005,
"mean_token_accuracy": 0.08417323902249337,
"num_tokens": 1233158.0,
"step": 560
},
{
"entropy": 7.401830005645752,
"epoch": 0.05607383882493053,
"grad_norm": 1.125,
"learning_rate": 0.00028199999999999997,
"loss": 6.9805,
"mean_token_accuracy": 0.08172584101557731,
"num_tokens": 1244552.0,
"step": 565
},
{
"entropy": 7.434595537185669,
"epoch": 0.05657006748709806,
"grad_norm": 1.0625,
"learning_rate": 0.0002845,
"loss": 7.0868,
"mean_token_accuracy": 0.07824886590242386,
"num_tokens": 1255478.0,
"step": 570
},
{
"entropy": 7.333680963516235,
"epoch": 0.05706629614926558,
"grad_norm": 1.1796875,
"learning_rate": 0.000287,
"loss": 6.9104,
"mean_token_accuracy": 0.09092385321855545,
"num_tokens": 1267507.0,
"step": 575
},
{
"entropy": 7.288129615783691,
"epoch": 0.057562524811433105,
"grad_norm": 0.953125,
"learning_rate": 0.0002895,
"loss": 6.9139,
"mean_token_accuracy": 0.08322850465774537,
"num_tokens": 1279377.0,
"step": 580
},
{
"entropy": 7.501294803619385,
"epoch": 0.058058753473600636,
"grad_norm": 1.28125,
"learning_rate": 0.000292,
"loss": 7.0026,
"mean_token_accuracy": 0.09129699394106865,
"num_tokens": 1290671.0,
"step": 585
},
{
"entropy": 7.294687795639038,
"epoch": 0.05855498213576816,
"grad_norm": 0.86328125,
"learning_rate": 0.0002945,
"loss": 6.9489,
"mean_token_accuracy": 0.09131397828459739,
"num_tokens": 1302792.0,
"step": 590
},
{
"entropy": 7.357182216644287,
"epoch": 0.05905121079793569,
"grad_norm": 1.015625,
"learning_rate": 0.000297,
"loss": 6.9027,
"mean_token_accuracy": 0.0812704510986805,
"num_tokens": 1313315.0,
"step": 595
},
{
"entropy": 7.276323318481445,
"epoch": 0.05954743946010321,
"grad_norm": 0.9375,
"learning_rate": 0.0002995,
"loss": 6.9766,
"mean_token_accuracy": 0.0856074646115303,
"num_tokens": 1326358.0,
"step": 600
},
{
"entropy": 7.370176696777344,
"epoch": 0.060043668122270744,
"grad_norm": 0.921875,
"learning_rate": 0.000302,
"loss": 6.859,
"mean_token_accuracy": 0.08850097134709359,
"num_tokens": 1336266.0,
"step": 605
},
{
"entropy": 7.235168743133545,
"epoch": 0.06053989678443827,
"grad_norm": 1.0546875,
"learning_rate": 0.0003045,
"loss": 6.8108,
"mean_token_accuracy": 0.0915584422647953,
"num_tokens": 1347610.0,
"step": 610
},
{
"entropy": 7.234266328811645,
"epoch": 0.0610361254466058,
"grad_norm": 1.078125,
"learning_rate": 0.000307,
"loss": 6.7687,
"mean_token_accuracy": 0.09413886740803719,
"num_tokens": 1358317.0,
"step": 615
},
{
"entropy": 7.267807722091675,
"epoch": 0.06153235410877332,
"grad_norm": 1.078125,
"learning_rate": 0.0003095,
"loss": 6.8764,
"mean_token_accuracy": 0.08720959722995758,
"num_tokens": 1370126.0,
"step": 620
},
{
"entropy": 7.298430395126343,
"epoch": 0.06202858277094085,
"grad_norm": 1.2578125,
"learning_rate": 0.000312,
"loss": 6.837,
"mean_token_accuracy": 0.0917629949748516,
"num_tokens": 1381571.0,
"step": 625
},
{
"entropy": 7.124814748764038,
"epoch": 0.06252481143310838,
"grad_norm": 1.1484375,
"learning_rate": 0.0003145,
"loss": 6.8635,
"mean_token_accuracy": 0.09271070286631584,
"num_tokens": 1392486.0,
"step": 630
},
{
"entropy": 7.334286689758301,
"epoch": 0.0630210400952759,
"grad_norm": 0.984375,
"learning_rate": 0.000317,
"loss": 6.8681,
"mean_token_accuracy": 0.08915356397628785,
"num_tokens": 1403086.0,
"step": 635
},
{
"entropy": 7.2069190502166744,
"epoch": 0.06351726875744343,
"grad_norm": 1.0234375,
"learning_rate": 0.0003195,
"loss": 6.8319,
"mean_token_accuracy": 0.09226809442043304,
"num_tokens": 1414537.0,
"step": 640
},
{
"entropy": 7.251292371749878,
"epoch": 0.06401349741961096,
"grad_norm": 1.3046875,
"learning_rate": 0.000322,
"loss": 6.8459,
"mean_token_accuracy": 0.09251495376229286,
"num_tokens": 1426998.0,
"step": 645
},
{
"entropy": 7.164758586883545,
"epoch": 0.06450972608177849,
"grad_norm": 1.109375,
"learning_rate": 0.00032450000000000003,
"loss": 6.874,
"mean_token_accuracy": 0.09265056326985359,
"num_tokens": 1437667.0,
"step": 650
},
{
"entropy": 7.226160907745362,
"epoch": 0.065005954743946,
"grad_norm": 1.09375,
"learning_rate": 0.00032700000000000003,
"loss": 6.7432,
"mean_token_accuracy": 0.0969434417784214,
"num_tokens": 1448280.0,
"step": 655
},
{
"entropy": 7.15698504447937,
"epoch": 0.06550218340611354,
"grad_norm": 1.1875,
"learning_rate": 0.00032950000000000004,
"loss": 6.8019,
"mean_token_accuracy": 0.0953099973499775,
"num_tokens": 1458608.0,
"step": 660
},
{
"entropy": 7.315108156204223,
"epoch": 0.06599841206828107,
"grad_norm": 1.5546875,
"learning_rate": 0.00033200000000000005,
"loss": 6.9802,
"mean_token_accuracy": 0.08709674105048179,
"num_tokens": 1470435.0,
"step": 665
},
{
"entropy": 7.140678405761719,
"epoch": 0.06649464073044858,
"grad_norm": 1.078125,
"learning_rate": 0.00033450000000000005,
"loss": 6.8258,
"mean_token_accuracy": 0.09219390302896499,
"num_tokens": 1481959.0,
"step": 670
},
{
"entropy": 7.185245370864868,
"epoch": 0.06699086939261611,
"grad_norm": 1.09375,
"learning_rate": 0.000337,
"loss": 6.8687,
"mean_token_accuracy": 0.08819933906197548,
"num_tokens": 1493798.0,
"step": 675
},
{
"entropy": 7.220840215682983,
"epoch": 0.06748709805478365,
"grad_norm": 1.0546875,
"learning_rate": 0.0003395,
"loss": 6.6833,
"mean_token_accuracy": 0.0931725412607193,
"num_tokens": 1502969.0,
"step": 680
},
{
"entropy": 7.103082990646362,
"epoch": 0.06798332671695118,
"grad_norm": 1.0546875,
"learning_rate": 0.000342,
"loss": 6.8289,
"mean_token_accuracy": 0.09088226556777954,
"num_tokens": 1513825.0,
"step": 685
},
{
"entropy": 7.194331789016724,
"epoch": 0.06847955537911869,
"grad_norm": 1.0625,
"learning_rate": 0.00034449999999999997,
"loss": 6.7387,
"mean_token_accuracy": 0.09087632670998573,
"num_tokens": 1525328.0,
"step": 690
},
{
"entropy": 7.107731008529663,
"epoch": 0.06897578404128622,
"grad_norm": 0.9609375,
"learning_rate": 0.000347,
"loss": 6.7526,
"mean_token_accuracy": 0.09229976609349251,
"num_tokens": 1537383.0,
"step": 695
},
{
"entropy": 7.252395153045654,
"epoch": 0.06947201270345375,
"grad_norm": 1.0390625,
"learning_rate": 0.0003495,
"loss": 6.8497,
"mean_token_accuracy": 0.09393966495990753,
"num_tokens": 1549671.0,
"step": 700
},
{
"entropy": 7.091624927520752,
"epoch": 0.06996824136562128,
"grad_norm": 0.95703125,
"learning_rate": 0.000352,
"loss": 6.728,
"mean_token_accuracy": 0.08959824368357658,
"num_tokens": 1559974.0,
"step": 705
},
{
"entropy": 7.1363232135772705,
"epoch": 0.0704644700277888,
"grad_norm": 0.8984375,
"learning_rate": 0.0003545,
"loss": 6.8239,
"mean_token_accuracy": 0.08678433373570442,
"num_tokens": 1570890.0,
"step": 710
},
{
"entropy": 7.107376194000244,
"epoch": 0.07096069868995633,
"grad_norm": 1.2265625,
"learning_rate": 0.000357,
"loss": 6.7323,
"mean_token_accuracy": 0.0978621043264866,
"num_tokens": 1581434.0,
"step": 715
},
{
"entropy": 7.089070653915405,
"epoch": 0.07145692735212386,
"grad_norm": 1.234375,
"learning_rate": 0.0003595,
"loss": 6.7863,
"mean_token_accuracy": 0.09178474843502045,
"num_tokens": 1593710.0,
"step": 720
},
{
"entropy": 7.1520946502685545,
"epoch": 0.07195315601429139,
"grad_norm": 1.1640625,
"learning_rate": 0.000362,
"loss": 6.7536,
"mean_token_accuracy": 0.09313018918037415,
"num_tokens": 1604635.0,
"step": 725
},
{
"entropy": 7.222806167602539,
"epoch": 0.07244938467645891,
"grad_norm": 0.9296875,
"learning_rate": 0.0003645,
"loss": 6.8202,
"mean_token_accuracy": 0.09525534659624099,
"num_tokens": 1615258.0,
"step": 730
},
{
"entropy": 7.074941730499267,
"epoch": 0.07294561333862644,
"grad_norm": 0.98828125,
"learning_rate": 0.000367,
"loss": 6.7125,
"mean_token_accuracy": 0.09651728793978691,
"num_tokens": 1625942.0,
"step": 735
},
{
"entropy": 7.0538249015808105,
"epoch": 0.07344184200079397,
"grad_norm": 1.140625,
"learning_rate": 0.0003695,
"loss": 6.7046,
"mean_token_accuracy": 0.1009585440158844,
"num_tokens": 1636270.0,
"step": 740
},
{
"entropy": 7.074357271194458,
"epoch": 0.07393807066296149,
"grad_norm": 1.0703125,
"learning_rate": 0.000372,
"loss": 6.6799,
"mean_token_accuracy": 0.0913244754076004,
"num_tokens": 1647029.0,
"step": 745
},
{
"entropy": 7.095760726928711,
"epoch": 0.07443429932512902,
"grad_norm": 1.3828125,
"learning_rate": 0.0003745,
"loss": 6.8498,
"mean_token_accuracy": 0.09200523346662522,
"num_tokens": 1659483.0,
"step": 750
},
{
"entropy": 6.94871416091919,
"epoch": 0.07493052798729655,
"grad_norm": 0.97265625,
"learning_rate": 0.000377,
"loss": 6.7055,
"mean_token_accuracy": 0.09416541829705238,
"num_tokens": 1669425.0,
"step": 755
},
{
"entropy": 7.080055141448975,
"epoch": 0.07542675664946408,
"grad_norm": 1.0078125,
"learning_rate": 0.0003795,
"loss": 6.6949,
"mean_token_accuracy": 0.09541306495666504,
"num_tokens": 1680426.0,
"step": 760
},
{
"entropy": 7.007839250564575,
"epoch": 0.0759229853116316,
"grad_norm": 0.83203125,
"learning_rate": 0.000382,
"loss": 6.7124,
"mean_token_accuracy": 0.09531945809721946,
"num_tokens": 1692803.0,
"step": 765
},
{
"entropy": 7.0682658672332765,
"epoch": 0.07641921397379912,
"grad_norm": 0.98828125,
"learning_rate": 0.0003845,
"loss": 6.7343,
"mean_token_accuracy": 0.09483553916215896,
"num_tokens": 1703919.0,
"step": 770
},
{
"entropy": 7.002940464019775,
"epoch": 0.07691544263596665,
"grad_norm": 0.96875,
"learning_rate": 0.00038700000000000003,
"loss": 6.7095,
"mean_token_accuracy": 0.10153669938445091,
"num_tokens": 1714487.0,
"step": 775
},
{
"entropy": 7.094982624053955,
"epoch": 0.07741167129813419,
"grad_norm": 1.1875,
"learning_rate": 0.00038950000000000003,
"loss": 6.7585,
"mean_token_accuracy": 0.09340820983052253,
"num_tokens": 1726601.0,
"step": 780
},
{
"entropy": 7.01391978263855,
"epoch": 0.0779078999603017,
"grad_norm": 1.140625,
"learning_rate": 0.00039200000000000004,
"loss": 6.6475,
"mean_token_accuracy": 0.09487889260053635,
"num_tokens": 1738743.0,
"step": 785
},
{
"entropy": 6.975992584228516,
"epoch": 0.07840412862246923,
"grad_norm": 1.15625,
"learning_rate": 0.00039450000000000005,
"loss": 6.7308,
"mean_token_accuracy": 0.0951120376586914,
"num_tokens": 1748286.0,
"step": 790
},
{
"entropy": 6.987929821014404,
"epoch": 0.07890035728463676,
"grad_norm": 1.1015625,
"learning_rate": 0.00039700000000000005,
"loss": 6.7515,
"mean_token_accuracy": 0.0957480825483799,
"num_tokens": 1758917.0,
"step": 795
},
{
"entropy": 6.972993183135986,
"epoch": 0.0793965859468043,
"grad_norm": 0.91015625,
"learning_rate": 0.0003995,
"loss": 6.5889,
"mean_token_accuracy": 0.09930885285139084,
"num_tokens": 1769179.0,
"step": 800
},
{
"entropy": 6.96280460357666,
"epoch": 0.07989281460897181,
"grad_norm": 0.9375,
"learning_rate": 0.000402,
"loss": 6.7138,
"mean_token_accuracy": 0.08964183181524277,
"num_tokens": 1782369.0,
"step": 805
},
{
"entropy": 6.994895315170288,
"epoch": 0.08038904327113934,
"grad_norm": 1.015625,
"learning_rate": 0.0004045,
"loss": 6.7162,
"mean_token_accuracy": 0.10041663944721221,
"num_tokens": 1795992.0,
"step": 810
},
{
"entropy": 6.964673948287964,
"epoch": 0.08088527193330687,
"grad_norm": 0.9296875,
"learning_rate": 0.00040699999999999997,
"loss": 6.6023,
"mean_token_accuracy": 0.10524952039122581,
"num_tokens": 1806938.0,
"step": 815
},
{
"entropy": 6.863758850097656,
"epoch": 0.0813815005954744,
"grad_norm": 0.91796875,
"learning_rate": 0.0004095,
"loss": 6.5447,
"mean_token_accuracy": 0.0992793746292591,
"num_tokens": 1817663.0,
"step": 820
},
{
"entropy": 6.998508739471435,
"epoch": 0.08187772925764192,
"grad_norm": 0.9609375,
"learning_rate": 0.000412,
"loss": 6.6978,
"mean_token_accuracy": 0.09649931266903877,
"num_tokens": 1828254.0,
"step": 825
},
{
"entropy": 6.933160877227783,
"epoch": 0.08237395791980945,
"grad_norm": 1.4296875,
"learning_rate": 0.0004145,
"loss": 6.6293,
"mean_token_accuracy": 0.09699171110987663,
"num_tokens": 1839001.0,
"step": 830
},
{
"entropy": 6.939033889770508,
"epoch": 0.08287018658197698,
"grad_norm": 1.015625,
"learning_rate": 0.000417,
"loss": 6.7288,
"mean_token_accuracy": 0.09662937819957733,
"num_tokens": 1849906.0,
"step": 835
},
{
"entropy": 6.815816164016724,
"epoch": 0.0833664152441445,
"grad_norm": 0.94140625,
"learning_rate": 0.0004195,
"loss": 6.6076,
"mean_token_accuracy": 0.09685141816735268,
"num_tokens": 1860982.0,
"step": 840
},
{
"entropy": 7.031003141403199,
"epoch": 0.08386264390631203,
"grad_norm": 0.94921875,
"learning_rate": 0.000422,
"loss": 6.5616,
"mean_token_accuracy": 0.10065253376960755,
"num_tokens": 1872177.0,
"step": 845
},
{
"entropy": 6.826659727096557,
"epoch": 0.08435887256847956,
"grad_norm": 1.0390625,
"learning_rate": 0.0004245,
"loss": 6.5362,
"mean_token_accuracy": 0.0993971161544323,
"num_tokens": 1882783.0,
"step": 850
},
{
"entropy": 7.006473255157471,
"epoch": 0.08485510123064709,
"grad_norm": 1.0390625,
"learning_rate": 0.000427,
"loss": 6.7184,
"mean_token_accuracy": 0.08898815661668777,
"num_tokens": 1893706.0,
"step": 855
},
{
"entropy": 6.823424482345581,
"epoch": 0.0853513298928146,
"grad_norm": 0.91796875,
"learning_rate": 0.0004295,
"loss": 6.5757,
"mean_token_accuracy": 0.09971656277775764,
"num_tokens": 1903852.0,
"step": 860
},
{
"entropy": 6.9098859310150145,
"epoch": 0.08584755855498213,
"grad_norm": 0.95703125,
"learning_rate": 0.000432,
"loss": 6.6087,
"mean_token_accuracy": 0.09987007081508636,
"num_tokens": 1914157.0,
"step": 865
},
{
"entropy": 6.880821752548218,
"epoch": 0.08634378721714966,
"grad_norm": 0.9453125,
"learning_rate": 0.0004345,
"loss": 6.5541,
"mean_token_accuracy": 0.09982468262314796,
"num_tokens": 1924684.0,
"step": 870
},
{
"entropy": 6.945840787887573,
"epoch": 0.0868400158793172,
"grad_norm": 0.93359375,
"learning_rate": 0.000437,
"loss": 6.7066,
"mean_token_accuracy": 0.0949609711766243,
"num_tokens": 1935657.0,
"step": 875
},
{
"entropy": 6.870462036132812,
"epoch": 0.08733624454148471,
"grad_norm": 0.9921875,
"learning_rate": 0.0004395,
"loss": 6.5897,
"mean_token_accuracy": 0.09813723415136337,
"num_tokens": 1946735.0,
"step": 880
},
{
"entropy": 6.932937145233154,
"epoch": 0.08783247320365224,
"grad_norm": 1.046875,
"learning_rate": 0.000442,
"loss": 6.7502,
"mean_token_accuracy": 0.09312896803021431,
"num_tokens": 1957871.0,
"step": 885
},
{
"entropy": 6.838114881515503,
"epoch": 0.08832870186581977,
"grad_norm": 1.1484375,
"learning_rate": 0.0004445,
"loss": 6.5663,
"mean_token_accuracy": 0.10240828990936279,
"num_tokens": 1969032.0,
"step": 890
},
{
"entropy": 6.92586464881897,
"epoch": 0.0888249305279873,
"grad_norm": 1.1484375,
"learning_rate": 0.000447,
"loss": 6.6063,
"mean_token_accuracy": 0.09725166410207749,
"num_tokens": 1980212.0,
"step": 895
},
{
"entropy": 6.780085325241089,
"epoch": 0.08932115919015482,
"grad_norm": 0.98828125,
"learning_rate": 0.00044950000000000003,
"loss": 6.5267,
"mean_token_accuracy": 0.10072881132364273,
"num_tokens": 1990333.0,
"step": 900
},
{
"entropy": 6.879197597503662,
"epoch": 0.08981738785232235,
"grad_norm": 1.0078125,
"learning_rate": 0.00045200000000000004,
"loss": 6.6025,
"mean_token_accuracy": 0.10396269112825393,
"num_tokens": 2001486.0,
"step": 905
},
{
"entropy": 6.857023143768311,
"epoch": 0.09031361651448988,
"grad_norm": 1.140625,
"learning_rate": 0.00045450000000000004,
"loss": 6.5679,
"mean_token_accuracy": 0.1022559143602848,
"num_tokens": 2012535.0,
"step": 910
},
{
"entropy": 6.766489458084107,
"epoch": 0.0908098451766574,
"grad_norm": 0.984375,
"learning_rate": 0.00045700000000000005,
"loss": 6.5318,
"mean_token_accuracy": 0.10583447441458702,
"num_tokens": 2024116.0,
"step": 915
},
{
"entropy": 6.746376848220825,
"epoch": 0.09130607383882493,
"grad_norm": 0.85546875,
"learning_rate": 0.00045950000000000006,
"loss": 6.5749,
"mean_token_accuracy": 0.09981027767062187,
"num_tokens": 2036341.0,
"step": 920
},
{
"entropy": 6.938368415832519,
"epoch": 0.09180230250099246,
"grad_norm": 0.9375,
"learning_rate": 0.000462,
"loss": 6.5504,
"mean_token_accuracy": 0.10081811845302582,
"num_tokens": 2046768.0,
"step": 925
},
{
"entropy": 6.7602132797241214,
"epoch": 0.09229853116315999,
"grad_norm": 0.98046875,
"learning_rate": 0.0004645,
"loss": 6.5178,
"mean_token_accuracy": 0.09979332089424134,
"num_tokens": 2057315.0,
"step": 930
},
{
"entropy": 6.767438125610352,
"epoch": 0.0927947598253275,
"grad_norm": 1.125,
"learning_rate": 0.000467,
"loss": 6.4968,
"mean_token_accuracy": 0.10354413539171219,
"num_tokens": 2068267.0,
"step": 935
},
{
"entropy": 6.767202854156494,
"epoch": 0.09329098848749504,
"grad_norm": 1.078125,
"learning_rate": 0.0004695,
"loss": 6.5673,
"mean_token_accuracy": 0.10121043622493744,
"num_tokens": 2078271.0,
"step": 940
},
{
"entropy": 6.74484601020813,
"epoch": 0.09378721714966257,
"grad_norm": 1.0390625,
"learning_rate": 0.000472,
"loss": 6.4897,
"mean_token_accuracy": 0.09888025969266892,
"num_tokens": 2088381.0,
"step": 945
},
{
"entropy": 6.789056253433228,
"epoch": 0.0942834458118301,
"grad_norm": 0.96484375,
"learning_rate": 0.0004745,
"loss": 6.451,
"mean_token_accuracy": 0.09835516735911369,
"num_tokens": 2098899.0,
"step": 950
},
{
"entropy": 6.749150371551513,
"epoch": 0.09477967447399761,
"grad_norm": 1.046875,
"learning_rate": 0.000477,
"loss": 6.5972,
"mean_token_accuracy": 0.09927409738302231,
"num_tokens": 2110057.0,
"step": 955
},
{
"entropy": 6.713812303543091,
"epoch": 0.09527590313616514,
"grad_norm": 1.015625,
"learning_rate": 0.0004795,
"loss": 6.5065,
"mean_token_accuracy": 0.09916102886199951,
"num_tokens": 2121758.0,
"step": 960
},
{
"entropy": 6.771513843536377,
"epoch": 0.09577213179833267,
"grad_norm": 0.96484375,
"learning_rate": 0.000482,
"loss": 6.5131,
"mean_token_accuracy": 0.10310358032584191,
"num_tokens": 2131840.0,
"step": 965
},
{
"entropy": 6.744855642318726,
"epoch": 0.0962683604605002,
"grad_norm": 1.2890625,
"learning_rate": 0.0004845,
"loss": 6.5007,
"mean_token_accuracy": 0.1013672724366188,
"num_tokens": 2142084.0,
"step": 970
},
{
"entropy": 6.749075984954834,
"epoch": 0.09676458912266772,
"grad_norm": 1.015625,
"learning_rate": 0.000487,
"loss": 6.5483,
"mean_token_accuracy": 0.10051179081201553,
"num_tokens": 2152785.0,
"step": 975
},
{
"entropy": 6.751393842697143,
"epoch": 0.09726081778483525,
"grad_norm": 1.015625,
"learning_rate": 0.0004895,
"loss": 6.5224,
"mean_token_accuracy": 0.1026117317378521,
"num_tokens": 2163774.0,
"step": 980
},
{
"entropy": 6.7088648796081545,
"epoch": 0.09775704644700278,
"grad_norm": 0.9140625,
"learning_rate": 0.000492,
"loss": 6.4589,
"mean_token_accuracy": 0.10463614240288735,
"num_tokens": 2175378.0,
"step": 985
},
{
"entropy": 6.910195636749267,
"epoch": 0.0982532751091703,
"grad_norm": 0.93359375,
"learning_rate": 0.0004945,
"loss": 6.6604,
"mean_token_accuracy": 0.09883119091391564,
"num_tokens": 2186510.0,
"step": 990
},
{
"entropy": 6.734104061126709,
"epoch": 0.09874950377133783,
"grad_norm": 0.91015625,
"learning_rate": 0.000497,
"loss": 6.4553,
"mean_token_accuracy": 0.10956107974052429,
"num_tokens": 2197482.0,
"step": 995
},
{
"entropy": 6.622636842727661,
"epoch": 0.09924573243350536,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995,
"loss": 6.4604,
"mean_token_accuracy": 0.10386316254734992,
"num_tokens": 2208329.0,
"step": 1000
},
{
"entropy": 6.609873104095459,
"epoch": 0.09974196109567289,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999999982145551,
"loss": 6.362,
"mean_token_accuracy": 0.10516428351402282,
"num_tokens": 2218632.0,
"step": 1005
},
{
"entropy": 6.709473085403443,
"epoch": 0.1002381897578404,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999999909611855,
"loss": 6.4814,
"mean_token_accuracy": 0.09935311302542686,
"num_tokens": 2231335.0,
"step": 1010
},
{
"entropy": 6.6939784526824955,
"epoch": 0.10073441842000794,
"grad_norm": 1.140625,
"learning_rate": 0.0004999999781283008,
"loss": 6.4733,
"mean_token_accuracy": 0.10079516991972923,
"num_tokens": 2241704.0,
"step": 1015
},
{
"entropy": 6.665507650375366,
"epoch": 0.10123064708217547,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999999597159017,
"loss": 6.505,
"mean_token_accuracy": 0.1012033224105835,
"num_tokens": 2252310.0,
"step": 1020
},
{
"entropy": 6.7658100605010985,
"epoch": 0.101726875744343,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999999357239882,
"loss": 6.4323,
"mean_token_accuracy": 0.10344030633568764,
"num_tokens": 2262270.0,
"step": 1025
},
{
"entropy": 6.723705816268921,
"epoch": 0.10222310440651051,
"grad_norm": 0.9375,
"learning_rate": 0.0004999999061525612,
"loss": 6.5677,
"mean_token_accuracy": 0.10005603730678558,
"num_tokens": 2273986.0,
"step": 1030
},
{
"entropy": 6.7837742328643795,
"epoch": 0.10271933306867805,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999998710016214,
"loss": 6.4629,
"mean_token_accuracy": 0.10525803044438362,
"num_tokens": 2283430.0,
"step": 1035
},
{
"entropy": 6.677142333984375,
"epoch": 0.10321556173084558,
"grad_norm": 0.953125,
"learning_rate": 0.0004999998302711696,
"loss": 6.4669,
"mean_token_accuracy": 0.10035635456442833,
"num_tokens": 2294201.0,
"step": 1040
},
{
"entropy": 6.609557294845581,
"epoch": 0.1037117903930131,
"grad_norm": 0.82421875,
"learning_rate": 0.0004999997839612069,
"loss": 6.3612,
"mean_token_accuracy": 0.1090113341808319,
"num_tokens": 2303876.0,
"step": 1045
},
{
"entropy": 6.626906967163086,
"epoch": 0.10420801905518062,
"grad_norm": 0.90625,
"learning_rate": 0.0004999997320717342,
"loss": 6.4015,
"mean_token_accuracy": 0.10890848338603973,
"num_tokens": 2313713.0,
"step": 1050
},
{
"entropy": 6.605928659439087,
"epoch": 0.10470424771734815,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999996746027531,
"loss": 6.4148,
"mean_token_accuracy": 0.10125040039420127,
"num_tokens": 2324862.0,
"step": 1055
},
{
"entropy": 6.717818832397461,
"epoch": 0.10520047637951568,
"grad_norm": 1.03125,
"learning_rate": 0.000499999611554265,
"loss": 6.4953,
"mean_token_accuracy": 0.10330296978354454,
"num_tokens": 2334974.0,
"step": 1060
},
{
"entropy": 6.62673773765564,
"epoch": 0.10569670504168321,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999995429262712,
"loss": 6.5196,
"mean_token_accuracy": 0.10395341366529465,
"num_tokens": 2345945.0,
"step": 1065
},
{
"entropy": 6.722126197814942,
"epoch": 0.10619293370385073,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999994687187737,
"loss": 6.4608,
"mean_token_accuracy": 0.10263997614383698,
"num_tokens": 2356749.0,
"step": 1070
},
{
"entropy": 6.70174617767334,
"epoch": 0.10668916236601826,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999993889317742,
"loss": 6.5289,
"mean_token_accuracy": 0.10041191279888154,
"num_tokens": 2368181.0,
"step": 1075
},
{
"entropy": 6.65612006187439,
"epoch": 0.10718539102818579,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999993035652746,
"loss": 6.432,
"mean_token_accuracy": 0.10550632029771805,
"num_tokens": 2377737.0,
"step": 1080
},
{
"entropy": 6.570956897735596,
"epoch": 0.10768161969035331,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999992126192771,
"loss": 6.3551,
"mean_token_accuracy": 0.10739153623580933,
"num_tokens": 2388944.0,
"step": 1085
},
{
"entropy": 6.695246934890747,
"epoch": 0.10817784835252084,
"grad_norm": 0.984375,
"learning_rate": 0.000499999116093784,
"loss": 6.4357,
"mean_token_accuracy": 0.10068753212690354,
"num_tokens": 2400758.0,
"step": 1090
},
{
"entropy": 6.534470844268799,
"epoch": 0.10867407701468837,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999990139887976,
"loss": 6.4253,
"mean_token_accuracy": 0.10705526322126388,
"num_tokens": 2411916.0,
"step": 1095
},
{
"entropy": 6.811460256576538,
"epoch": 0.1091703056768559,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999989063043206,
"loss": 6.463,
"mean_token_accuracy": 0.10148089230060578,
"num_tokens": 2423310.0,
"step": 1100
},
{
"entropy": 6.5493433475494385,
"epoch": 0.10966653433902342,
"grad_norm": 1.015625,
"learning_rate": 0.0004999987930403555,
"loss": 6.3854,
"mean_token_accuracy": 0.10383407324552536,
"num_tokens": 2432100.0,
"step": 1105
},
{
"entropy": 6.6149815082550045,
"epoch": 0.11016276300119095,
"grad_norm": 0.9375,
"learning_rate": 0.0004999986741969051,
"loss": 6.4639,
"mean_token_accuracy": 0.10388477817177773,
"num_tokens": 2443563.0,
"step": 1110
},
{
"entropy": 6.726413249969482,
"epoch": 0.11065899166335848,
"grad_norm": 1.0,
"learning_rate": 0.0004999985497739724,
"loss": 6.4566,
"mean_token_accuracy": 0.10313622802495956,
"num_tokens": 2455435.0,
"step": 1115
},
{
"entropy": 6.534851551055908,
"epoch": 0.11115522032552601,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999984197715606,
"loss": 6.3981,
"mean_token_accuracy": 0.10903462842106819,
"num_tokens": 2466173.0,
"step": 1120
},
{
"entropy": 6.549497175216675,
"epoch": 0.11165144898769352,
"grad_norm": 0.890625,
"learning_rate": 0.0004999982841896728,
"loss": 6.3521,
"mean_token_accuracy": 0.10385128781199456,
"num_tokens": 2476349.0,
"step": 1125
},
{
"entropy": 6.66382737159729,
"epoch": 0.11214767764986106,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999981430283123,
"loss": 6.4763,
"mean_token_accuracy": 0.10946066528558732,
"num_tokens": 2487864.0,
"step": 1130
},
{
"entropy": 6.5958610534667965,
"epoch": 0.11264390631202859,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999979962874827,
"loss": 6.4164,
"mean_token_accuracy": 0.10289325043559075,
"num_tokens": 2499188.0,
"step": 1135
},
{
"entropy": 6.616852045059204,
"epoch": 0.11314013497419612,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999978439671877,
"loss": 6.3896,
"mean_token_accuracy": 0.10375408679246903,
"num_tokens": 2509813.0,
"step": 1140
},
{
"entropy": 6.66924524307251,
"epoch": 0.11363636363636363,
"grad_norm": 0.8046875,
"learning_rate": 0.0004999976860674309,
"loss": 6.4185,
"mean_token_accuracy": 0.10202216356992722,
"num_tokens": 2520813.0,
"step": 1145
},
{
"entropy": 6.483026504516602,
"epoch": 0.11413259229853116,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999975225882164,
"loss": 6.4173,
"mean_token_accuracy": 0.1033550575375557,
"num_tokens": 2531802.0,
"step": 1150
},
{
"entropy": 6.586180353164673,
"epoch": 0.1146288209606987,
"grad_norm": 1.0078125,
"learning_rate": 0.000499997353529548,
"loss": 6.2121,
"mean_token_accuracy": 0.1153721533715725,
"num_tokens": 2541935.0,
"step": 1155
},
{
"entropy": 6.49102201461792,
"epoch": 0.11512504962286621,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999971788914303,
"loss": 6.3252,
"mean_token_accuracy": 0.10732494816184043,
"num_tokens": 2553224.0,
"step": 1160
},
{
"entropy": 6.557978200912475,
"epoch": 0.11562127828503374,
"grad_norm": 0.7421875,
"learning_rate": 0.0004999969986738672,
"loss": 6.3578,
"mean_token_accuracy": 0.10425093993544579,
"num_tokens": 2565355.0,
"step": 1165
},
{
"entropy": 6.631026411056519,
"epoch": 0.11611750694720127,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999968128768634,
"loss": 6.4405,
"mean_token_accuracy": 0.10526153668761254,
"num_tokens": 2575962.0,
"step": 1170
},
{
"entropy": 6.586783170700073,
"epoch": 0.1166137356093688,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999966215004236,
"loss": 6.4478,
"mean_token_accuracy": 0.11039511635899543,
"num_tokens": 2588136.0,
"step": 1175
},
{
"entropy": 6.563306188583374,
"epoch": 0.11710996427153632,
"grad_norm": 0.890625,
"learning_rate": 0.0004999964245445522,
"loss": 6.3313,
"mean_token_accuracy": 0.10910526439547538,
"num_tokens": 2597785.0,
"step": 1180
},
{
"entropy": 6.518818330764771,
"epoch": 0.11760619293370385,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999962220092545,
"loss": 6.3785,
"mean_token_accuracy": 0.11630177646875381,
"num_tokens": 2609738.0,
"step": 1185
},
{
"entropy": 6.455830097198486,
"epoch": 0.11810242159587138,
"grad_norm": 0.81640625,
"learning_rate": 0.0004999960138945351,
"loss": 6.2671,
"mean_token_accuracy": 0.11771797314286232,
"num_tokens": 2620179.0,
"step": 1190
},
{
"entropy": 6.5756360530853275,
"epoch": 0.11859865025803891,
"grad_norm": 0.9375,
"learning_rate": 0.0004999958002003996,
"loss": 6.3035,
"mean_token_accuracy": 0.11005059778690338,
"num_tokens": 2631550.0,
"step": 1195
},
{
"entropy": 6.5420623302459715,
"epoch": 0.11909487892020643,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999955809268529,
"loss": 6.3474,
"mean_token_accuracy": 0.1102082036435604,
"num_tokens": 2642467.0,
"step": 1200
},
{
"entropy": 6.430347681045532,
"epoch": 0.11959110758237396,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999953560739008,
"loss": 6.3182,
"mean_token_accuracy": 0.1130327932536602,
"num_tokens": 2652132.0,
"step": 1205
},
{
"entropy": 6.609172010421753,
"epoch": 0.12008733624454149,
"grad_norm": 0.80859375,
"learning_rate": 0.0004999951256415485,
"loss": 6.3786,
"mean_token_accuracy": 0.1051353357732296,
"num_tokens": 2663336.0,
"step": 1210
},
{
"entropy": 6.602077627182007,
"epoch": 0.12058356490670902,
"grad_norm": 0.95703125,
"learning_rate": 0.000499994889629802,
"loss": 6.4557,
"mean_token_accuracy": 0.09855944886803628,
"num_tokens": 2674201.0,
"step": 1215
},
{
"entropy": 6.591161394119263,
"epoch": 0.12107979356887653,
"grad_norm": 0.93359375,
"learning_rate": 0.000499994648038667,
"loss": 6.3755,
"mean_token_accuracy": 0.11215761303901672,
"num_tokens": 2685073.0,
"step": 1220
},
{
"entropy": 6.588635969161987,
"epoch": 0.12157602223104406,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999944008681496,
"loss": 6.3189,
"mean_token_accuracy": 0.11205925047397614,
"num_tokens": 2695024.0,
"step": 1225
},
{
"entropy": 6.502772092819214,
"epoch": 0.1220722508932116,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999941481182559,
"loss": 6.4082,
"mean_token_accuracy": 0.10132400244474411,
"num_tokens": 2706232.0,
"step": 1230
},
{
"entropy": 6.634514093399048,
"epoch": 0.12256847955537911,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999938897889921,
"loss": 6.3776,
"mean_token_accuracy": 0.11380477249622345,
"num_tokens": 2715874.0,
"step": 1235
},
{
"entropy": 6.467554330825806,
"epoch": 0.12306470821754664,
"grad_norm": 0.80859375,
"learning_rate": 0.0004999936258803647,
"loss": 6.3171,
"mean_token_accuracy": 0.11320996284484863,
"num_tokens": 2727409.0,
"step": 1240
},
{
"entropy": 6.4485221862792965,
"epoch": 0.12356093687971417,
"grad_norm": 0.96875,
"learning_rate": 0.0004999933563923802,
"loss": 6.2336,
"mean_token_accuracy": 0.11560766994953156,
"num_tokens": 2737259.0,
"step": 1245
},
{
"entropy": 6.5671069622039795,
"epoch": 0.1240571655418817,
"grad_norm": 0.7734375,
"learning_rate": 0.0004999930813250451,
"loss": 6.3636,
"mean_token_accuracy": 0.10681080147624016,
"num_tokens": 2748416.0,
"step": 1250
},
{
"entropy": 6.504324531555175,
"epoch": 0.12455339420404922,
"grad_norm": 0.890625,
"learning_rate": 0.0004999928006783667,
"loss": 6.3299,
"mean_token_accuracy": 0.10958396941423416,
"num_tokens": 2760470.0,
"step": 1255
},
{
"entropy": 6.538706254959107,
"epoch": 0.12504962286621676,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999925144523515,
"loss": 6.3608,
"mean_token_accuracy": 0.10441428348422051,
"num_tokens": 2770222.0,
"step": 1260
},
{
"entropy": 6.533433151245117,
"epoch": 0.12554585152838427,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999922226470066,
"loss": 6.2804,
"mean_token_accuracy": 0.10863575115799903,
"num_tokens": 2780212.0,
"step": 1265
},
{
"entropy": 6.4365581512451175,
"epoch": 0.1260420801905518,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999919252623397,
"loss": 6.2603,
"mean_token_accuracy": 0.11084286272525787,
"num_tokens": 2790239.0,
"step": 1270
},
{
"entropy": 6.580913019180298,
"epoch": 0.12653830885271933,
"grad_norm": 1.0,
"learning_rate": 0.0004999916222983577,
"loss": 6.2971,
"mean_token_accuracy": 0.1081952191889286,
"num_tokens": 2801902.0,
"step": 1275
},
{
"entropy": 6.360956907272339,
"epoch": 0.12703453751488686,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999913137550682,
"loss": 6.2987,
"mean_token_accuracy": 0.10948901474475861,
"num_tokens": 2813268.0,
"step": 1280
},
{
"entropy": 6.5253222465515135,
"epoch": 0.1275307661770544,
"grad_norm": 0.80078125,
"learning_rate": 0.0004999909996324791,
"loss": 6.3583,
"mean_token_accuracy": 0.1083506740629673,
"num_tokens": 2824506.0,
"step": 1285
},
{
"entropy": 6.4849005222320555,
"epoch": 0.12802699483922192,
"grad_norm": 0.84375,
"learning_rate": 0.000499990679930598,
"loss": 6.2729,
"mean_token_accuracy": 0.11370046511292457,
"num_tokens": 2835144.0,
"step": 1290
},
{
"entropy": 6.4290557384490965,
"epoch": 0.12852322350138945,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999903546494327,
"loss": 6.2832,
"mean_token_accuracy": 0.1095927357673645,
"num_tokens": 2845382.0,
"step": 1295
},
{
"entropy": 6.500859642028809,
"epoch": 0.12901945216355698,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999900237889916,
"loss": 6.2942,
"mean_token_accuracy": 0.11329072564840317,
"num_tokens": 2854697.0,
"step": 1300
},
{
"entropy": 6.496008634567261,
"epoch": 0.12951568082572448,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999896873492827,
"loss": 6.313,
"mean_token_accuracy": 0.10269764885306358,
"num_tokens": 2866216.0,
"step": 1305
},
{
"entropy": 6.478988075256348,
"epoch": 0.130011909487892,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999893453303143,
"loss": 6.2662,
"mean_token_accuracy": 0.11140501946210861,
"num_tokens": 2877734.0,
"step": 1310
},
{
"entropy": 6.496612596511841,
"epoch": 0.13050813815005954,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999889977320949,
"loss": 6.3097,
"mean_token_accuracy": 0.11451938673853874,
"num_tokens": 2889445.0,
"step": 1315
},
{
"entropy": 6.386645030975342,
"epoch": 0.13100436681222707,
"grad_norm": 0.796875,
"learning_rate": 0.0004999886445546332,
"loss": 6.2526,
"mean_token_accuracy": 0.117067901045084,
"num_tokens": 2900485.0,
"step": 1320
},
{
"entropy": 6.536042022705078,
"epoch": 0.1315005954743946,
"grad_norm": 0.8828125,
"learning_rate": 0.000499988285797938,
"loss": 6.2462,
"mean_token_accuracy": 0.11725969165563584,
"num_tokens": 2909795.0,
"step": 1325
},
{
"entropy": 6.310559272766113,
"epoch": 0.13199682413656214,
"grad_norm": 0.859375,
"learning_rate": 0.000499987921462018,
"loss": 6.2739,
"mean_token_accuracy": 0.120875982940197,
"num_tokens": 2920153.0,
"step": 1330
},
{
"entropy": 6.545910024642945,
"epoch": 0.13249305279872967,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999875515468825,
"loss": 6.2668,
"mean_token_accuracy": 0.11391363963484764,
"num_tokens": 2930457.0,
"step": 1335
},
{
"entropy": 6.387372589111328,
"epoch": 0.13298928146089717,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999871760525403,
"loss": 6.3293,
"mean_token_accuracy": 0.11570871099829674,
"num_tokens": 2941305.0,
"step": 1340
},
{
"entropy": 6.462648534774781,
"epoch": 0.1334855101230647,
"grad_norm": 1.046875,
"learning_rate": 0.0004999867949790011,
"loss": 6.2594,
"mean_token_accuracy": 0.11346142515540122,
"num_tokens": 2951798.0,
"step": 1345
},
{
"entropy": 6.408753728866577,
"epoch": 0.13398173878523223,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999864083262743,
"loss": 6.1704,
"mean_token_accuracy": 0.11709224805235863,
"num_tokens": 2962486.0,
"step": 1350
},
{
"entropy": 6.384300184249878,
"epoch": 0.13447796744739976,
"grad_norm": 0.90625,
"learning_rate": 0.0004999860160943691,
"loss": 6.2681,
"mean_token_accuracy": 0.11269479319453239,
"num_tokens": 2974317.0,
"step": 1355
},
{
"entropy": 6.450789546966552,
"epoch": 0.1349741961095673,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999856182832958,
"loss": 6.1904,
"mean_token_accuracy": 0.11520234048366547,
"num_tokens": 2985440.0,
"step": 1360
},
{
"entropy": 6.400120830535888,
"epoch": 0.13547042477173482,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999852148930637,
"loss": 6.2605,
"mean_token_accuracy": 0.10958047136664391,
"num_tokens": 2997044.0,
"step": 1365
},
{
"entropy": 6.447021579742431,
"epoch": 0.13596665343390235,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999848059236832,
"loss": 6.2687,
"mean_token_accuracy": 0.1083021655678749,
"num_tokens": 3005965.0,
"step": 1370
},
{
"entropy": 6.3724065780639645,
"epoch": 0.13646288209606988,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999843913751643,
"loss": 6.1383,
"mean_token_accuracy": 0.11827749311923981,
"num_tokens": 3015031.0,
"step": 1375
},
{
"entropy": 6.378967905044556,
"epoch": 0.13695911075823738,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999839712475174,
"loss": 6.2074,
"mean_token_accuracy": 0.1124014399945736,
"num_tokens": 3026275.0,
"step": 1380
},
{
"entropy": 6.385042381286621,
"epoch": 0.13745533942040492,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999835455407526,
"loss": 6.2493,
"mean_token_accuracy": 0.11228244602680207,
"num_tokens": 3036778.0,
"step": 1385
},
{
"entropy": 6.488673353195191,
"epoch": 0.13795156808257245,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999831142548808,
"loss": 6.3516,
"mean_token_accuracy": 0.11366798952221871,
"num_tokens": 3048149.0,
"step": 1390
},
{
"entropy": 6.426563882827759,
"epoch": 0.13844779674473998,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999826773899124,
"loss": 6.2154,
"mean_token_accuracy": 0.11358957141637802,
"num_tokens": 3059149.0,
"step": 1395
},
{
"entropy": 6.404417276382446,
"epoch": 0.1389440254069075,
"grad_norm": 0.74609375,
"learning_rate": 0.0004999822349458586,
"loss": 6.2257,
"mean_token_accuracy": 0.11496646702289581,
"num_tokens": 3071613.0,
"step": 1400
},
{
"entropy": 6.5001030445098875,
"epoch": 0.13944025406907504,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999817869227301,
"loss": 6.369,
"mean_token_accuracy": 0.1097478836774826,
"num_tokens": 3084742.0,
"step": 1405
},
{
"entropy": 6.461193561553955,
"epoch": 0.13993648273124257,
"grad_norm": 1.015625,
"learning_rate": 0.0004999813333205381,
"loss": 6.2857,
"mean_token_accuracy": 0.11472547352313996,
"num_tokens": 3095356.0,
"step": 1410
},
{
"entropy": 6.409702301025391,
"epoch": 0.14043271139341007,
"grad_norm": 0.921875,
"learning_rate": 0.0004999808741392937,
"loss": 6.2669,
"mean_token_accuracy": 0.10798696279525757,
"num_tokens": 3107717.0,
"step": 1415
},
{
"entropy": 6.36297869682312,
"epoch": 0.1409289400555776,
"grad_norm": 0.80078125,
"learning_rate": 0.0004999804093790086,
"loss": 6.2518,
"mean_token_accuracy": 0.1127444364130497,
"num_tokens": 3119205.0,
"step": 1420
},
{
"entropy": 6.448381614685059,
"epoch": 0.14142516871774513,
"grad_norm": 0.86328125,
"learning_rate": 0.000499979939039694,
"loss": 6.3038,
"mean_token_accuracy": 0.11262011229991913,
"num_tokens": 3130564.0,
"step": 1425
},
{
"entropy": 6.450186538696289,
"epoch": 0.14192139737991266,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999794631213618,
"loss": 6.2395,
"mean_token_accuracy": 0.11082697659730911,
"num_tokens": 3141883.0,
"step": 1430
},
{
"entropy": 6.420169115066528,
"epoch": 0.1424176260420802,
"grad_norm": 0.828125,
"learning_rate": 0.0004999789816240236,
"loss": 6.2394,
"mean_token_accuracy": 0.10963290259242058,
"num_tokens": 3152817.0,
"step": 1435
},
{
"entropy": 6.358748054504394,
"epoch": 0.14291385470424772,
"grad_norm": 1.015625,
"learning_rate": 0.0004999784945476915,
"loss": 6.1567,
"mean_token_accuracy": 0.11966277733445167,
"num_tokens": 3163546.0,
"step": 1440
},
{
"entropy": 6.412687873840332,
"epoch": 0.14341008336641525,
"grad_norm": 1.015625,
"learning_rate": 0.0004999780018923775,
"loss": 6.207,
"mean_token_accuracy": 0.10822459608316422,
"num_tokens": 3173899.0,
"step": 1445
},
{
"entropy": 6.287933111190796,
"epoch": 0.14390631202858278,
"grad_norm": 0.875,
"learning_rate": 0.0004999775036580939,
"loss": 6.11,
"mean_token_accuracy": 0.11830191910266877,
"num_tokens": 3184433.0,
"step": 1450
},
{
"entropy": 6.4349367141723635,
"epoch": 0.1444025406907503,
"grad_norm": 0.875,
"learning_rate": 0.0004999769998448529,
"loss": 6.22,
"mean_token_accuracy": 0.11617036014795304,
"num_tokens": 3196077.0,
"step": 1455
},
{
"entropy": 6.256659460067749,
"epoch": 0.14489876935291782,
"grad_norm": 0.96875,
"learning_rate": 0.0004999764904526672,
"loss": 6.1422,
"mean_token_accuracy": 0.11813490390777588,
"num_tokens": 3206645.0,
"step": 1460
},
{
"entropy": 6.442942762374878,
"epoch": 0.14539499801508535,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999759754815492,
"loss": 6.2781,
"mean_token_accuracy": 0.11664444580674171,
"num_tokens": 3216786.0,
"step": 1465
},
{
"entropy": 6.370675897598266,
"epoch": 0.14589122667725288,
"grad_norm": 0.82421875,
"learning_rate": 0.0004999754549315118,
"loss": 6.1909,
"mean_token_accuracy": 0.11156717166304589,
"num_tokens": 3227384.0,
"step": 1470
},
{
"entropy": 6.271988248825073,
"epoch": 0.1463874553394204,
"grad_norm": 0.80859375,
"learning_rate": 0.000499974928802568,
"loss": 6.0496,
"mean_token_accuracy": 0.1194495476782322,
"num_tokens": 3238800.0,
"step": 1475
},
{
"entropy": 6.343875980377197,
"epoch": 0.14688368400158794,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999743970947306,
"loss": 6.1983,
"mean_token_accuracy": 0.11200206577777863,
"num_tokens": 3248627.0,
"step": 1480
},
{
"entropy": 6.364979076385498,
"epoch": 0.14737991266375547,
"grad_norm": 0.890625,
"learning_rate": 0.000499973859808013,
"loss": 6.2264,
"mean_token_accuracy": 0.11476045995950698,
"num_tokens": 3258989.0,
"step": 1485
},
{
"entropy": 6.351506185531616,
"epoch": 0.14787614132592297,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999733169424284,
"loss": 6.1945,
"mean_token_accuracy": 0.12000712603330613,
"num_tokens": 3269474.0,
"step": 1490
},
{
"entropy": 6.345127201080322,
"epoch": 0.1483723699880905,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999727684979904,
"loss": 6.1851,
"mean_token_accuracy": 0.11789674162864686,
"num_tokens": 3280453.0,
"step": 1495
},
{
"entropy": 6.31245641708374,
"epoch": 0.14886859865025803,
"grad_norm": 0.890625,
"learning_rate": 0.0004999722144747125,
"loss": 6.1203,
"mean_token_accuracy": 0.11437854543328285,
"num_tokens": 3290234.0,
"step": 1500
},
{
"entropy": 6.290363025665283,
"epoch": 0.14936482731242556,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999716548726085,
"loss": 6.1151,
"mean_token_accuracy": 0.12060677483677865,
"num_tokens": 3300943.0,
"step": 1505
},
{
"entropy": 6.312642621994018,
"epoch": 0.1498610559745931,
"grad_norm": 0.90625,
"learning_rate": 0.0004999710896916921,
"loss": 6.0684,
"mean_token_accuracy": 0.11790815740823746,
"num_tokens": 3311296.0,
"step": 1510
},
{
"entropy": 6.33518762588501,
"epoch": 0.15035728463676062,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999705189319774,
"loss": 6.2183,
"mean_token_accuracy": 0.11259444132447242,
"num_tokens": 3322310.0,
"step": 1515
},
{
"entropy": 6.35755672454834,
"epoch": 0.15085351329892815,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999699425934788,
"loss": 6.2299,
"mean_token_accuracy": 0.11464733257889748,
"num_tokens": 3334176.0,
"step": 1520
},
{
"entropy": 6.353241586685181,
"epoch": 0.15134974196109569,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999693606762103,
"loss": 6.2134,
"mean_token_accuracy": 0.11368680894374847,
"num_tokens": 3344839.0,
"step": 1525
},
{
"entropy": 6.369286584854126,
"epoch": 0.1518459706232632,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999687731801865,
"loss": 6.1841,
"mean_token_accuracy": 0.11411794796586036,
"num_tokens": 3357046.0,
"step": 1530
},
{
"entropy": 6.3504150867462155,
"epoch": 0.15234219928543072,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999681801054217,
"loss": 6.2434,
"mean_token_accuracy": 0.11438931301236152,
"num_tokens": 3367783.0,
"step": 1535
},
{
"entropy": 6.324097442626953,
"epoch": 0.15283842794759825,
"grad_norm": 0.91015625,
"learning_rate": 0.000499967581451931,
"loss": 6.1145,
"mean_token_accuracy": 0.11726456582546234,
"num_tokens": 3378529.0,
"step": 1540
},
{
"entropy": 6.375859260559082,
"epoch": 0.15333465660976578,
"grad_norm": 1.0078125,
"learning_rate": 0.000499966977219729,
"loss": 6.3061,
"mean_token_accuracy": 0.11245756894350052,
"num_tokens": 3389318.0,
"step": 1545
},
{
"entropy": 6.410177373886109,
"epoch": 0.1538308852719333,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999663674088306,
"loss": 6.3082,
"mean_token_accuracy": 0.1074009008705616,
"num_tokens": 3400603.0,
"step": 1550
},
{
"entropy": 6.406285333633423,
"epoch": 0.15432711393410084,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999657520192511,
"loss": 6.2016,
"mean_token_accuracy": 0.11285926327109337,
"num_tokens": 3410316.0,
"step": 1555
},
{
"entropy": 6.356779527664185,
"epoch": 0.15482334259626837,
"grad_norm": 0.79296875,
"learning_rate": 0.0004999651310510058,
"loss": 6.1416,
"mean_token_accuracy": 0.1178070843219757,
"num_tokens": 3422155.0,
"step": 1560
},
{
"entropy": 6.345295333862305,
"epoch": 0.15531957125843587,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999645045041099,
"loss": 6.0819,
"mean_token_accuracy": 0.11718083545565605,
"num_tokens": 3432892.0,
"step": 1565
},
{
"entropy": 6.321739959716797,
"epoch": 0.1558157999206034,
"grad_norm": 0.83984375,
"learning_rate": 0.0004999638723785791,
"loss": 6.2004,
"mean_token_accuracy": 0.11840709745883941,
"num_tokens": 3443967.0,
"step": 1570
},
{
"entropy": 6.270855236053467,
"epoch": 0.15631202858277093,
"grad_norm": 0.89453125,
"learning_rate": 0.000499963234674429,
"loss": 6.0916,
"mean_token_accuracy": 0.12166468054056168,
"num_tokens": 3454354.0,
"step": 1575
},
{
"entropy": 6.233002376556397,
"epoch": 0.15680825724493846,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999625913916754,
"loss": 6.1319,
"mean_token_accuracy": 0.11802673935890198,
"num_tokens": 3465232.0,
"step": 1580
},
{
"entropy": 6.242019701004028,
"epoch": 0.157304485907106,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999619425303343,
"loss": 6.0925,
"mean_token_accuracy": 0.11856844201683998,
"num_tokens": 3475443.0,
"step": 1585
},
{
"entropy": 6.3711628913879395,
"epoch": 0.15780071456927353,
"grad_norm": 0.828125,
"learning_rate": 0.0004999612880904219,
"loss": 6.1274,
"mean_token_accuracy": 0.11918400153517723,
"num_tokens": 3486643.0,
"step": 1590
},
{
"entropy": 6.28782172203064,
"epoch": 0.15829694323144106,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999606280719542,
"loss": 6.185,
"mean_token_accuracy": 0.11784008368849755,
"num_tokens": 3497557.0,
"step": 1595
},
{
"entropy": 6.421589040756226,
"epoch": 0.1587931718936086,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999599624749475,
"loss": 6.1637,
"mean_token_accuracy": 0.11806859821081161,
"num_tokens": 3509790.0,
"step": 1600
},
{
"entropy": 6.215350532531739,
"epoch": 0.1592894005557761,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999592912994187,
"loss": 6.0908,
"mean_token_accuracy": 0.11676915585994721,
"num_tokens": 3521577.0,
"step": 1605
},
{
"entropy": 6.3262897491455075,
"epoch": 0.15978562921794362,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999586145453842,
"loss": 6.1432,
"mean_token_accuracy": 0.11143859475851059,
"num_tokens": 3532126.0,
"step": 1610
},
{
"entropy": 6.2068733215332035,
"epoch": 0.16028185788011115,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999579322128608,
"loss": 5.972,
"mean_token_accuracy": 0.11642664968967438,
"num_tokens": 3542560.0,
"step": 1615
},
{
"entropy": 6.257067346572876,
"epoch": 0.16077808654227868,
"grad_norm": 0.921875,
"learning_rate": 0.0004999572443018654,
"loss": 6.1404,
"mean_token_accuracy": 0.11960313320159913,
"num_tokens": 3553863.0,
"step": 1620
},
{
"entropy": 6.282857131958008,
"epoch": 0.1612743152044462,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999565508124151,
"loss": 6.1603,
"mean_token_accuracy": 0.12152265384793282,
"num_tokens": 3564953.0,
"step": 1625
},
{
"entropy": 6.252542781829834,
"epoch": 0.16177054386661374,
"grad_norm": 0.78125,
"learning_rate": 0.000499955851744527,
"loss": 6.1588,
"mean_token_accuracy": 0.1182580165565014,
"num_tokens": 3576247.0,
"step": 1630
},
{
"entropy": 6.4327959537506105,
"epoch": 0.16226677252878127,
"grad_norm": 0.83984375,
"learning_rate": 0.0004999551470982185,
"loss": 6.1834,
"mean_token_accuracy": 0.11378399357199669,
"num_tokens": 3587600.0,
"step": 1635
},
{
"entropy": 6.352491140365601,
"epoch": 0.1627630011909488,
"grad_norm": 1.015625,
"learning_rate": 0.0004999544368735073,
"loss": 6.1676,
"mean_token_accuracy": 0.1208123430609703,
"num_tokens": 3597727.0,
"step": 1640
},
{
"entropy": 6.288443946838379,
"epoch": 0.1632592298531163,
"grad_norm": 0.90625,
"learning_rate": 0.0004999537210704106,
"loss": 6.1383,
"mean_token_accuracy": 0.1199887067079544,
"num_tokens": 3608066.0,
"step": 1645
},
{
"entropy": 6.309280347824097,
"epoch": 0.16375545851528384,
"grad_norm": 0.84375,
"learning_rate": 0.0004999529996889463,
"loss": 6.2006,
"mean_token_accuracy": 0.11445491313934326,
"num_tokens": 3618263.0,
"step": 1650
},
{
"entropy": 6.344433927536011,
"epoch": 0.16425168717745137,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999522727291325,
"loss": 6.1369,
"mean_token_accuracy": 0.12186064049601555,
"num_tokens": 3629465.0,
"step": 1655
},
{
"entropy": 6.342315721511841,
"epoch": 0.1647479158396189,
"grad_norm": 0.88671875,
"learning_rate": 0.000499951540190987,
"loss": 6.1583,
"mean_token_accuracy": 0.11373883932828903,
"num_tokens": 3639657.0,
"step": 1660
},
{
"entropy": 6.328229236602783,
"epoch": 0.16524414450178643,
"grad_norm": 0.859375,
"learning_rate": 0.0004999508020745281,
"loss": 6.1261,
"mean_token_accuracy": 0.11836064383387565,
"num_tokens": 3652374.0,
"step": 1665
},
{
"entropy": 6.265553331375122,
"epoch": 0.16574037316395396,
"grad_norm": 0.80859375,
"learning_rate": 0.0004999500583797739,
"loss": 6.1072,
"mean_token_accuracy": 0.12078006193041801,
"num_tokens": 3664083.0,
"step": 1670
},
{
"entropy": 6.2652623653411865,
"epoch": 0.1662366018261215,
"grad_norm": 0.84375,
"learning_rate": 0.0004999493091067431,
"loss": 6.1452,
"mean_token_accuracy": 0.12134512290358543,
"num_tokens": 3675314.0,
"step": 1675
},
{
"entropy": 6.273673057556152,
"epoch": 0.166732830488289,
"grad_norm": 0.796875,
"learning_rate": 0.0004999485542554541,
"loss": 6.0998,
"mean_token_accuracy": 0.11119712963700294,
"num_tokens": 3686019.0,
"step": 1680
},
{
"entropy": 6.370255327224731,
"epoch": 0.16722905915045652,
"grad_norm": 0.84375,
"learning_rate": 0.0004999477938259256,
"loss": 6.2015,
"mean_token_accuracy": 0.1111437164247036,
"num_tokens": 3698170.0,
"step": 1685
},
{
"entropy": 6.2584208965301515,
"epoch": 0.16772528781262405,
"grad_norm": 0.796875,
"learning_rate": 0.0004999470278181766,
"loss": 6.181,
"mean_token_accuracy": 0.11148267313838005,
"num_tokens": 3709243.0,
"step": 1690
},
{
"entropy": 6.344109964370728,
"epoch": 0.16822151647479158,
"grad_norm": 1.0,
"learning_rate": 0.0004999462562322261,
"loss": 6.1492,
"mean_token_accuracy": 0.11816644594073296,
"num_tokens": 3720196.0,
"step": 1695
},
{
"entropy": 6.374466037750244,
"epoch": 0.1687177451369591,
"grad_norm": 0.83984375,
"learning_rate": 0.000499945479068093,
"loss": 6.1937,
"mean_token_accuracy": 0.1157242514193058,
"num_tokens": 3732714.0,
"step": 1700
},
{
"entropy": 6.284525728225708,
"epoch": 0.16921397379912664,
"grad_norm": 0.72265625,
"learning_rate": 0.0004999446963257969,
"loss": 6.1734,
"mean_token_accuracy": 0.11130392551422119,
"num_tokens": 3746474.0,
"step": 1705
},
{
"entropy": 6.26992883682251,
"epoch": 0.16971020246129417,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999439080053571,
"loss": 6.0446,
"mean_token_accuracy": 0.12372053563594818,
"num_tokens": 3757528.0,
"step": 1710
},
{
"entropy": 6.382873582839966,
"epoch": 0.1702064311234617,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999431141067928,
"loss": 6.18,
"mean_token_accuracy": 0.11336961910128593,
"num_tokens": 3768360.0,
"step": 1715
},
{
"entropy": 6.226411628723144,
"epoch": 0.1707026597856292,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999423146301242,
"loss": 6.053,
"mean_token_accuracy": 0.11719192564487457,
"num_tokens": 3779809.0,
"step": 1720
},
{
"entropy": 6.197938585281372,
"epoch": 0.17119888844779674,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999415095753709,
"loss": 6.0517,
"mean_token_accuracy": 0.11497935056686401,
"num_tokens": 3791455.0,
"step": 1725
},
{
"entropy": 6.194182014465332,
"epoch": 0.17169511710996427,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999406989425527,
"loss": 6.0768,
"mean_token_accuracy": 0.11933901458978653,
"num_tokens": 3803209.0,
"step": 1730
},
{
"entropy": 6.272842121124268,
"epoch": 0.1721913457721318,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999398827316901,
"loss": 6.118,
"mean_token_accuracy": 0.11888522580265999,
"num_tokens": 3813553.0,
"step": 1735
},
{
"entropy": 6.242817163467407,
"epoch": 0.17268757443429933,
"grad_norm": 0.87890625,
"learning_rate": 0.000499939060942803,
"loss": 6.1491,
"mean_token_accuracy": 0.1162618339061737,
"num_tokens": 3824856.0,
"step": 1740
},
{
"entropy": 6.3598274230957035,
"epoch": 0.17318380309646686,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999382335759119,
"loss": 6.226,
"mean_token_accuracy": 0.1098976731300354,
"num_tokens": 3835489.0,
"step": 1745
},
{
"entropy": 6.3387504577636715,
"epoch": 0.1736800317586344,
"grad_norm": 1.09375,
"learning_rate": 0.0004999374006310372,
"loss": 6.0514,
"mean_token_accuracy": 0.11634560897946358,
"num_tokens": 3844838.0,
"step": 1750
},
{
"entropy": 6.168754911422729,
"epoch": 0.1741762604208019,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999365621081999,
"loss": 6.0342,
"mean_token_accuracy": 0.11432853490114211,
"num_tokens": 3855288.0,
"step": 1755
},
{
"entropy": 6.248133897781372,
"epoch": 0.17467248908296942,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999357180074204,
"loss": 6.0702,
"mean_token_accuracy": 0.12020740732550621,
"num_tokens": 3865585.0,
"step": 1760
},
{
"entropy": 6.307549571990966,
"epoch": 0.17516871774513695,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999348683287199,
"loss": 6.0612,
"mean_token_accuracy": 0.11682183891534806,
"num_tokens": 3877694.0,
"step": 1765
},
{
"entropy": 6.190867567062378,
"epoch": 0.17566494640730448,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999340130721192,
"loss": 6.0438,
"mean_token_accuracy": 0.12537719830870628,
"num_tokens": 3888802.0,
"step": 1770
},
{
"entropy": 6.209321069717407,
"epoch": 0.17616117506947201,
"grad_norm": 0.875,
"learning_rate": 0.0004999331522376396,
"loss": 6.0208,
"mean_token_accuracy": 0.11791236400604248,
"num_tokens": 3900294.0,
"step": 1775
},
{
"entropy": 6.202359390258789,
"epoch": 0.17665740373163955,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999322858253027,
"loss": 6.0417,
"mean_token_accuracy": 0.12317918911576271,
"num_tokens": 3910865.0,
"step": 1780
},
{
"entropy": 6.220766210556031,
"epoch": 0.17715363239380708,
"grad_norm": 0.890625,
"learning_rate": 0.0004999314138351297,
"loss": 6.1082,
"mean_token_accuracy": 0.1180627316236496,
"num_tokens": 3921960.0,
"step": 1785
},
{
"entropy": 6.266336345672608,
"epoch": 0.1776498610559746,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999305362671424,
"loss": 6.1053,
"mean_token_accuracy": 0.11726907566189766,
"num_tokens": 3934085.0,
"step": 1790
},
{
"entropy": 6.166120910644532,
"epoch": 0.1781460897181421,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999296531213624,
"loss": 5.9587,
"mean_token_accuracy": 0.1321515955030918,
"num_tokens": 3944532.0,
"step": 1795
},
{
"entropy": 6.186281824111939,
"epoch": 0.17864231838030964,
"grad_norm": 0.921875,
"learning_rate": 0.0004999287643978117,
"loss": 5.9965,
"mean_token_accuracy": 0.12600135430693626,
"num_tokens": 3954995.0,
"step": 1800
},
{
"entropy": 6.13634295463562,
"epoch": 0.17913854704247717,
"grad_norm": 0.796875,
"learning_rate": 0.0004999278700965124,
"loss": 6.045,
"mean_token_accuracy": 0.12481234520673752,
"num_tokens": 3966621.0,
"step": 1805
},
{
"entropy": 6.31223611831665,
"epoch": 0.1796347757046447,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999269702174866,
"loss": 6.1086,
"mean_token_accuracy": 0.11732375845313073,
"num_tokens": 3977568.0,
"step": 1810
},
{
"entropy": 6.178112936019898,
"epoch": 0.18013100436681223,
"grad_norm": 0.96875,
"learning_rate": 0.0004999260647607565,
"loss": 6.1296,
"mean_token_accuracy": 0.11571917533874512,
"num_tokens": 3988571.0,
"step": 1815
},
{
"entropy": 6.251215600967408,
"epoch": 0.18062723302897976,
"grad_norm": 0.953125,
"learning_rate": 0.0004999251537263448,
"loss": 5.9563,
"mean_token_accuracy": 0.12610897570848464,
"num_tokens": 3998380.0,
"step": 1820
},
{
"entropy": 6.153609037399292,
"epoch": 0.1811234616911473,
"grad_norm": 0.83984375,
"learning_rate": 0.0004999242371142739,
"loss": 6.1257,
"mean_token_accuracy": 0.12231617867946624,
"num_tokens": 4010017.0,
"step": 1825
},
{
"entropy": 6.336256694793701,
"epoch": 0.1816196903533148,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999233149245665,
"loss": 6.0924,
"mean_token_accuracy": 0.11703047156333923,
"num_tokens": 4020922.0,
"step": 1830
},
{
"entropy": 6.280692195892334,
"epoch": 0.18211591901548232,
"grad_norm": 0.875,
"learning_rate": 0.0004999223871572456,
"loss": 6.0664,
"mean_token_accuracy": 0.1190974049270153,
"num_tokens": 4030927.0,
"step": 1835
},
{
"entropy": 6.214899921417237,
"epoch": 0.18261214767764986,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999214538123342,
"loss": 6.1446,
"mean_token_accuracy": 0.11380955725908279,
"num_tokens": 4041375.0,
"step": 1840
},
{
"entropy": 6.2237224102020265,
"epoch": 0.18310837633981739,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999205148898554,
"loss": 6.1269,
"mean_token_accuracy": 0.11845389157533645,
"num_tokens": 4053213.0,
"step": 1845
},
{
"entropy": 6.275475740432739,
"epoch": 0.18360460500198492,
"grad_norm": 1.015625,
"learning_rate": 0.0004999195703898325,
"loss": 5.9784,
"mean_token_accuracy": 0.12606104388833045,
"num_tokens": 4064536.0,
"step": 1850
},
{
"entropy": 6.190648603439331,
"epoch": 0.18410083366415245,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999186203122889,
"loss": 6.06,
"mean_token_accuracy": 0.12409329041838646,
"num_tokens": 4075282.0,
"step": 1855
},
{
"entropy": 6.171055507659912,
"epoch": 0.18459706232631998,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999176646572482,
"loss": 6.057,
"mean_token_accuracy": 0.12450164556503296,
"num_tokens": 4085501.0,
"step": 1860
},
{
"entropy": 6.222785091400146,
"epoch": 0.1850932909884875,
"grad_norm": 0.890625,
"learning_rate": 0.000499916703424734,
"loss": 6.0077,
"mean_token_accuracy": 0.11859095394611359,
"num_tokens": 4095697.0,
"step": 1865
},
{
"entropy": 6.199232578277588,
"epoch": 0.185589519650655,
"grad_norm": 0.82421875,
"learning_rate": 0.0004999157366147702,
"loss": 6.0605,
"mean_token_accuracy": 0.1195147417485714,
"num_tokens": 4106459.0,
"step": 1870
},
{
"entropy": 6.221849536895752,
"epoch": 0.18608574831282254,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999147642273809,
"loss": 5.9945,
"mean_token_accuracy": 0.12537158504128457,
"num_tokens": 4118320.0,
"step": 1875
},
{
"entropy": 6.132491636276245,
"epoch": 0.18658197697499007,
"grad_norm": 0.87109375,
"learning_rate": 0.00049991378626259,
"loss": 6.018,
"mean_token_accuracy": 0.12773770093917847,
"num_tokens": 4131249.0,
"step": 1880
},
{
"entropy": 6.259493064880371,
"epoch": 0.1870782056371576,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999128027204219,
"loss": 6.0923,
"mean_token_accuracy": 0.12224588692188262,
"num_tokens": 4141480.0,
"step": 1885
},
{
"entropy": 6.227820491790771,
"epoch": 0.18757443429932513,
"grad_norm": 0.9296875,
"learning_rate": 0.000499911813600901,
"loss": 6.1457,
"mean_token_accuracy": 0.11545902863144875,
"num_tokens": 4152344.0,
"step": 1890
},
{
"entropy": 6.2940919399261475,
"epoch": 0.18807066296149266,
"grad_norm": 0.890625,
"learning_rate": 0.0004999108189040516,
"loss": 6.1088,
"mean_token_accuracy": 0.12004400044679642,
"num_tokens": 4164772.0,
"step": 1895
},
{
"entropy": 6.163094329833984,
"epoch": 0.1885668916236602,
"grad_norm": 0.90625,
"learning_rate": 0.0004999098186298986,
"loss": 6.0398,
"mean_token_accuracy": 0.11989247053861618,
"num_tokens": 4175511.0,
"step": 1900
},
{
"entropy": 6.184041833877563,
"epoch": 0.1890631202858277,
"grad_norm": 0.890625,
"learning_rate": 0.0004999088127784667,
"loss": 6.0138,
"mean_token_accuracy": 0.1198397345840931,
"num_tokens": 4187445.0,
"step": 1905
},
{
"entropy": 6.224808263778686,
"epoch": 0.18955934894799523,
"grad_norm": 0.828125,
"learning_rate": 0.000499907801349781,
"loss": 5.9992,
"mean_token_accuracy": 0.1206267774105072,
"num_tokens": 4198124.0,
"step": 1910
},
{
"entropy": 6.183636951446533,
"epoch": 0.19005557761016276,
"grad_norm": 0.9375,
"learning_rate": 0.0004999067843438664,
"loss": 6.0454,
"mean_token_accuracy": 0.11894752085208893,
"num_tokens": 4208954.0,
"step": 1915
},
{
"entropy": 6.204626941680909,
"epoch": 0.1905518062723303,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999057617607482,
"loss": 6.1093,
"mean_token_accuracy": 0.1197871819138527,
"num_tokens": 4218896.0,
"step": 1920
},
{
"entropy": 6.1618143081665036,
"epoch": 0.19104803493449782,
"grad_norm": 0.90625,
"learning_rate": 0.0004999047336004517,
"loss": 6.0051,
"mean_token_accuracy": 0.11776421442627907,
"num_tokens": 4228851.0,
"step": 1925
},
{
"entropy": 6.110290908813477,
"epoch": 0.19154426359666535,
"grad_norm": 0.96875,
"learning_rate": 0.0004999036998630024,
"loss": 5.9683,
"mean_token_accuracy": 0.12174378857016563,
"num_tokens": 4240377.0,
"step": 1930
},
{
"entropy": 6.116988277435302,
"epoch": 0.19204049225883288,
"grad_norm": 0.88671875,
"learning_rate": 0.000499902660548426,
"loss": 5.9615,
"mean_token_accuracy": 0.12193656638264656,
"num_tokens": 4251341.0,
"step": 1935
},
{
"entropy": 6.247050094604492,
"epoch": 0.1925367209210004,
"grad_norm": 0.9375,
"learning_rate": 0.0004999016156567482,
"loss": 6.0695,
"mean_token_accuracy": 0.12280801013112068,
"num_tokens": 4261206.0,
"step": 1940
},
{
"entropy": 6.215968275070191,
"epoch": 0.1930329495831679,
"grad_norm": 0.8203125,
"learning_rate": 0.000499900565187995,
"loss": 6.0677,
"mean_token_accuracy": 0.1141348622739315,
"num_tokens": 4272662.0,
"step": 1945
},
{
"entropy": 6.240064001083374,
"epoch": 0.19352917824533544,
"grad_norm": 0.859375,
"learning_rate": 0.0004998995091421924,
"loss": 6.0459,
"mean_token_accuracy": 0.11888503208756447,
"num_tokens": 4283302.0,
"step": 1950
},
{
"entropy": 6.114093589782715,
"epoch": 0.19402540690750297,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998984475193665,
"loss": 5.9975,
"mean_token_accuracy": 0.1221638023853302,
"num_tokens": 4296087.0,
"step": 1955
},
{
"entropy": 6.181862306594849,
"epoch": 0.1945216355696705,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998973803195439,
"loss": 6.0053,
"mean_token_accuracy": 0.12597885355353355,
"num_tokens": 4305918.0,
"step": 1960
},
{
"entropy": 6.123769617080688,
"epoch": 0.19501786423183803,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998963075427508,
"loss": 6.0646,
"mean_token_accuracy": 0.12296420186758042,
"num_tokens": 4316507.0,
"step": 1965
},
{
"entropy": 6.2180873394012455,
"epoch": 0.19551409289400556,
"grad_norm": 0.76953125,
"learning_rate": 0.0004998952291890138,
"loss": 6.0,
"mean_token_accuracy": 0.12596616744995118,
"num_tokens": 4329091.0,
"step": 1970
},
{
"entropy": 6.206251239776611,
"epoch": 0.1960103215561731,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998941452583599,
"loss": 6.0603,
"mean_token_accuracy": 0.12475561425089836,
"num_tokens": 4339266.0,
"step": 1975
},
{
"entropy": 6.0974133014678955,
"epoch": 0.1965065502183406,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998930557508156,
"loss": 5.984,
"mean_token_accuracy": 0.1275751493871212,
"num_tokens": 4349300.0,
"step": 1980
},
{
"entropy": 6.0830080032348635,
"epoch": 0.19700277888050813,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998919606664082,
"loss": 5.9019,
"mean_token_accuracy": 0.13283863738179208,
"num_tokens": 4360314.0,
"step": 1985
},
{
"entropy": 6.183152389526367,
"epoch": 0.19749900754267566,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998908600051648,
"loss": 5.9678,
"mean_token_accuracy": 0.1221761591732502,
"num_tokens": 4370077.0,
"step": 1990
},
{
"entropy": 6.084688329696656,
"epoch": 0.1979952362048432,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998897537671126,
"loss": 5.9713,
"mean_token_accuracy": 0.12218144908547401,
"num_tokens": 4381000.0,
"step": 1995
},
{
"entropy": 6.211720275878906,
"epoch": 0.19849146486701072,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998886419522792,
"loss": 6.1942,
"mean_token_accuracy": 0.11456988155841827,
"num_tokens": 4393485.0,
"step": 2000
},
{
"entropy": 6.298164701461792,
"epoch": 0.19898769352917825,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998875245606919,
"loss": 6.0114,
"mean_token_accuracy": 0.12225927039980888,
"num_tokens": 4404354.0,
"step": 2005
},
{
"entropy": 6.105899810791016,
"epoch": 0.19948392219134578,
"grad_norm": 0.984375,
"learning_rate": 0.0004998864015923788,
"loss": 6.1597,
"mean_token_accuracy": 0.12076655179262161,
"num_tokens": 4414818.0,
"step": 2010
},
{
"entropy": 6.200961780548096,
"epoch": 0.1999801508535133,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998852730473674,
"loss": 6.0344,
"mean_token_accuracy": 0.11524304151535034,
"num_tokens": 4426271.0,
"step": 2015
},
{
"entropy": 6.12263708114624,
"epoch": 0.2004763795156808,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998841389256859,
"loss": 6.0903,
"mean_token_accuracy": 0.12121652588248252,
"num_tokens": 4436981.0,
"step": 2020
},
{
"entropy": 6.139917802810669,
"epoch": 0.20097260817784834,
"grad_norm": 0.96875,
"learning_rate": 0.0004998829992273622,
"loss": 5.9254,
"mean_token_accuracy": 0.12804987505078316,
"num_tokens": 4446494.0,
"step": 2025
},
{
"entropy": 6.094701433181763,
"epoch": 0.20146883684001587,
"grad_norm": 0.89453125,
"learning_rate": 0.0004998818539524248,
"loss": 6.007,
"mean_token_accuracy": 0.12410036325454712,
"num_tokens": 4458535.0,
"step": 2030
},
{
"entropy": 6.231687307357788,
"epoch": 0.2019650655021834,
"grad_norm": 0.8984375,
"learning_rate": 0.000499880703100902,
"loss": 5.9882,
"mean_token_accuracy": 0.1257662445306778,
"num_tokens": 4469025.0,
"step": 2035
},
{
"entropy": 6.119232892990112,
"epoch": 0.20246129416435094,
"grad_norm": 0.953125,
"learning_rate": 0.0004998795466728223,
"loss": 5.9832,
"mean_token_accuracy": 0.1266017124056816,
"num_tokens": 4478431.0,
"step": 2040
},
{
"entropy": 6.141779518127441,
"epoch": 0.20295752282651847,
"grad_norm": 0.984375,
"learning_rate": 0.0004998783846682144,
"loss": 6.1012,
"mean_token_accuracy": 0.12176022753119468,
"num_tokens": 4490215.0,
"step": 2045
},
{
"entropy": 6.160814476013184,
"epoch": 0.203453751488686,
"grad_norm": 0.76171875,
"learning_rate": 0.0004998772170871072,
"loss": 5.9883,
"mean_token_accuracy": 0.12457352355122567,
"num_tokens": 4502895.0,
"step": 2050
},
{
"entropy": 6.102251768112183,
"epoch": 0.20394998015085353,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998760439295295,
"loss": 6.0066,
"mean_token_accuracy": 0.12112522423267365,
"num_tokens": 4514468.0,
"step": 2055
},
{
"entropy": 6.218370962142944,
"epoch": 0.20444620881302103,
"grad_norm": 0.953125,
"learning_rate": 0.0004998748651955105,
"loss": 6.0276,
"mean_token_accuracy": 0.11916945278644561,
"num_tokens": 4525787.0,
"step": 2060
},
{
"entropy": 6.161030578613281,
"epoch": 0.20494243747518856,
"grad_norm": 0.765625,
"learning_rate": 0.0004998736808850795,
"loss": 6.1047,
"mean_token_accuracy": 0.11898912787437439,
"num_tokens": 4537536.0,
"step": 2065
},
{
"entropy": 6.2428888320922855,
"epoch": 0.2054386661373561,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998724909982656,
"loss": 6.0472,
"mean_token_accuracy": 0.12252518236637115,
"num_tokens": 4549619.0,
"step": 2070
},
{
"entropy": 6.097667264938354,
"epoch": 0.20593489479952362,
"grad_norm": 0.86328125,
"learning_rate": 0.0004998712955350986,
"loss": 5.9866,
"mean_token_accuracy": 0.12310111299157142,
"num_tokens": 4560248.0,
"step": 2075
},
{
"entropy": 6.195798301696778,
"epoch": 0.20643112346169115,
"grad_norm": 0.79296875,
"learning_rate": 0.0004998700944956081,
"loss": 6.0975,
"mean_token_accuracy": 0.12141188234090805,
"num_tokens": 4571888.0,
"step": 2080
},
{
"entropy": 6.179491949081421,
"epoch": 0.20692735212385868,
"grad_norm": 0.84765625,
"learning_rate": 0.0004998688878798237,
"loss": 5.9365,
"mean_token_accuracy": 0.1305774599313736,
"num_tokens": 4583234.0,
"step": 2085
},
{
"entropy": 6.161084604263306,
"epoch": 0.2074235807860262,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998676756877753,
"loss": 6.1601,
"mean_token_accuracy": 0.11846253573894501,
"num_tokens": 4595252.0,
"step": 2090
},
{
"entropy": 6.131542110443116,
"epoch": 0.20791980944819372,
"grad_norm": 0.83203125,
"learning_rate": 0.0004998664579194934,
"loss": 5.9236,
"mean_token_accuracy": 0.1237091027200222,
"num_tokens": 4607091.0,
"step": 2095
},
{
"entropy": 6.140899753570556,
"epoch": 0.20841603811036125,
"grad_norm": 0.8515625,
"learning_rate": 0.0004998652345750077,
"loss": 5.8958,
"mean_token_accuracy": 0.1278851717710495,
"num_tokens": 4618003.0,
"step": 2100
},
{
"entropy": 6.041016483306885,
"epoch": 0.20891226677252878,
"grad_norm": 0.80859375,
"learning_rate": 0.0004998640056543487,
"loss": 5.8996,
"mean_token_accuracy": 0.12397338300943375,
"num_tokens": 4628429.0,
"step": 2105
},
{
"entropy": 6.15621018409729,
"epoch": 0.2094084954346963,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998627711575469,
"loss": 5.9676,
"mean_token_accuracy": 0.12933185324072838,
"num_tokens": 4639414.0,
"step": 2110
},
{
"entropy": 6.167100429534912,
"epoch": 0.20990472409686384,
"grad_norm": 0.8046875,
"learning_rate": 0.000499861531084633,
"loss": 6.0398,
"mean_token_accuracy": 0.12240175604820251,
"num_tokens": 4652594.0,
"step": 2115
},
{
"entropy": 6.079341125488281,
"epoch": 0.21040095275903137,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998602854356376,
"loss": 5.8855,
"mean_token_accuracy": 0.1307040624320507,
"num_tokens": 4662981.0,
"step": 2120
},
{
"entropy": 6.053981876373291,
"epoch": 0.2108971814211989,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998590342105917,
"loss": 5.9951,
"mean_token_accuracy": 0.12460904493927956,
"num_tokens": 4672469.0,
"step": 2125
},
{
"entropy": 6.146594381332397,
"epoch": 0.21139341008336643,
"grad_norm": 0.90625,
"learning_rate": 0.0004998577774095262,
"loss": 5.9799,
"mean_token_accuracy": 0.12103921249508857,
"num_tokens": 4683640.0,
"step": 2130
},
{
"entropy": 6.100749874114991,
"epoch": 0.21188963874553393,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998565150324724,
"loss": 5.9284,
"mean_token_accuracy": 0.12689334973692895,
"num_tokens": 4694032.0,
"step": 2135
},
{
"entropy": 6.005304336547852,
"epoch": 0.21238586740770146,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998552470794616,
"loss": 5.8162,
"mean_token_accuracy": 0.12930263206362724,
"num_tokens": 4704253.0,
"step": 2140
},
{
"entropy": 6.0581738471984865,
"epoch": 0.212882096069869,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998539735505251,
"loss": 5.9544,
"mean_token_accuracy": 0.12062570005655289,
"num_tokens": 4715327.0,
"step": 2145
},
{
"entropy": 6.2206943988800045,
"epoch": 0.21337832473203652,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998526944456946,
"loss": 5.9851,
"mean_token_accuracy": 0.12153983116149902,
"num_tokens": 4726450.0,
"step": 2150
},
{
"entropy": 6.0526422500610355,
"epoch": 0.21387455339420405,
"grad_norm": 0.83203125,
"learning_rate": 0.0004998514097650018,
"loss": 6.0067,
"mean_token_accuracy": 0.12227371484041213,
"num_tokens": 4736630.0,
"step": 2155
},
{
"entropy": 6.161049175262451,
"epoch": 0.21437078205637158,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998501195084786,
"loss": 5.8963,
"mean_token_accuracy": 0.12918635457754135,
"num_tokens": 4746592.0,
"step": 2160
},
{
"entropy": 6.086433601379395,
"epoch": 0.21486701071853911,
"grad_norm": 0.890625,
"learning_rate": 0.0004998488236761569,
"loss": 6.072,
"mean_token_accuracy": 0.12488910630345344,
"num_tokens": 4757824.0,
"step": 2165
},
{
"entropy": 6.144498586654663,
"epoch": 0.21536323938070662,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998475222680688,
"loss": 5.9959,
"mean_token_accuracy": 0.1257177084684372,
"num_tokens": 4768453.0,
"step": 2170
},
{
"entropy": 6.051328659057617,
"epoch": 0.21585946804287415,
"grad_norm": 0.90625,
"learning_rate": 0.0004998462152842468,
"loss": 5.9771,
"mean_token_accuracy": 0.12266893535852433,
"num_tokens": 4779664.0,
"step": 2175
},
{
"entropy": 6.203923177719116,
"epoch": 0.21635569670504168,
"grad_norm": 0.84765625,
"learning_rate": 0.000499844902724723,
"loss": 5.997,
"mean_token_accuracy": 0.12400704175233841,
"num_tokens": 4791012.0,
"step": 2180
},
{
"entropy": 6.080739498138428,
"epoch": 0.2168519253672092,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998435845895302,
"loss": 5.8727,
"mean_token_accuracy": 0.1246095396578312,
"num_tokens": 4800578.0,
"step": 2185
},
{
"entropy": 6.023191165924072,
"epoch": 0.21734815402937674,
"grad_norm": 0.90625,
"learning_rate": 0.000499842260878701,
"loss": 5.8847,
"mean_token_accuracy": 0.12296021655201912,
"num_tokens": 4810940.0,
"step": 2190
},
{
"entropy": 6.089705514907837,
"epoch": 0.21784438269154427,
"grad_norm": 0.7734375,
"learning_rate": 0.0004998409315922683,
"loss": 6.0182,
"mean_token_accuracy": 0.12229126617312432,
"num_tokens": 4822696.0,
"step": 2195
},
{
"entropy": 6.09975118637085,
"epoch": 0.2183406113537118,
"grad_norm": 0.90625,
"learning_rate": 0.0004998395967302649,
"loss": 5.8893,
"mean_token_accuracy": 0.12959264144301413,
"num_tokens": 4833669.0,
"step": 2200
},
{
"entropy": 6.1851764678955075,
"epoch": 0.21883684001587933,
"grad_norm": 0.86328125,
"learning_rate": 0.000499838256292724,
"loss": 6.0744,
"mean_token_accuracy": 0.12145713344216347,
"num_tokens": 4844021.0,
"step": 2205
},
{
"entropy": 6.052931642532348,
"epoch": 0.21933306867804683,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998369102796787,
"loss": 5.9505,
"mean_token_accuracy": 0.12860263735055924,
"num_tokens": 4855336.0,
"step": 2210
},
{
"entropy": 6.104745626449585,
"epoch": 0.21982929734021436,
"grad_norm": 0.7734375,
"learning_rate": 0.0004998355586911627,
"loss": 5.9443,
"mean_token_accuracy": 0.12385106608271598,
"num_tokens": 4867713.0,
"step": 2215
},
{
"entropy": 6.0858691215515135,
"epoch": 0.2203255260023819,
"grad_norm": 0.78515625,
"learning_rate": 0.0004998342015272093,
"loss": 5.9244,
"mean_token_accuracy": 0.1223891668021679,
"num_tokens": 4879134.0,
"step": 2220
},
{
"entropy": 6.108216285705566,
"epoch": 0.22082175466454942,
"grad_norm": 0.796875,
"learning_rate": 0.0004998328387878521,
"loss": 6.0808,
"mean_token_accuracy": 0.12277590483427048,
"num_tokens": 4891755.0,
"step": 2225
},
{
"entropy": 6.2470619678497314,
"epoch": 0.22131798332671695,
"grad_norm": 0.90625,
"learning_rate": 0.0004998314704731249,
"loss": 6.0916,
"mean_token_accuracy": 0.12046385034918786,
"num_tokens": 4903264.0,
"step": 2230
},
{
"entropy": 6.077593660354614,
"epoch": 0.22181421198888449,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998300965830618,
"loss": 5.9,
"mean_token_accuracy": 0.12378231808543205,
"num_tokens": 4912533.0,
"step": 2235
},
{
"entropy": 5.9839921474456785,
"epoch": 0.22231044065105202,
"grad_norm": 0.8359375,
"learning_rate": 0.0004998287171176968,
"loss": 5.8996,
"mean_token_accuracy": 0.1311133660376072,
"num_tokens": 4923288.0,
"step": 2240
},
{
"entropy": 6.105050659179687,
"epoch": 0.22280666931321952,
"grad_norm": 0.83984375,
"learning_rate": 0.0004998273320770641,
"loss": 5.9154,
"mean_token_accuracy": 0.12592516243457794,
"num_tokens": 4934759.0,
"step": 2245
},
{
"entropy": 6.072722053527832,
"epoch": 0.22330289797538705,
"grad_norm": 0.96484375,
"learning_rate": 0.000499825941461198,
"loss": 5.9621,
"mean_token_accuracy": 0.12398277670145035,
"num_tokens": 4947024.0,
"step": 2250
},
{
"entropy": 6.092696714401245,
"epoch": 0.22379912663755458,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998245452701329,
"loss": 5.9152,
"mean_token_accuracy": 0.12754956483840943,
"num_tokens": 4959281.0,
"step": 2255
},
{
"entropy": 6.13616452217102,
"epoch": 0.2242953552997221,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998231435039036,
"loss": 5.9725,
"mean_token_accuracy": 0.12141189724206924,
"num_tokens": 4970466.0,
"step": 2260
},
{
"entropy": 6.069933271408081,
"epoch": 0.22479158396188964,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998217361625449,
"loss": 5.9236,
"mean_token_accuracy": 0.12711238041520118,
"num_tokens": 4981135.0,
"step": 2265
},
{
"entropy": 6.105356216430664,
"epoch": 0.22528781262405717,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998203232460915,
"loss": 5.9797,
"mean_token_accuracy": 0.12980691865086555,
"num_tokens": 4991292.0,
"step": 2270
},
{
"entropy": 6.059658765792847,
"epoch": 0.2257840412862247,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998189047545785,
"loss": 5.9345,
"mean_token_accuracy": 0.13011907190084457,
"num_tokens": 5002806.0,
"step": 2275
},
{
"entropy": 6.100206136703491,
"epoch": 0.22628026994839223,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998174806880411,
"loss": 5.8843,
"mean_token_accuracy": 0.1296177588403225,
"num_tokens": 5013474.0,
"step": 2280
},
{
"entropy": 6.06908540725708,
"epoch": 0.22677649861055973,
"grad_norm": 0.8515625,
"learning_rate": 0.0004998160510465146,
"loss": 5.9249,
"mean_token_accuracy": 0.1318567730486393,
"num_tokens": 5025151.0,
"step": 2285
},
{
"entropy": 5.967230415344238,
"epoch": 0.22727272727272727,
"grad_norm": 0.8203125,
"learning_rate": 0.0004998146158300345,
"loss": 5.9196,
"mean_token_accuracy": 0.1306504562497139,
"num_tokens": 5037166.0,
"step": 2290
},
{
"entropy": 6.142732429504394,
"epoch": 0.2277689559348948,
"grad_norm": 0.82421875,
"learning_rate": 0.0004998131750386364,
"loss": 5.9706,
"mean_token_accuracy": 0.12619220688939095,
"num_tokens": 5047452.0,
"step": 2295
},
{
"entropy": 6.048137044906616,
"epoch": 0.22826518459706233,
"grad_norm": 0.875,
"learning_rate": 0.0004998117286723559,
"loss": 5.9756,
"mean_token_accuracy": 0.12324338406324387,
"num_tokens": 5058501.0,
"step": 2300
},
{
"entropy": 6.132171249389648,
"epoch": 0.22876141325922986,
"grad_norm": 0.9296875,
"learning_rate": 0.000499810276731229,
"loss": 5.974,
"mean_token_accuracy": 0.12393502593040466,
"num_tokens": 5069716.0,
"step": 2305
},
{
"entropy": 6.158532667160034,
"epoch": 0.2292576419213974,
"grad_norm": 0.96875,
"learning_rate": 0.0004998088192152918,
"loss": 5.9587,
"mean_token_accuracy": 0.12171528786420822,
"num_tokens": 5078509.0,
"step": 2310
},
{
"entropy": 6.056684732437134,
"epoch": 0.22975387058356492,
"grad_norm": 0.90625,
"learning_rate": 0.0004998073561245801,
"loss": 5.9908,
"mean_token_accuracy": 0.12283852398395538,
"num_tokens": 5090355.0,
"step": 2315
},
{
"entropy": 6.118961763381958,
"epoch": 0.23025009924573242,
"grad_norm": 0.83203125,
"learning_rate": 0.0004998058874591305,
"loss": 5.8839,
"mean_token_accuracy": 0.12737965807318688,
"num_tokens": 5100582.0,
"step": 2320
},
{
"entropy": 6.098464822769165,
"epoch": 0.23074632790789995,
"grad_norm": 0.76953125,
"learning_rate": 0.0004998044132189793,
"loss": 5.9514,
"mean_token_accuracy": 0.12292122989892959,
"num_tokens": 5112176.0,
"step": 2325
},
{
"entropy": 6.076971912384034,
"epoch": 0.23124255657006748,
"grad_norm": 1.0546875,
"learning_rate": 0.000499802933404163,
"loss": 5.8787,
"mean_token_accuracy": 0.1285627380013466,
"num_tokens": 5122840.0,
"step": 2330
},
{
"entropy": 6.050347518920899,
"epoch": 0.231738785232235,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998014480147183,
"loss": 5.8587,
"mean_token_accuracy": 0.13070411905646323,
"num_tokens": 5132926.0,
"step": 2335
},
{
"entropy": 6.0548614978790285,
"epoch": 0.23223501389440254,
"grad_norm": 0.921875,
"learning_rate": 0.0004997999570506823,
"loss": 5.8476,
"mean_token_accuracy": 0.12983267456293107,
"num_tokens": 5143394.0,
"step": 2340
},
{
"entropy": 5.978992509841919,
"epoch": 0.23273124255657007,
"grad_norm": 0.80078125,
"learning_rate": 0.0004997984605120917,
"loss": 5.8757,
"mean_token_accuracy": 0.12346786707639694,
"num_tokens": 5155710.0,
"step": 2345
},
{
"entropy": 5.942594051361084,
"epoch": 0.2332274712187376,
"grad_norm": 0.96875,
"learning_rate": 0.0004997969583989835,
"loss": 5.7929,
"mean_token_accuracy": 0.12445841357111931,
"num_tokens": 5167065.0,
"step": 2350
},
{
"entropy": 6.06046576499939,
"epoch": 0.23372369988090513,
"grad_norm": 0.890625,
"learning_rate": 0.0004997954507113952,
"loss": 5.9018,
"mean_token_accuracy": 0.12778859511017798,
"num_tokens": 5177535.0,
"step": 2355
},
{
"entropy": 6.021834182739258,
"epoch": 0.23421992854307264,
"grad_norm": 0.859375,
"learning_rate": 0.0004997939374493642,
"loss": 5.8305,
"mean_token_accuracy": 0.12911173850297927,
"num_tokens": 5188324.0,
"step": 2360
},
{
"entropy": 6.018092298507691,
"epoch": 0.23471615720524017,
"grad_norm": 0.88671875,
"learning_rate": 0.000499792418612928,
"loss": 5.9024,
"mean_token_accuracy": 0.13002333790063858,
"num_tokens": 5199223.0,
"step": 2365
},
{
"entropy": 5.9740711688995365,
"epoch": 0.2352123858674077,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997908942021241,
"loss": 5.89,
"mean_token_accuracy": 0.13267461881041526,
"num_tokens": 5210080.0,
"step": 2370
},
{
"entropy": 6.135750961303711,
"epoch": 0.23570861452957523,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997893642169904,
"loss": 6.0108,
"mean_token_accuracy": 0.12664430662989618,
"num_tokens": 5220867.0,
"step": 2375
},
{
"entropy": 6.101852035522461,
"epoch": 0.23620484319174276,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997878286575648,
"loss": 5.8597,
"mean_token_accuracy": 0.12812694609165193,
"num_tokens": 5231451.0,
"step": 2380
},
{
"entropy": 6.028659343719482,
"epoch": 0.2367010718539103,
"grad_norm": 0.8828125,
"learning_rate": 0.0004997862875238855,
"loss": 5.9642,
"mean_token_accuracy": 0.12428494766354561,
"num_tokens": 5244643.0,
"step": 2385
},
{
"entropy": 6.113288068771363,
"epoch": 0.23719730051607782,
"grad_norm": 0.984375,
"learning_rate": 0.0004997847408159907,
"loss": 5.979,
"mean_token_accuracy": 0.12018007338047028,
"num_tokens": 5256279.0,
"step": 2390
},
{
"entropy": 6.0652384757995605,
"epoch": 0.23769352917824532,
"grad_norm": 0.84375,
"learning_rate": 0.0004997831885339186,
"loss": 5.9475,
"mean_token_accuracy": 0.1281052775681019,
"num_tokens": 5268498.0,
"step": 2395
},
{
"entropy": 6.014727401733398,
"epoch": 0.23818975784041285,
"grad_norm": 0.83984375,
"learning_rate": 0.0004997816306777077,
"loss": 5.908,
"mean_token_accuracy": 0.1289649024605751,
"num_tokens": 5280869.0,
"step": 2400
},
{
"entropy": 6.235253620147705,
"epoch": 0.23868598650258038,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997800672473969,
"loss": 6.0058,
"mean_token_accuracy": 0.12169916182756424,
"num_tokens": 5293082.0,
"step": 2405
},
{
"entropy": 5.997746467590332,
"epoch": 0.2391822151647479,
"grad_norm": 1.03125,
"learning_rate": 0.0004997784982430247,
"loss": 5.8106,
"mean_token_accuracy": 0.13437482044100763,
"num_tokens": 5304039.0,
"step": 2410
},
{
"entropy": 5.944843721389771,
"epoch": 0.23967844382691544,
"grad_norm": 0.84375,
"learning_rate": 0.0004997769236646301,
"loss": 5.8851,
"mean_token_accuracy": 0.1293253131210804,
"num_tokens": 5315016.0,
"step": 2415
},
{
"entropy": 6.173792791366577,
"epoch": 0.24017467248908297,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997753435122522,
"loss": 6.0082,
"mean_token_accuracy": 0.12303022742271423,
"num_tokens": 5326090.0,
"step": 2420
},
{
"entropy": 6.016424989700317,
"epoch": 0.2406709011512505,
"grad_norm": 0.91796875,
"learning_rate": 0.00049977375778593,
"loss": 5.7936,
"mean_token_accuracy": 0.1281319923698902,
"num_tokens": 5336423.0,
"step": 2425
},
{
"entropy": 6.052116632461548,
"epoch": 0.24116712981341804,
"grad_norm": 0.8671875,
"learning_rate": 0.0004997721664857032,
"loss": 5.9728,
"mean_token_accuracy": 0.12626754492521286,
"num_tokens": 5346982.0,
"step": 2430
},
{
"entropy": 5.9658690929412845,
"epoch": 0.24166335847558554,
"grad_norm": 0.90625,
"learning_rate": 0.0004997705696116108,
"loss": 5.8618,
"mean_token_accuracy": 0.1352499283850193,
"num_tokens": 5359210.0,
"step": 2435
},
{
"entropy": 6.060595321655273,
"epoch": 0.24215958713775307,
"grad_norm": 0.8125,
"learning_rate": 0.0004997689671636926,
"loss": 5.8833,
"mean_token_accuracy": 0.1232607200741768,
"num_tokens": 5369891.0,
"step": 2440
},
{
"entropy": 6.024409484863281,
"epoch": 0.2426558157999206,
"grad_norm": 0.80078125,
"learning_rate": 0.0004997673591419884,
"loss": 5.8619,
"mean_token_accuracy": 0.1305900402367115,
"num_tokens": 5380850.0,
"step": 2445
},
{
"entropy": 6.061957502365113,
"epoch": 0.24315204446208813,
"grad_norm": 0.7890625,
"learning_rate": 0.0004997657455465381,
"loss": 5.928,
"mean_token_accuracy": 0.13309514969587327,
"num_tokens": 5391903.0,
"step": 2450
},
{
"entropy": 6.0940954208374025,
"epoch": 0.24364827312425566,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997641263773815,
"loss": 5.978,
"mean_token_accuracy": 0.12671018466353418,
"num_tokens": 5402517.0,
"step": 2455
},
{
"entropy": 5.941332578659058,
"epoch": 0.2441445017864232,
"grad_norm": 0.84765625,
"learning_rate": 0.000499762501634559,
"loss": 5.8878,
"mean_token_accuracy": 0.12774158492684365,
"num_tokens": 5414110.0,
"step": 2460
},
{
"entropy": 6.125403642654419,
"epoch": 0.24464073044859072,
"grad_norm": 0.8828125,
"learning_rate": 0.0004997608713181108,
"loss": 5.9006,
"mean_token_accuracy": 0.12905448973178862,
"num_tokens": 5424466.0,
"step": 2465
},
{
"entropy": 6.040532350540161,
"epoch": 0.24513695911075822,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997592354280772,
"loss": 5.9095,
"mean_token_accuracy": 0.134385783970356,
"num_tokens": 5434461.0,
"step": 2470
},
{
"entropy": 5.967836189270019,
"epoch": 0.24563318777292575,
"grad_norm": 0.91796875,
"learning_rate": 0.000499757593964499,
"loss": 5.8913,
"mean_token_accuracy": 0.11896291747689247,
"num_tokens": 5445921.0,
"step": 2475
},
{
"entropy": 6.038508701324463,
"epoch": 0.24612941643509328,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997559469274166,
"loss": 5.8771,
"mean_token_accuracy": 0.13031587675213813,
"num_tokens": 5456283.0,
"step": 2480
},
{
"entropy": 6.024613332748413,
"epoch": 0.24662564509726081,
"grad_norm": 0.85546875,
"learning_rate": 0.0004997542943168712,
"loss": 5.9114,
"mean_token_accuracy": 0.1300675705075264,
"num_tokens": 5467616.0,
"step": 2485
},
{
"entropy": 6.012180805206299,
"epoch": 0.24712187375942835,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997526361329035,
"loss": 5.8744,
"mean_token_accuracy": 0.13522633239626886,
"num_tokens": 5478050.0,
"step": 2490
},
{
"entropy": 6.102998495101929,
"epoch": 0.24761810242159588,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997509723755547,
"loss": 5.8714,
"mean_token_accuracy": 0.12875011935830116,
"num_tokens": 5487185.0,
"step": 2495
},
{
"entropy": 5.998979711532593,
"epoch": 0.2481143310837634,
"grad_norm": 0.83203125,
"learning_rate": 0.0004997493030448662,
"loss": 5.8924,
"mean_token_accuracy": 0.12383673936128617,
"num_tokens": 5498802.0,
"step": 2500
},
{
"entropy": 6.0036334037780765,
"epoch": 0.24861055974593094,
"grad_norm": 0.83984375,
"learning_rate": 0.0004997476281408793,
"loss": 5.9009,
"mean_token_accuracy": 0.12360760122537613,
"num_tokens": 5510285.0,
"step": 2505
},
{
"entropy": 6.045953416824341,
"epoch": 0.24910678840809844,
"grad_norm": 0.84375,
"learning_rate": 0.0004997459476636354,
"loss": 5.7868,
"mean_token_accuracy": 0.13784014880657197,
"num_tokens": 5520028.0,
"step": 2510
},
{
"entropy": 5.9735174655914305,
"epoch": 0.24960301707026597,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997442616131763,
"loss": 5.912,
"mean_token_accuracy": 0.12859713956713675,
"num_tokens": 5529271.0,
"step": 2515
},
{
"entropy": 6.054106569290161,
"epoch": 0.25009924573243353,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997425699895438,
"loss": 5.826,
"mean_token_accuracy": 0.12208940386772156,
"num_tokens": 5539396.0,
"step": 2520
},
{
"entropy": 6.049839782714844,
"epoch": 0.25059547439460106,
"grad_norm": 0.76953125,
"learning_rate": 0.0004997408727927798,
"loss": 5.9723,
"mean_token_accuracy": 0.1294700786471367,
"num_tokens": 5551674.0,
"step": 2525
},
{
"entropy": 6.055530405044555,
"epoch": 0.25109170305676853,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997391700229264,
"loss": 5.8948,
"mean_token_accuracy": 0.1298077531158924,
"num_tokens": 5562749.0,
"step": 2530
},
{
"entropy": 6.030540132522583,
"epoch": 0.25158793171893606,
"grad_norm": 0.85546875,
"learning_rate": 0.000499737461680026,
"loss": 5.9,
"mean_token_accuracy": 0.12933559864759445,
"num_tokens": 5573478.0,
"step": 2535
},
{
"entropy": 5.963146209716797,
"epoch": 0.2520841603811036,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997357477641207,
"loss": 5.8908,
"mean_token_accuracy": 0.13493381813168526,
"num_tokens": 5584546.0,
"step": 2540
},
{
"entropy": 5.964659547805786,
"epoch": 0.2525803890432711,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997340282752532,
"loss": 5.7847,
"mean_token_accuracy": 0.1342097096145153,
"num_tokens": 5594004.0,
"step": 2545
},
{
"entropy": 6.085031366348266,
"epoch": 0.25307661770543866,
"grad_norm": 0.87890625,
"learning_rate": 0.000499732303213466,
"loss": 5.8353,
"mean_token_accuracy": 0.12420367822051048,
"num_tokens": 5606095.0,
"step": 2550
},
{
"entropy": 6.014924669265747,
"epoch": 0.2535728463676062,
"grad_norm": 1.03125,
"learning_rate": 0.0004997305725788019,
"loss": 5.8899,
"mean_token_accuracy": 0.12461917325854302,
"num_tokens": 5615596.0,
"step": 2555
},
{
"entropy": 5.9634756565094,
"epoch": 0.2540690750297737,
"grad_norm": 0.83984375,
"learning_rate": 0.000499728836371304,
"loss": 5.8139,
"mean_token_accuracy": 0.12745841071009636,
"num_tokens": 5626753.0,
"step": 2560
},
{
"entropy": 5.963172578811646,
"epoch": 0.25456530369194125,
"grad_norm": 0.88671875,
"learning_rate": 0.0004997270945910151,
"loss": 5.8388,
"mean_token_accuracy": 0.12756719589233398,
"num_tokens": 5636135.0,
"step": 2565
},
{
"entropy": 6.079827070236206,
"epoch": 0.2550615323541088,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997253472379785,
"loss": 5.9086,
"mean_token_accuracy": 0.12276982814073563,
"num_tokens": 5647074.0,
"step": 2570
},
{
"entropy": 6.04095311164856,
"epoch": 0.2555577610162763,
"grad_norm": 0.78515625,
"learning_rate": 0.0004997235943122375,
"loss": 5.9298,
"mean_token_accuracy": 0.12193780019879341,
"num_tokens": 5658600.0,
"step": 2575
},
{
"entropy": 5.954120779037476,
"epoch": 0.25605398967844384,
"grad_norm": 0.90625,
"learning_rate": 0.0004997218358138356,
"loss": 5.8184,
"mean_token_accuracy": 0.12850069105625153,
"num_tokens": 5668453.0,
"step": 2580
},
{
"entropy": 6.052977800369263,
"epoch": 0.25655021834061137,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997200717428165,
"loss": 5.8802,
"mean_token_accuracy": 0.12774283811450005,
"num_tokens": 5678855.0,
"step": 2585
},
{
"entropy": 6.0446836948394775,
"epoch": 0.2570464470027789,
"grad_norm": 0.88671875,
"learning_rate": 0.0004997183020992237,
"loss": 5.9438,
"mean_token_accuracy": 0.12405224218964576,
"num_tokens": 5690063.0,
"step": 2590
},
{
"entropy": 5.979913902282715,
"epoch": 0.25754267566494643,
"grad_norm": 0.8984375,
"learning_rate": 0.0004997165268831014,
"loss": 5.843,
"mean_token_accuracy": 0.1319989301264286,
"num_tokens": 5701487.0,
"step": 2595
},
{
"entropy": 5.977937984466553,
"epoch": 0.25803890432711396,
"grad_norm": 0.81640625,
"learning_rate": 0.0004997147460944933,
"loss": 5.8313,
"mean_token_accuracy": 0.131146914511919,
"num_tokens": 5713034.0,
"step": 2600
},
{
"entropy": 6.07123966217041,
"epoch": 0.25853513298928144,
"grad_norm": 0.80078125,
"learning_rate": 0.0004997129597334438,
"loss": 5.8811,
"mean_token_accuracy": 0.13053280711174012,
"num_tokens": 5725469.0,
"step": 2605
},
{
"entropy": 5.960973072052002,
"epoch": 0.25903136165144897,
"grad_norm": 0.7890625,
"learning_rate": 0.0004997111677999971,
"loss": 5.87,
"mean_token_accuracy": 0.12475038319826126,
"num_tokens": 5737099.0,
"step": 2610
},
{
"entropy": 6.032576656341552,
"epoch": 0.2595275903136165,
"grad_norm": 0.81640625,
"learning_rate": 0.0004997093702941976,
"loss": 5.872,
"mean_token_accuracy": 0.1248766042292118,
"num_tokens": 5748756.0,
"step": 2615
},
{
"entropy": 6.0417296409606935,
"epoch": 0.260023818975784,
"grad_norm": 0.94140625,
"learning_rate": 0.00049970756721609,
"loss": 5.8727,
"mean_token_accuracy": 0.12974709644913673,
"num_tokens": 5759598.0,
"step": 2620
},
{
"entropy": 6.012347316741943,
"epoch": 0.26052004763795156,
"grad_norm": 0.796875,
"learning_rate": 0.000499705758565719,
"loss": 5.9157,
"mean_token_accuracy": 0.12809571251273155,
"num_tokens": 5770397.0,
"step": 2625
},
{
"entropy": 6.098965167999268,
"epoch": 0.2610162763001191,
"grad_norm": 0.8359375,
"learning_rate": 0.0004997039443431292,
"loss": 5.983,
"mean_token_accuracy": 0.11926335170865059,
"num_tokens": 5780935.0,
"step": 2630
},
{
"entropy": 6.08192253112793,
"epoch": 0.2615125049622866,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997021245483659,
"loss": 5.8366,
"mean_token_accuracy": 0.12855549082159995,
"num_tokens": 5791557.0,
"step": 2635
},
{
"entropy": 5.944977807998657,
"epoch": 0.26200873362445415,
"grad_norm": 0.85546875,
"learning_rate": 0.000499700299181474,
"loss": 5.874,
"mean_token_accuracy": 0.12412384524941444,
"num_tokens": 5802233.0,
"step": 2640
},
{
"entropy": 5.984538125991821,
"epoch": 0.2625049622866217,
"grad_norm": 0.86328125,
"learning_rate": 0.0004996984682424989,
"loss": 5.7929,
"mean_token_accuracy": 0.13563756346702577,
"num_tokens": 5813424.0,
"step": 2645
},
{
"entropy": 5.963291454315185,
"epoch": 0.2630011909487892,
"grad_norm": 0.74609375,
"learning_rate": 0.000499696631731486,
"loss": 5.7802,
"mean_token_accuracy": 0.13106578662991525,
"num_tokens": 5825652.0,
"step": 2650
},
{
"entropy": 6.034928655624389,
"epoch": 0.26349741961095674,
"grad_norm": 0.86328125,
"learning_rate": 0.0004996947896484809,
"loss": 5.865,
"mean_token_accuracy": 0.13201481997966766,
"num_tokens": 5836603.0,
"step": 2655
},
{
"entropy": 5.982569742202759,
"epoch": 0.26399364827312427,
"grad_norm": 0.890625,
"learning_rate": 0.000499692941993529,
"loss": 5.9128,
"mean_token_accuracy": 0.1318881943821907,
"num_tokens": 5849365.0,
"step": 2660
},
{
"entropy": 6.044679498672485,
"epoch": 0.2644898769352918,
"grad_norm": 0.828125,
"learning_rate": 0.0004996910887666765,
"loss": 5.8354,
"mean_token_accuracy": 0.1336039572954178,
"num_tokens": 5860305.0,
"step": 2665
},
{
"entropy": 5.926130056381226,
"epoch": 0.26498610559745933,
"grad_norm": 0.828125,
"learning_rate": 0.000499689229967969,
"loss": 5.7666,
"mean_token_accuracy": 0.13250631913542749,
"num_tokens": 5872272.0,
"step": 2670
},
{
"entropy": 5.9937135696411135,
"epoch": 0.26548233425962686,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996873655974529,
"loss": 5.8152,
"mean_token_accuracy": 0.12750425711274146,
"num_tokens": 5884107.0,
"step": 2675
},
{
"entropy": 5.906749153137207,
"epoch": 0.26597856292179434,
"grad_norm": 0.92578125,
"learning_rate": 0.0004996854956551743,
"loss": 5.8092,
"mean_token_accuracy": 0.13456383347511292,
"num_tokens": 5894060.0,
"step": 2680
},
{
"entropy": 5.9789345264434814,
"epoch": 0.26647479158396187,
"grad_norm": 0.8671875,
"learning_rate": 0.0004996836201411794,
"loss": 5.8238,
"mean_token_accuracy": 0.1331551544368267,
"num_tokens": 5904804.0,
"step": 2685
},
{
"entropy": 6.056610536575318,
"epoch": 0.2669710202461294,
"grad_norm": 1.0234375,
"learning_rate": 0.000499681739055515,
"loss": 5.8861,
"mean_token_accuracy": 0.1274963565170765,
"num_tokens": 5916496.0,
"step": 2690
},
{
"entropy": 5.9887939453125,
"epoch": 0.26746724890829693,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996798523982276,
"loss": 5.831,
"mean_token_accuracy": 0.1256636656820774,
"num_tokens": 5926920.0,
"step": 2695
},
{
"entropy": 6.095494651794434,
"epoch": 0.26796347757046446,
"grad_norm": 0.875,
"learning_rate": 0.000499677960169364,
"loss": 5.935,
"mean_token_accuracy": 0.12686496153473853,
"num_tokens": 5940098.0,
"step": 2700
},
{
"entropy": 5.972264385223388,
"epoch": 0.268459706232632,
"grad_norm": 0.90234375,
"learning_rate": 0.0004996760623689711,
"loss": 5.7841,
"mean_token_accuracy": 0.13265340775251389,
"num_tokens": 5949687.0,
"step": 2705
},
{
"entropy": 5.992936277389527,
"epoch": 0.2689559348947995,
"grad_norm": 0.8125,
"learning_rate": 0.000499674158997096,
"loss": 5.9045,
"mean_token_accuracy": 0.1267220564186573,
"num_tokens": 5960777.0,
"step": 2710
},
{
"entropy": 6.038538122177124,
"epoch": 0.26945216355696705,
"grad_norm": 0.8828125,
"learning_rate": 0.0004996722500537859,
"loss": 5.8095,
"mean_token_accuracy": 0.12924240455031394,
"num_tokens": 5972298.0,
"step": 2715
},
{
"entropy": 5.961693143844604,
"epoch": 0.2699483922191346,
"grad_norm": 0.8515625,
"learning_rate": 0.0004996703355390881,
"loss": 5.9214,
"mean_token_accuracy": 0.12794365584850312,
"num_tokens": 5985613.0,
"step": 2720
},
{
"entropy": 6.043231201171875,
"epoch": 0.2704446208813021,
"grad_norm": 0.8515625,
"learning_rate": 0.00049966841545305,
"loss": 5.9656,
"mean_token_accuracy": 0.1257025733590126,
"num_tokens": 5997530.0,
"step": 2725
},
{
"entropy": 6.031536388397217,
"epoch": 0.27094084954346964,
"grad_norm": 0.83203125,
"learning_rate": 0.0004996664897957195,
"loss": 5.8365,
"mean_token_accuracy": 0.12756068184971808,
"num_tokens": 6008123.0,
"step": 2730
},
{
"entropy": 6.018695735931397,
"epoch": 0.2714370782056372,
"grad_norm": 0.85546875,
"learning_rate": 0.0004996645585671441,
"loss": 5.852,
"mean_token_accuracy": 0.13391014859080314,
"num_tokens": 6020578.0,
"step": 2735
},
{
"entropy": 5.907762765884399,
"epoch": 0.2719333068678047,
"grad_norm": 0.85546875,
"learning_rate": 0.0004996626217673717,
"loss": 5.7793,
"mean_token_accuracy": 0.1337601587176323,
"num_tokens": 6031625.0,
"step": 2740
},
{
"entropy": 5.958955860137939,
"epoch": 0.27242953552997223,
"grad_norm": 0.8203125,
"learning_rate": 0.0004996606793964504,
"loss": 5.8286,
"mean_token_accuracy": 0.13649252206087112,
"num_tokens": 6041320.0,
"step": 2745
},
{
"entropy": 6.01719069480896,
"epoch": 0.27292576419213976,
"grad_norm": 0.8359375,
"learning_rate": 0.0004996587314544284,
"loss": 5.8369,
"mean_token_accuracy": 0.12664347812533377,
"num_tokens": 6052627.0,
"step": 2750
},
{
"entropy": 5.954709434509278,
"epoch": 0.27342199285430724,
"grad_norm": 0.82421875,
"learning_rate": 0.0004996567779413539,
"loss": 5.8036,
"mean_token_accuracy": 0.13628075420856475,
"num_tokens": 6063468.0,
"step": 2755
},
{
"entropy": 5.991281652450562,
"epoch": 0.27391822151647477,
"grad_norm": 0.84765625,
"learning_rate": 0.0004996548188572755,
"loss": 5.7814,
"mean_token_accuracy": 0.1363678827881813,
"num_tokens": 6074208.0,
"step": 2760
},
{
"entropy": 5.994483947753906,
"epoch": 0.2744144501786423,
"grad_norm": 0.90625,
"learning_rate": 0.0004996528542022416,
"loss": 5.8366,
"mean_token_accuracy": 0.12587395310401917,
"num_tokens": 6085068.0,
"step": 2765
},
{
"entropy": 5.959670734405518,
"epoch": 0.27491067884080983,
"grad_norm": 0.8046875,
"learning_rate": 0.000499650883976301,
"loss": 5.844,
"mean_token_accuracy": 0.12782294154167176,
"num_tokens": 6096445.0,
"step": 2770
},
{
"entropy": 5.91315975189209,
"epoch": 0.27540690750297736,
"grad_norm": 0.90234375,
"learning_rate": 0.0004996489081795026,
"loss": 5.7401,
"mean_token_accuracy": 0.13289162740111352,
"num_tokens": 6106527.0,
"step": 2775
},
{
"entropy": 5.91305251121521,
"epoch": 0.2759031361651449,
"grad_norm": 1.015625,
"learning_rate": 0.0004996469268118953,
"loss": 5.7589,
"mean_token_accuracy": 0.13679778575897217,
"num_tokens": 6115923.0,
"step": 2780
},
{
"entropy": 5.868895196914673,
"epoch": 0.2763993648273124,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996449398735283,
"loss": 5.6526,
"mean_token_accuracy": 0.13616614267230034,
"num_tokens": 6125861.0,
"step": 2785
},
{
"entropy": 6.010179805755615,
"epoch": 0.27689559348947995,
"grad_norm": 0.80859375,
"learning_rate": 0.0004996429473644509,
"loss": 5.8537,
"mean_token_accuracy": 0.13069555088877677,
"num_tokens": 6137189.0,
"step": 2790
},
{
"entropy": 5.97644534111023,
"epoch": 0.2773918221516475,
"grad_norm": 0.77734375,
"learning_rate": 0.0004996409492847124,
"loss": 5.7804,
"mean_token_accuracy": 0.13484650626778602,
"num_tokens": 6147920.0,
"step": 2795
},
{
"entropy": 5.939912557601929,
"epoch": 0.277888050813815,
"grad_norm": 0.828125,
"learning_rate": 0.0004996389456343624,
"loss": 5.7712,
"mean_token_accuracy": 0.13595055788755417,
"num_tokens": 6159371.0,
"step": 2800
},
{
"entropy": 6.066695976257324,
"epoch": 0.27838427947598254,
"grad_norm": 0.75,
"learning_rate": 0.0004996369364134507,
"loss": 5.9699,
"mean_token_accuracy": 0.12258584052324295,
"num_tokens": 6172679.0,
"step": 2805
},
{
"entropy": 5.927506589889527,
"epoch": 0.2788805081381501,
"grad_norm": 0.8359375,
"learning_rate": 0.000499634921622027,
"loss": 5.8298,
"mean_token_accuracy": 0.12628915533423424,
"num_tokens": 6184244.0,
"step": 2810
},
{
"entropy": 5.933115148544312,
"epoch": 0.2793767368003176,
"grad_norm": 0.859375,
"learning_rate": 0.0004996329012601413,
"loss": 5.706,
"mean_token_accuracy": 0.1285858377814293,
"num_tokens": 6194826.0,
"step": 2815
},
{
"entropy": 5.979615592956543,
"epoch": 0.27987296546248513,
"grad_norm": 0.921875,
"learning_rate": 0.0004996308753278437,
"loss": 5.8456,
"mean_token_accuracy": 0.13074849992990495,
"num_tokens": 6206808.0,
"step": 2820
},
{
"entropy": 6.013004970550537,
"epoch": 0.28036919412465267,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996288438251843,
"loss": 5.8605,
"mean_token_accuracy": 0.12552075758576392,
"num_tokens": 6217303.0,
"step": 2825
},
{
"entropy": 5.992705678939819,
"epoch": 0.28086542278682014,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996268067522137,
"loss": 5.8603,
"mean_token_accuracy": 0.12485205754637718,
"num_tokens": 6227965.0,
"step": 2830
},
{
"entropy": 6.082558679580688,
"epoch": 0.28136165144898767,
"grad_norm": 0.953125,
"learning_rate": 0.0004996247641089823,
"loss": 5.919,
"mean_token_accuracy": 0.12674327939748764,
"num_tokens": 6239011.0,
"step": 2835
},
{
"entropy": 5.880522966384888,
"epoch": 0.2818578801111552,
"grad_norm": 0.96875,
"learning_rate": 0.0004996227158955409,
"loss": 5.7612,
"mean_token_accuracy": 0.13352836593985556,
"num_tokens": 6249084.0,
"step": 2840
},
{
"entropy": 5.958843946456909,
"epoch": 0.28235410877332273,
"grad_norm": 0.87109375,
"learning_rate": 0.00049962066211194,
"loss": 5.8334,
"mean_token_accuracy": 0.12839528918266296,
"num_tokens": 6260706.0,
"step": 2845
},
{
"entropy": 5.931930923461914,
"epoch": 0.28285033743549026,
"grad_norm": 0.90625,
"learning_rate": 0.0004996186027582308,
"loss": 5.7522,
"mean_token_accuracy": 0.13635355457663537,
"num_tokens": 6272848.0,
"step": 2850
},
{
"entropy": 5.946898508071899,
"epoch": 0.2833465660976578,
"grad_norm": 0.92578125,
"learning_rate": 0.0004996165378344642,
"loss": 5.7663,
"mean_token_accuracy": 0.13342537730932236,
"num_tokens": 6283820.0,
"step": 2855
},
{
"entropy": 5.972999238967896,
"epoch": 0.2838427947598253,
"grad_norm": 0.890625,
"learning_rate": 0.0004996144673406914,
"loss": 5.9339,
"mean_token_accuracy": 0.12071966528892517,
"num_tokens": 6294193.0,
"step": 2860
},
{
"entropy": 5.960216379165649,
"epoch": 0.28433902342199285,
"grad_norm": 0.8828125,
"learning_rate": 0.0004996123912769639,
"loss": 5.7676,
"mean_token_accuracy": 0.1383022852241993,
"num_tokens": 6304407.0,
"step": 2865
},
{
"entropy": 5.893539810180664,
"epoch": 0.2848352520841604,
"grad_norm": 0.80859375,
"learning_rate": 0.0004996103096433332,
"loss": 5.8421,
"mean_token_accuracy": 0.12886647060513495,
"num_tokens": 6316311.0,
"step": 2870
},
{
"entropy": 6.00091586112976,
"epoch": 0.2853314807463279,
"grad_norm": 0.82421875,
"learning_rate": 0.0004996082224398506,
"loss": 5.8144,
"mean_token_accuracy": 0.13291785418987273,
"num_tokens": 6327248.0,
"step": 2875
},
{
"entropy": 5.986534309387207,
"epoch": 0.28582770940849545,
"grad_norm": 0.828125,
"learning_rate": 0.0004996061296665682,
"loss": 5.8325,
"mean_token_accuracy": 0.12824170365929605,
"num_tokens": 6338494.0,
"step": 2880
},
{
"entropy": 5.9684656143188475,
"epoch": 0.286323938070663,
"grad_norm": 0.921875,
"learning_rate": 0.0004996040313235377,
"loss": 5.7233,
"mean_token_accuracy": 0.13372908607125283,
"num_tokens": 6349038.0,
"step": 2885
},
{
"entropy": 5.902253913879394,
"epoch": 0.2868201667328305,
"grad_norm": 0.80078125,
"learning_rate": 0.0004996019274108112,
"loss": 5.8233,
"mean_token_accuracy": 0.13003271594643592,
"num_tokens": 6360180.0,
"step": 2890
},
{
"entropy": 5.906910562515259,
"epoch": 0.28731639539499804,
"grad_norm": 0.875,
"learning_rate": 0.000499599817928441,
"loss": 5.7596,
"mean_token_accuracy": 0.13310305699706076,
"num_tokens": 6370931.0,
"step": 2895
},
{
"entropy": 5.877241802215576,
"epoch": 0.28781262405716557,
"grad_norm": 0.82421875,
"learning_rate": 0.0004995977028764792,
"loss": 5.7375,
"mean_token_accuracy": 0.13817125037312508,
"num_tokens": 6381810.0,
"step": 2900
},
{
"entropy": 5.958535528182983,
"epoch": 0.28830885271933304,
"grad_norm": 0.8203125,
"learning_rate": 0.0004995955822549783,
"loss": 5.7447,
"mean_token_accuracy": 0.1337203674018383,
"num_tokens": 6392268.0,
"step": 2905
},
{
"entropy": 5.885456800460815,
"epoch": 0.2888050813815006,
"grad_norm": 0.8359375,
"learning_rate": 0.0004995934560639909,
"loss": 5.8172,
"mean_token_accuracy": 0.13473332896828652,
"num_tokens": 6403650.0,
"step": 2910
},
{
"entropy": 5.995740032196045,
"epoch": 0.2893013100436681,
"grad_norm": 0.8984375,
"learning_rate": 0.0004995913243035699,
"loss": 5.7959,
"mean_token_accuracy": 0.12757119312882423,
"num_tokens": 6414672.0,
"step": 2915
},
{
"entropy": 5.970022487640381,
"epoch": 0.28979753870583563,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995891869737678,
"loss": 5.8593,
"mean_token_accuracy": 0.12999678775668144,
"num_tokens": 6424916.0,
"step": 2920
},
{
"entropy": 5.960818958282471,
"epoch": 0.29029376736800316,
"grad_norm": 0.85546875,
"learning_rate": 0.0004995870440746379,
"loss": 5.8059,
"mean_token_accuracy": 0.13466615676879884,
"num_tokens": 6435950.0,
"step": 2925
},
{
"entropy": 5.905315399169922,
"epoch": 0.2907899960301707,
"grad_norm": 0.8125,
"learning_rate": 0.0004995848956062333,
"loss": 5.7742,
"mean_token_accuracy": 0.1341630332171917,
"num_tokens": 6446336.0,
"step": 2930
},
{
"entropy": 5.917783546447754,
"epoch": 0.2912862246923382,
"grad_norm": 0.83984375,
"learning_rate": 0.0004995827415686071,
"loss": 5.8452,
"mean_token_accuracy": 0.13145026564598083,
"num_tokens": 6457655.0,
"step": 2935
},
{
"entropy": 5.985751056671143,
"epoch": 0.29178245335450576,
"grad_norm": 0.92578125,
"learning_rate": 0.000499580581961813,
"loss": 5.7235,
"mean_token_accuracy": 0.13669102862477303,
"num_tokens": 6469627.0,
"step": 2940
},
{
"entropy": 5.918372678756714,
"epoch": 0.2922786820166733,
"grad_norm": 0.8359375,
"learning_rate": 0.0004995784167859043,
"loss": 5.825,
"mean_token_accuracy": 0.129074639827013,
"num_tokens": 6480351.0,
"step": 2945
},
{
"entropy": 6.008989477157593,
"epoch": 0.2927749106788408,
"grad_norm": 0.890625,
"learning_rate": 0.0004995762460409349,
"loss": 5.8548,
"mean_token_accuracy": 0.1337804116308689,
"num_tokens": 6491651.0,
"step": 2950
},
{
"entropy": 5.920342969894409,
"epoch": 0.29327113934100835,
"grad_norm": 0.79296875,
"learning_rate": 0.0004995740697269584,
"loss": 5.6923,
"mean_token_accuracy": 0.13204924538731574,
"num_tokens": 6502531.0,
"step": 2955
},
{
"entropy": 5.878091621398926,
"epoch": 0.2937673680031759,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995718878440289,
"loss": 5.7131,
"mean_token_accuracy": 0.14013013020157813,
"num_tokens": 6511938.0,
"step": 2960
},
{
"entropy": 6.0145423412323,
"epoch": 0.2942635966653434,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995697003922006,
"loss": 5.8628,
"mean_token_accuracy": 0.1239828608930111,
"num_tokens": 6522775.0,
"step": 2965
},
{
"entropy": 5.953707361221314,
"epoch": 0.29475982532751094,
"grad_norm": 0.80859375,
"learning_rate": 0.0004995675073715274,
"loss": 5.681,
"mean_token_accuracy": 0.13056408539414405,
"num_tokens": 6534660.0,
"step": 2970
},
{
"entropy": 5.901784515380859,
"epoch": 0.29525605398967847,
"grad_norm": 0.90625,
"learning_rate": 0.0004995653087820641,
"loss": 5.8299,
"mean_token_accuracy": 0.13245153054594994,
"num_tokens": 6545673.0,
"step": 2975
},
{
"entropy": 5.9352250576019285,
"epoch": 0.29575228265184594,
"grad_norm": 0.96484375,
"learning_rate": 0.000499563104623865,
"loss": 5.7862,
"mean_token_accuracy": 0.1360513038933277,
"num_tokens": 6556912.0,
"step": 2980
},
{
"entropy": 5.981804084777832,
"epoch": 0.2962485113140135,
"grad_norm": 0.859375,
"learning_rate": 0.0004995608948969849,
"loss": 5.8789,
"mean_token_accuracy": 0.12770932540297508,
"num_tokens": 6567532.0,
"step": 2985
},
{
"entropy": 5.955900192260742,
"epoch": 0.296744739976181,
"grad_norm": 0.98828125,
"learning_rate": 0.0004995586796014784,
"loss": 5.8553,
"mean_token_accuracy": 0.13003772795200347,
"num_tokens": 6577830.0,
"step": 2990
},
{
"entropy": 5.927827978134156,
"epoch": 0.29724096863834853,
"grad_norm": 0.859375,
"learning_rate": 0.0004995564587374005,
"loss": 5.7476,
"mean_token_accuracy": 0.13181802555918692,
"num_tokens": 6587254.0,
"step": 2995
},
{
"entropy": 5.962104320526123,
"epoch": 0.29773719730051607,
"grad_norm": 0.87890625,
"learning_rate": 0.0004995542323048063,
"loss": 5.934,
"mean_token_accuracy": 0.12504435703158379,
"num_tokens": 6598294.0,
"step": 3000
},
{
"epoch": 0.29773719730051607,
"eval_entropy": 5.7928955505681055,
"eval_loss": 5.8702263832092285,
"eval_mean_token_accuracy": 0.13254506512458533,
"eval_num_tokens": 6598294.0,
"eval_runtime": 20.7058,
"eval_samples_per_second": 1626.066,
"eval_steps_per_second": 203.276,
"step": 3000
},
{
"entropy": 6.02299394607544,
"epoch": 0.2982334259626836,
"grad_norm": 0.88671875,
"learning_rate": 0.000499552000303751,
"loss": 5.8132,
"mean_token_accuracy": 0.13359044566750528,
"num_tokens": 6608772.0,
"step": 3005
},
{
"entropy": 5.907866477966309,
"epoch": 0.2987296546248511,
"grad_norm": 0.84765625,
"learning_rate": 0.00049954976273429,
"loss": 5.8093,
"mean_token_accuracy": 0.1337615132331848,
"num_tokens": 6619973.0,
"step": 3010
},
{
"entropy": 5.9264623641967775,
"epoch": 0.29922588328701866,
"grad_norm": 0.90625,
"learning_rate": 0.0004995475195964787,
"loss": 5.8292,
"mean_token_accuracy": 0.13057641312479973,
"num_tokens": 6631172.0,
"step": 3015
},
{
"entropy": 5.9544103145599365,
"epoch": 0.2997221119491862,
"grad_norm": 0.8671875,
"learning_rate": 0.0004995452708903727,
"loss": 5.8188,
"mean_token_accuracy": 0.12836946919560432,
"num_tokens": 6641343.0,
"step": 3020
},
{
"entropy": 5.912361097335816,
"epoch": 0.3002183406113537,
"grad_norm": 0.84765625,
"learning_rate": 0.000499543016616028,
"loss": 5.7628,
"mean_token_accuracy": 0.13423253819346428,
"num_tokens": 6652892.0,
"step": 3025
},
{
"entropy": 5.965789175033569,
"epoch": 0.30071456927352125,
"grad_norm": 0.82421875,
"learning_rate": 0.0004995407567735002,
"loss": 5.7887,
"mean_token_accuracy": 0.13260234743356705,
"num_tokens": 6664177.0,
"step": 3030
},
{
"entropy": 5.903903007507324,
"epoch": 0.3012107979356888,
"grad_norm": 0.953125,
"learning_rate": 0.0004995384913628456,
"loss": 5.7612,
"mean_token_accuracy": 0.13183784186840058,
"num_tokens": 6674402.0,
"step": 3035
},
{
"entropy": 5.972148370742798,
"epoch": 0.3017070265978563,
"grad_norm": 0.87109375,
"learning_rate": 0.0004995362203841202,
"loss": 5.8938,
"mean_token_accuracy": 0.12839276492595672,
"num_tokens": 6684886.0,
"step": 3040
},
{
"entropy": 5.954754209518432,
"epoch": 0.30220325526002384,
"grad_norm": 0.92578125,
"learning_rate": 0.0004995339438373804,
"loss": 5.7202,
"mean_token_accuracy": 0.135014508664608,
"num_tokens": 6694774.0,
"step": 3045
},
{
"entropy": 5.888301420211792,
"epoch": 0.30269948392219137,
"grad_norm": 0.89453125,
"learning_rate": 0.0004995316617226825,
"loss": 5.7818,
"mean_token_accuracy": 0.13432007953524588,
"num_tokens": 6705346.0,
"step": 3050
},
{
"entropy": 5.871857500076294,
"epoch": 0.30319571258435885,
"grad_norm": 0.86328125,
"learning_rate": 0.0004995293740400833,
"loss": 5.7312,
"mean_token_accuracy": 0.13942001089453698,
"num_tokens": 6715579.0,
"step": 3055
},
{
"entropy": 5.8535606384277346,
"epoch": 0.3036919412465264,
"grad_norm": 0.859375,
"learning_rate": 0.0004995270807896395,
"loss": 5.6946,
"mean_token_accuracy": 0.13676538467407226,
"num_tokens": 6726382.0,
"step": 3060
},
{
"entropy": 5.935056352615357,
"epoch": 0.3041881699086939,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995247819714079,
"loss": 5.7698,
"mean_token_accuracy": 0.1337558187544346,
"num_tokens": 6737245.0,
"step": 3065
},
{
"entropy": 5.979394245147705,
"epoch": 0.30468439857086144,
"grad_norm": 0.8671875,
"learning_rate": 0.0004995224775854454,
"loss": 5.9624,
"mean_token_accuracy": 0.12612314149737358,
"num_tokens": 6749113.0,
"step": 3070
},
{
"entropy": 6.009639263153076,
"epoch": 0.30518062723302897,
"grad_norm": 0.78125,
"learning_rate": 0.0004995201676318093,
"loss": 5.8219,
"mean_token_accuracy": 0.1318033404648304,
"num_tokens": 6761987.0,
"step": 3075
},
{
"entropy": 5.9234983921051025,
"epoch": 0.3056768558951965,
"grad_norm": 0.8984375,
"learning_rate": 0.0004995178521105568,
"loss": 5.766,
"mean_token_accuracy": 0.12556311190128328,
"num_tokens": 6771279.0,
"step": 3080
},
{
"entropy": 5.889677429199219,
"epoch": 0.30617308455736403,
"grad_norm": 0.921875,
"learning_rate": 0.0004995155310217454,
"loss": 5.6911,
"mean_token_accuracy": 0.1361276350915432,
"num_tokens": 6781038.0,
"step": 3085
},
{
"entropy": 5.8956520557403564,
"epoch": 0.30666931321953156,
"grad_norm": 0.83984375,
"learning_rate": 0.0004995132043654327,
"loss": 5.8346,
"mean_token_accuracy": 0.13275470435619355,
"num_tokens": 6792686.0,
"step": 3090
},
{
"entropy": 5.951267910003662,
"epoch": 0.3071655418816991,
"grad_norm": 0.80859375,
"learning_rate": 0.0004995108721416763,
"loss": 5.8074,
"mean_token_accuracy": 0.139174947142601,
"num_tokens": 6804414.0,
"step": 3095
},
{
"entropy": 5.929559230804443,
"epoch": 0.3076617705438666,
"grad_norm": 0.83984375,
"learning_rate": 0.0004995085343505339,
"loss": 5.7652,
"mean_token_accuracy": 0.13180396631360053,
"num_tokens": 6816180.0,
"step": 3100
},
{
"entropy": 5.945132160186768,
"epoch": 0.30815799920603415,
"grad_norm": 0.875,
"learning_rate": 0.0004995061909920638,
"loss": 5.778,
"mean_token_accuracy": 0.1298554077744484,
"num_tokens": 6827114.0,
"step": 3105
},
{
"entropy": 5.969542980194092,
"epoch": 0.3086542278682017,
"grad_norm": 0.94140625,
"learning_rate": 0.0004995038420663238,
"loss": 5.8373,
"mean_token_accuracy": 0.13185039311647415,
"num_tokens": 6836953.0,
"step": 3110
},
{
"entropy": 5.944265985488892,
"epoch": 0.3091504565303692,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995014875733724,
"loss": 5.8746,
"mean_token_accuracy": 0.12786564826965333,
"num_tokens": 6847963.0,
"step": 3115
},
{
"entropy": 5.981411457061768,
"epoch": 0.30964668519253674,
"grad_norm": 0.859375,
"learning_rate": 0.0004994991275132678,
"loss": 5.8869,
"mean_token_accuracy": 0.12628084421157837,
"num_tokens": 6858264.0,
"step": 3120
},
{
"entropy": 5.948660182952881,
"epoch": 0.31014291385470427,
"grad_norm": 0.86328125,
"learning_rate": 0.0004994967618860685,
"loss": 5.7736,
"mean_token_accuracy": 0.13805429711937905,
"num_tokens": 6868861.0,
"step": 3125
},
{
"entropy": 5.981192207336425,
"epoch": 0.31063914251687175,
"grad_norm": 0.92578125,
"learning_rate": 0.0004994943906918333,
"loss": 5.7972,
"mean_token_accuracy": 0.13693608418107034,
"num_tokens": 6879797.0,
"step": 3130
},
{
"entropy": 5.884922504425049,
"epoch": 0.3111353711790393,
"grad_norm": 0.82421875,
"learning_rate": 0.0004994920139306211,
"loss": 5.7463,
"mean_token_accuracy": 0.13194068968296052,
"num_tokens": 6891323.0,
"step": 3135
},
{
"entropy": 5.909702634811401,
"epoch": 0.3116315998412068,
"grad_norm": 0.984375,
"learning_rate": 0.0004994896316024906,
"loss": 5.8044,
"mean_token_accuracy": 0.13013609647750854,
"num_tokens": 6903345.0,
"step": 3140
},
{
"entropy": 5.932398700714112,
"epoch": 0.31212782850337434,
"grad_norm": 0.9375,
"learning_rate": 0.000499487243707501,
"loss": 5.7547,
"mean_token_accuracy": 0.13403790220618247,
"num_tokens": 6914342.0,
"step": 3145
},
{
"entropy": 5.918168926239014,
"epoch": 0.31262405716554187,
"grad_norm": 0.79296875,
"learning_rate": 0.0004994848502457114,
"loss": 5.8166,
"mean_token_accuracy": 0.13116133585572243,
"num_tokens": 6926691.0,
"step": 3150
},
{
"entropy": 5.885048723220825,
"epoch": 0.3131202858277094,
"grad_norm": 0.86328125,
"learning_rate": 0.0004994824512171812,
"loss": 5.7052,
"mean_token_accuracy": 0.13244526460766792,
"num_tokens": 6937841.0,
"step": 3155
},
{
"entropy": 5.895932054519653,
"epoch": 0.31361651448987693,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994800466219701,
"loss": 5.781,
"mean_token_accuracy": 0.12934989854693413,
"num_tokens": 6948601.0,
"step": 3160
},
{
"entropy": 5.95720477104187,
"epoch": 0.31411274315204446,
"grad_norm": 0.89453125,
"learning_rate": 0.0004994776364601374,
"loss": 5.7171,
"mean_token_accuracy": 0.13609042316675185,
"num_tokens": 6959775.0,
"step": 3165
},
{
"entropy": 5.85569396018982,
"epoch": 0.314608971814212,
"grad_norm": 0.921875,
"learning_rate": 0.0004994752207317431,
"loss": 5.7471,
"mean_token_accuracy": 0.1365107588469982,
"num_tokens": 6969202.0,
"step": 3170
},
{
"entropy": 5.846676874160766,
"epoch": 0.3151052004763795,
"grad_norm": 0.8828125,
"learning_rate": 0.000499472799436847,
"loss": 5.7296,
"mean_token_accuracy": 0.136111731082201,
"num_tokens": 6978633.0,
"step": 3175
},
{
"entropy": 6.046626424789428,
"epoch": 0.31560142913854705,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994703725755092,
"loss": 5.7934,
"mean_token_accuracy": 0.13145529180765153,
"num_tokens": 6989638.0,
"step": 3180
},
{
"entropy": 5.903738021850586,
"epoch": 0.3160976578007146,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994679401477899,
"loss": 5.7616,
"mean_token_accuracy": 0.1309955298900604,
"num_tokens": 7001606.0,
"step": 3185
},
{
"entropy": 5.860150480270386,
"epoch": 0.3165938864628821,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994655021537492,
"loss": 5.7334,
"mean_token_accuracy": 0.13141413554549217,
"num_tokens": 7012915.0,
"step": 3190
},
{
"entropy": 5.89152774810791,
"epoch": 0.31709011512504964,
"grad_norm": 0.875,
"learning_rate": 0.0004994630585934478,
"loss": 5.7898,
"mean_token_accuracy": 0.12930853962898253,
"num_tokens": 7023191.0,
"step": 3195
},
{
"entropy": 5.914811658859253,
"epoch": 0.3175863437872172,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994606094669463,
"loss": 5.681,
"mean_token_accuracy": 0.13937111124396323,
"num_tokens": 7032506.0,
"step": 3200
},
{
"entropy": 5.89855432510376,
"epoch": 0.3180825724493847,
"grad_norm": 0.89453125,
"learning_rate": 0.0004994581547743053,
"loss": 5.814,
"mean_token_accuracy": 0.12698504626750945,
"num_tokens": 7046098.0,
"step": 3205
},
{
"entropy": 6.016089820861817,
"epoch": 0.3185788011115522,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994556945155857,
"loss": 5.8264,
"mean_token_accuracy": 0.13558279499411582,
"num_tokens": 7057434.0,
"step": 3210
},
{
"entropy": 5.82761664390564,
"epoch": 0.3190750297737197,
"grad_norm": 0.85546875,
"learning_rate": 0.0004994532286908486,
"loss": 5.7577,
"mean_token_accuracy": 0.12784940525889396,
"num_tokens": 7067674.0,
"step": 3215
},
{
"entropy": 5.943560075759888,
"epoch": 0.31957125843588724,
"grad_norm": 1.0625,
"learning_rate": 0.000499450757300155,
"loss": 5.8249,
"mean_token_accuracy": 0.13068325966596603,
"num_tokens": 7077898.0,
"step": 3220
},
{
"entropy": 5.973318004608155,
"epoch": 0.32006748709805477,
"grad_norm": 0.98046875,
"learning_rate": 0.0004994482803435664,
"loss": 5.7882,
"mean_token_accuracy": 0.12848865017294883,
"num_tokens": 7088566.0,
"step": 3225
},
{
"entropy": 5.90263147354126,
"epoch": 0.3205637157602223,
"grad_norm": 0.84765625,
"learning_rate": 0.000499445797821144,
"loss": 5.7808,
"mean_token_accuracy": 0.13472721502184867,
"num_tokens": 7099533.0,
"step": 3230
},
{
"entropy": 5.94389238357544,
"epoch": 0.32105994442238983,
"grad_norm": 0.87890625,
"learning_rate": 0.0004994433097329494,
"loss": 5.7632,
"mean_token_accuracy": 0.13370973244309425,
"num_tokens": 7112254.0,
"step": 3235
},
{
"entropy": 5.9783838272094725,
"epoch": 0.32155617308455736,
"grad_norm": 0.8515625,
"learning_rate": 0.0004994408160790444,
"loss": 5.7846,
"mean_token_accuracy": 0.13278804942965508,
"num_tokens": 7124704.0,
"step": 3240
},
{
"entropy": 6.045426225662231,
"epoch": 0.3220524017467249,
"grad_norm": 0.77734375,
"learning_rate": 0.0004994383168594907,
"loss": 5.9735,
"mean_token_accuracy": 0.11922514736652375,
"num_tokens": 7137741.0,
"step": 3245
},
{
"entropy": 5.94087586402893,
"epoch": 0.3225486304088924,
"grad_norm": 0.90625,
"learning_rate": 0.0004994358120743506,
"loss": 5.7579,
"mean_token_accuracy": 0.12984591871500015,
"num_tokens": 7147888.0,
"step": 3250
},
{
"entropy": 5.7819092750549315,
"epoch": 0.32304485907105995,
"grad_norm": 0.85546875,
"learning_rate": 0.0004994333017236858,
"loss": 5.7076,
"mean_token_accuracy": 0.14143425077199936,
"num_tokens": 7158950.0,
"step": 3255
},
{
"entropy": 5.94332447052002,
"epoch": 0.3235410877332275,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994307858075587,
"loss": 5.8421,
"mean_token_accuracy": 0.12570593357086182,
"num_tokens": 7170156.0,
"step": 3260
},
{
"entropy": 5.932934379577636,
"epoch": 0.324037316395395,
"grad_norm": 0.96484375,
"learning_rate": 0.0004994282643260319,
"loss": 5.7048,
"mean_token_accuracy": 0.13710035309195517,
"num_tokens": 7180576.0,
"step": 3265
},
{
"entropy": 5.811103487014771,
"epoch": 0.32453354505756254,
"grad_norm": 0.859375,
"learning_rate": 0.0004994257372791676,
"loss": 5.7565,
"mean_token_accuracy": 0.13245908543467522,
"num_tokens": 7192380.0,
"step": 3270
},
{
"entropy": 5.963694858551025,
"epoch": 0.3250297737197301,
"grad_norm": 0.78515625,
"learning_rate": 0.0004994232046670286,
"loss": 5.8183,
"mean_token_accuracy": 0.13459155336022377,
"num_tokens": 7203898.0,
"step": 3275
},
{
"entropy": 5.989812850952148,
"epoch": 0.3255260023818976,
"grad_norm": 0.875,
"learning_rate": 0.0004994206664896778,
"loss": 5.7841,
"mean_token_accuracy": 0.13584727346897124,
"num_tokens": 7215241.0,
"step": 3280
},
{
"entropy": 5.931274271011352,
"epoch": 0.3260222310440651,
"grad_norm": 0.875,
"learning_rate": 0.0004994181227471779,
"loss": 5.7383,
"mean_token_accuracy": 0.13464230597019194,
"num_tokens": 7226039.0,
"step": 3285
},
{
"entropy": 5.8311902523040775,
"epoch": 0.3265184597062326,
"grad_norm": 0.86328125,
"learning_rate": 0.0004994155734395923,
"loss": 5.7687,
"mean_token_accuracy": 0.13406949862837791,
"num_tokens": 7237148.0,
"step": 3290
},
{
"entropy": 5.923242807388306,
"epoch": 0.32701468836840014,
"grad_norm": 0.90625,
"learning_rate": 0.000499413018566984,
"loss": 5.7621,
"mean_token_accuracy": 0.1315522961318493,
"num_tokens": 7249284.0,
"step": 3295
},
{
"entropy": 5.875470447540283,
"epoch": 0.32751091703056767,
"grad_norm": 0.85546875,
"learning_rate": 0.0004994104581294163,
"loss": 5.7631,
"mean_token_accuracy": 0.12935446053743363,
"num_tokens": 7259967.0,
"step": 3300
},
{
"entropy": 5.827996301651001,
"epoch": 0.3280071456927352,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994078921269529,
"loss": 5.7235,
"mean_token_accuracy": 0.13562380596995355,
"num_tokens": 7271145.0,
"step": 3305
},
{
"entropy": 6.002976369857788,
"epoch": 0.32850337435490273,
"grad_norm": 0.86328125,
"learning_rate": 0.0004994053205596573,
"loss": 5.7351,
"mean_token_accuracy": 0.13544280007481574,
"num_tokens": 7281872.0,
"step": 3310
},
{
"entropy": 5.815413761138916,
"epoch": 0.32899960301707026,
"grad_norm": 0.8046875,
"learning_rate": 0.0004994027434275934,
"loss": 5.7639,
"mean_token_accuracy": 0.1359103187918663,
"num_tokens": 7293330.0,
"step": 3315
},
{
"entropy": 5.9990698337554935,
"epoch": 0.3294958316792378,
"grad_norm": 0.765625,
"learning_rate": 0.0004994001607308249,
"loss": 5.8214,
"mean_token_accuracy": 0.12711838632822037,
"num_tokens": 7304187.0,
"step": 3320
},
{
"entropy": 5.941404533386231,
"epoch": 0.3299920603414053,
"grad_norm": 0.8046875,
"learning_rate": 0.000499397572469416,
"loss": 5.7195,
"mean_token_accuracy": 0.13139874935150148,
"num_tokens": 7314756.0,
"step": 3325
},
{
"entropy": 5.826371097564698,
"epoch": 0.33048828900357285,
"grad_norm": 0.78515625,
"learning_rate": 0.0004993949786434308,
"loss": 5.7561,
"mean_token_accuracy": 0.13403744399547576,
"num_tokens": 7325711.0,
"step": 3330
},
{
"entropy": 5.872751712799072,
"epoch": 0.3309845176657404,
"grad_norm": 0.9765625,
"learning_rate": 0.0004993923792529337,
"loss": 5.6831,
"mean_token_accuracy": 0.1406965471804142,
"num_tokens": 7335639.0,
"step": 3335
},
{
"entropy": 5.956606531143189,
"epoch": 0.3314807463279079,
"grad_norm": 0.84375,
"learning_rate": 0.0004993897742979892,
"loss": 5.8027,
"mean_token_accuracy": 0.13377716019749641,
"num_tokens": 7346621.0,
"step": 3340
},
{
"entropy": 5.9286905288696286,
"epoch": 0.33197697499007545,
"grad_norm": 0.84765625,
"learning_rate": 0.0004993871637786616,
"loss": 5.7705,
"mean_token_accuracy": 0.13041678220033645,
"num_tokens": 7358950.0,
"step": 3345
},
{
"entropy": 5.821332550048828,
"epoch": 0.332473203652243,
"grad_norm": 0.8671875,
"learning_rate": 0.000499384547695016,
"loss": 5.626,
"mean_token_accuracy": 0.13846936598420143,
"num_tokens": 7369629.0,
"step": 3350
},
{
"entropy": 5.872072839736939,
"epoch": 0.3329694323144105,
"grad_norm": 0.9296875,
"learning_rate": 0.000499381926047117,
"loss": 5.7262,
"mean_token_accuracy": 0.13408969193696976,
"num_tokens": 7379853.0,
"step": 3355
},
{
"entropy": 5.865009212493897,
"epoch": 0.333465660976578,
"grad_norm": 0.90625,
"learning_rate": 0.00049937929883503,
"loss": 5.7563,
"mean_token_accuracy": 0.13046114146709442,
"num_tokens": 7391323.0,
"step": 3360
},
{
"entropy": 5.887697219848633,
"epoch": 0.3339618896387455,
"grad_norm": 1.078125,
"learning_rate": 0.0004993766660588196,
"loss": 5.7304,
"mean_token_accuracy": 0.13350997790694236,
"num_tokens": 7402201.0,
"step": 3365
},
{
"entropy": 6.01474552154541,
"epoch": 0.33445811830091304,
"grad_norm": 0.84765625,
"learning_rate": 0.0004993740277185515,
"loss": 5.9623,
"mean_token_accuracy": 0.12484050020575524,
"num_tokens": 7413783.0,
"step": 3370
},
{
"entropy": 5.840613555908203,
"epoch": 0.3349543469630806,
"grad_norm": 0.859375,
"learning_rate": 0.000499371383814291,
"loss": 5.7223,
"mean_token_accuracy": 0.13400618061423303,
"num_tokens": 7425909.0,
"step": 3375
},
{
"entropy": 5.847539567947388,
"epoch": 0.3354505756252481,
"grad_norm": 0.8671875,
"learning_rate": 0.0004993687343461036,
"loss": 5.6318,
"mean_token_accuracy": 0.13809972107410431,
"num_tokens": 7436346.0,
"step": 3380
},
{
"entropy": 5.904675722122192,
"epoch": 0.33594680428741563,
"grad_norm": 0.90234375,
"learning_rate": 0.0004993660793140551,
"loss": 5.6827,
"mean_token_accuracy": 0.14161865711212157,
"num_tokens": 7446596.0,
"step": 3385
},
{
"entropy": 5.782830667495728,
"epoch": 0.33644303294958317,
"grad_norm": 0.96875,
"learning_rate": 0.0004993634187182113,
"loss": 5.7883,
"mean_token_accuracy": 0.12615693137049674,
"num_tokens": 7457697.0,
"step": 3390
},
{
"entropy": 6.065376043319702,
"epoch": 0.3369392616117507,
"grad_norm": 0.83984375,
"learning_rate": 0.0004993607525586381,
"loss": 5.8802,
"mean_token_accuracy": 0.131008081138134,
"num_tokens": 7468541.0,
"step": 3395
},
{
"entropy": 5.9343438148498535,
"epoch": 0.3374354902739182,
"grad_norm": 0.8046875,
"learning_rate": 0.0004993580808354018,
"loss": 5.7764,
"mean_token_accuracy": 0.13551992177963257,
"num_tokens": 7479781.0,
"step": 3400
},
{
"entropy": 5.837701988220215,
"epoch": 0.33793171893608576,
"grad_norm": 0.86328125,
"learning_rate": 0.0004993554035485685,
"loss": 5.6816,
"mean_token_accuracy": 0.1342918574810028,
"num_tokens": 7490722.0,
"step": 3405
},
{
"entropy": 5.948045158386231,
"epoch": 0.3384279475982533,
"grad_norm": 0.92578125,
"learning_rate": 0.0004993527206982047,
"loss": 5.7867,
"mean_token_accuracy": 0.13522917330265044,
"num_tokens": 7502293.0,
"step": 3410
},
{
"entropy": 5.918159008026123,
"epoch": 0.3389241762604208,
"grad_norm": 0.75390625,
"learning_rate": 0.0004993500322843767,
"loss": 5.7208,
"mean_token_accuracy": 0.13539437279105188,
"num_tokens": 7513846.0,
"step": 3415
},
{
"entropy": 5.830818223953247,
"epoch": 0.33942040492258835,
"grad_norm": 0.8671875,
"learning_rate": 0.0004993473383071514,
"loss": 5.7466,
"mean_token_accuracy": 0.13467329666018485,
"num_tokens": 7524062.0,
"step": 3420
},
{
"entropy": 5.901687240600586,
"epoch": 0.3399166335847559,
"grad_norm": 0.83984375,
"learning_rate": 0.0004993446387665956,
"loss": 5.7989,
"mean_token_accuracy": 0.13709105178713799,
"num_tokens": 7535533.0,
"step": 3425
},
{
"entropy": 5.914133930206299,
"epoch": 0.3404128622469234,
"grad_norm": 0.87109375,
"learning_rate": 0.0004993419336627761,
"loss": 5.6902,
"mean_token_accuracy": 0.13548238351941108,
"num_tokens": 7545951.0,
"step": 3430
},
{
"entropy": 5.817704439163208,
"epoch": 0.3409090909090909,
"grad_norm": 0.90234375,
"learning_rate": 0.0004993392229957602,
"loss": 5.6334,
"mean_token_accuracy": 0.1382601298391819,
"num_tokens": 7557320.0,
"step": 3435
},
{
"entropy": 5.805458402633667,
"epoch": 0.3414053195712584,
"grad_norm": 0.8359375,
"learning_rate": 0.0004993365067656148,
"loss": 5.7513,
"mean_token_accuracy": 0.1329974986612797,
"num_tokens": 7568078.0,
"step": 3440
},
{
"entropy": 5.883556461334228,
"epoch": 0.34190154823342594,
"grad_norm": 0.83203125,
"learning_rate": 0.0004993337849724076,
"loss": 5.7412,
"mean_token_accuracy": 0.13214808478951454,
"num_tokens": 7578940.0,
"step": 3445
},
{
"entropy": 5.893185567855835,
"epoch": 0.3423977768955935,
"grad_norm": 0.99609375,
"learning_rate": 0.0004993310576162059,
"loss": 5.7479,
"mean_token_accuracy": 0.13639405518770217,
"num_tokens": 7589864.0,
"step": 3450
},
{
"entropy": 5.922738981246948,
"epoch": 0.342894005557761,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993283246970772,
"loss": 5.7057,
"mean_token_accuracy": 0.13013398125767708,
"num_tokens": 7599795.0,
"step": 3455
},
{
"entropy": 5.854125308990478,
"epoch": 0.34339023421992854,
"grad_norm": 0.8828125,
"learning_rate": 0.0004993255862150896,
"loss": 5.6876,
"mean_token_accuracy": 0.13669478967785836,
"num_tokens": 7610056.0,
"step": 3460
},
{
"entropy": 5.802547788619995,
"epoch": 0.34388646288209607,
"grad_norm": 0.875,
"learning_rate": 0.0004993228421703108,
"loss": 5.6874,
"mean_token_accuracy": 0.13286467269062996,
"num_tokens": 7620218.0,
"step": 3465
},
{
"entropy": 5.884554147720337,
"epoch": 0.3443826915442636,
"grad_norm": 0.796875,
"learning_rate": 0.0004993200925628088,
"loss": 5.7779,
"mean_token_accuracy": 0.1346588559448719,
"num_tokens": 7632781.0,
"step": 3470
},
{
"entropy": 5.8774271488189695,
"epoch": 0.3448789202064311,
"grad_norm": 0.85546875,
"learning_rate": 0.000499317337392652,
"loss": 5.7513,
"mean_token_accuracy": 0.13203491717576982,
"num_tokens": 7643845.0,
"step": 3475
},
{
"entropy": 5.910117483139038,
"epoch": 0.34537514886859866,
"grad_norm": 0.8515625,
"learning_rate": 0.0004993145766599085,
"loss": 5.7793,
"mean_token_accuracy": 0.13024887666106225,
"num_tokens": 7654583.0,
"step": 3480
},
{
"entropy": 5.8562994480133055,
"epoch": 0.3458713775307662,
"grad_norm": 0.89453125,
"learning_rate": 0.0004993118103646469,
"loss": 5.6299,
"mean_token_accuracy": 0.14034538567066193,
"num_tokens": 7665714.0,
"step": 3485
},
{
"entropy": 5.8359884262084964,
"epoch": 0.3463676061929337,
"grad_norm": 0.86328125,
"learning_rate": 0.0004993090385069357,
"loss": 5.6924,
"mean_token_accuracy": 0.1398855820298195,
"num_tokens": 7676011.0,
"step": 3490
},
{
"entropy": 5.825574779510498,
"epoch": 0.34686383485510125,
"grad_norm": 0.84765625,
"learning_rate": 0.0004993062610868437,
"loss": 5.7376,
"mean_token_accuracy": 0.13499814942479132,
"num_tokens": 7686927.0,
"step": 3495
},
{
"entropy": 5.846639966964721,
"epoch": 0.3473600635172688,
"grad_norm": 0.82421875,
"learning_rate": 0.0004993034781044398,
"loss": 5.6989,
"mean_token_accuracy": 0.135499270260334,
"num_tokens": 7698419.0,
"step": 3500
},
{
"entropy": 5.902319765090942,
"epoch": 0.3478562921794363,
"grad_norm": 0.859375,
"learning_rate": 0.000499300689559793,
"loss": 5.6731,
"mean_token_accuracy": 0.13935840278863906,
"num_tokens": 7709695.0,
"step": 3505
},
{
"entropy": 5.833289051055909,
"epoch": 0.3483525208416038,
"grad_norm": 1.078125,
"learning_rate": 0.0004992978954529724,
"loss": 5.6456,
"mean_token_accuracy": 0.1389060892164707,
"num_tokens": 7721309.0,
"step": 3510
},
{
"entropy": 5.848011827468872,
"epoch": 0.3488487495037713,
"grad_norm": 0.8515625,
"learning_rate": 0.0004992950957840473,
"loss": 5.7182,
"mean_token_accuracy": 0.13931151628494262,
"num_tokens": 7732074.0,
"step": 3515
},
{
"entropy": 5.895964670181274,
"epoch": 0.34934497816593885,
"grad_norm": 0.95703125,
"learning_rate": 0.000499292290553087,
"loss": 5.7953,
"mean_token_accuracy": 0.13158057406544685,
"num_tokens": 7742869.0,
"step": 3520
},
{
"entropy": 5.8405561447143555,
"epoch": 0.3498412068281064,
"grad_norm": 0.8515625,
"learning_rate": 0.0004992894797601614,
"loss": 5.6901,
"mean_token_accuracy": 0.13555181324481963,
"num_tokens": 7754895.0,
"step": 3525
},
{
"entropy": 5.8825901508331295,
"epoch": 0.3503374354902739,
"grad_norm": 0.90625,
"learning_rate": 0.00049928666340534,
"loss": 5.6677,
"mean_token_accuracy": 0.13643246665596961,
"num_tokens": 7765262.0,
"step": 3530
},
{
"entropy": 5.8095235347747805,
"epoch": 0.35083366415244144,
"grad_norm": 0.87890625,
"learning_rate": 0.0004992838414886925,
"loss": 5.739,
"mean_token_accuracy": 0.1327873632311821,
"num_tokens": 7777348.0,
"step": 3535
},
{
"entropy": 5.916355848312378,
"epoch": 0.35132989281460897,
"grad_norm": 0.84765625,
"learning_rate": 0.0004992810140102891,
"loss": 5.7297,
"mean_token_accuracy": 0.13413830399513244,
"num_tokens": 7788161.0,
"step": 3540
},
{
"entropy": 5.840317058563232,
"epoch": 0.3518261214767765,
"grad_norm": 0.86328125,
"learning_rate": 0.0004992781809701999,
"loss": 5.7161,
"mean_token_accuracy": 0.13731137067079544,
"num_tokens": 7798629.0,
"step": 3545
},
{
"entropy": 5.829579782485962,
"epoch": 0.35232235013894403,
"grad_norm": 0.83984375,
"learning_rate": 0.0004992753423684951,
"loss": 5.7198,
"mean_token_accuracy": 0.13580114766955376,
"num_tokens": 7810148.0,
"step": 3550
},
{
"entropy": 5.863365554809571,
"epoch": 0.35281857880111156,
"grad_norm": 0.921875,
"learning_rate": 0.0004992724982052449,
"loss": 5.71,
"mean_token_accuracy": 0.13254974409937859,
"num_tokens": 7821667.0,
"step": 3555
},
{
"entropy": 5.798212766647339,
"epoch": 0.3533148074632791,
"grad_norm": 0.7890625,
"learning_rate": 0.00049926964848052,
"loss": 5.6957,
"mean_token_accuracy": 0.13729101940989494,
"num_tokens": 7832184.0,
"step": 3560
},
{
"entropy": 5.927416610717773,
"epoch": 0.3538110361254466,
"grad_norm": 0.875,
"learning_rate": 0.0004992667931943913,
"loss": 5.7183,
"mean_token_accuracy": 0.13577507585287094,
"num_tokens": 7842983.0,
"step": 3565
},
{
"entropy": 5.856418895721435,
"epoch": 0.35430726478761415,
"grad_norm": 0.7734375,
"learning_rate": 0.0004992639323469293,
"loss": 5.6354,
"mean_token_accuracy": 0.14334375709295272,
"num_tokens": 7854783.0,
"step": 3570
},
{
"entropy": 5.773731470108032,
"epoch": 0.3548034934497817,
"grad_norm": 0.91015625,
"learning_rate": 0.000499261065938205,
"loss": 5.6583,
"mean_token_accuracy": 0.1432967111468315,
"num_tokens": 7864684.0,
"step": 3575
},
{
"entropy": 5.895873069763184,
"epoch": 0.3552997221119492,
"grad_norm": 0.9296875,
"learning_rate": 0.0004992581939682895,
"loss": 5.8056,
"mean_token_accuracy": 0.1261812210083008,
"num_tokens": 7876846.0,
"step": 3580
},
{
"entropy": 5.935712575912476,
"epoch": 0.3557959507741167,
"grad_norm": 0.890625,
"learning_rate": 0.000499255316437254,
"loss": 5.6779,
"mean_token_accuracy": 0.1337505742907524,
"num_tokens": 7886826.0,
"step": 3585
},
{
"entropy": 5.801439237594605,
"epoch": 0.3562921794362842,
"grad_norm": 0.875,
"learning_rate": 0.0004992524333451699,
"loss": 5.7537,
"mean_token_accuracy": 0.13139742091298104,
"num_tokens": 7898941.0,
"step": 3590
},
{
"entropy": 5.849826717376709,
"epoch": 0.35678840809845175,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992495446921087,
"loss": 5.6674,
"mean_token_accuracy": 0.13365738093852997,
"num_tokens": 7909916.0,
"step": 3595
},
{
"entropy": 5.808553171157837,
"epoch": 0.3572846367606193,
"grad_norm": 0.78125,
"learning_rate": 0.0004992466504781419,
"loss": 5.6792,
"mean_token_accuracy": 0.13627909868955612,
"num_tokens": 7922137.0,
"step": 3600
},
{
"entropy": 5.79820442199707,
"epoch": 0.3577808654227868,
"grad_norm": 0.91015625,
"learning_rate": 0.0004992437507033415,
"loss": 5.616,
"mean_token_accuracy": 0.13959808573126792,
"num_tokens": 7932940.0,
"step": 3605
},
{
"entropy": 5.881356430053711,
"epoch": 0.35827709408495434,
"grad_norm": 0.80078125,
"learning_rate": 0.0004992408453677792,
"loss": 5.7851,
"mean_token_accuracy": 0.12732164859771727,
"num_tokens": 7943735.0,
"step": 3610
},
{
"entropy": 5.798199939727783,
"epoch": 0.35877332274712187,
"grad_norm": 0.96484375,
"learning_rate": 0.0004992379344715273,
"loss": 5.672,
"mean_token_accuracy": 0.13556718379259108,
"num_tokens": 7955323.0,
"step": 3615
},
{
"entropy": 5.827555418014526,
"epoch": 0.3592695514092894,
"grad_norm": 0.94140625,
"learning_rate": 0.0004992350180146577,
"loss": 5.6371,
"mean_token_accuracy": 0.13796332776546477,
"num_tokens": 7966073.0,
"step": 3620
},
{
"entropy": 5.883342456817627,
"epoch": 0.35976578007145693,
"grad_norm": 0.97265625,
"learning_rate": 0.0004992320959972428,
"loss": 5.7951,
"mean_token_accuracy": 0.13291794434189796,
"num_tokens": 7977027.0,
"step": 3625
},
{
"entropy": 5.816143798828125,
"epoch": 0.36026200873362446,
"grad_norm": 0.859375,
"learning_rate": 0.0004992291684193551,
"loss": 5.6621,
"mean_token_accuracy": 0.13837531432509423,
"num_tokens": 7987368.0,
"step": 3630
},
{
"entropy": 5.862054634094238,
"epoch": 0.360758237395792,
"grad_norm": 0.78125,
"learning_rate": 0.0004992262352810672,
"loss": 5.7323,
"mean_token_accuracy": 0.13465685471892358,
"num_tokens": 8000060.0,
"step": 3635
},
{
"entropy": 5.904985523223877,
"epoch": 0.3612544660579595,
"grad_norm": 0.875,
"learning_rate": 0.0004992232965824518,
"loss": 5.779,
"mean_token_accuracy": 0.13045103549957277,
"num_tokens": 8011388.0,
"step": 3640
},
{
"entropy": 5.78095817565918,
"epoch": 0.36175069472012705,
"grad_norm": 0.87890625,
"learning_rate": 0.000499220352323582,
"loss": 5.6446,
"mean_token_accuracy": 0.13995237201452254,
"num_tokens": 8021682.0,
"step": 3645
},
{
"entropy": 5.828157138824463,
"epoch": 0.3622469233822946,
"grad_norm": 0.88671875,
"learning_rate": 0.0004992174025045304,
"loss": 5.6637,
"mean_token_accuracy": 0.1378515474498272,
"num_tokens": 8032799.0,
"step": 3650
},
{
"entropy": 5.791543912887573,
"epoch": 0.3627431520444621,
"grad_norm": 0.88671875,
"learning_rate": 0.0004992144471253704,
"loss": 5.5542,
"mean_token_accuracy": 0.14514420479536055,
"num_tokens": 8042742.0,
"step": 3655
},
{
"entropy": 5.836719179153443,
"epoch": 0.3632393807066296,
"grad_norm": 0.79296875,
"learning_rate": 0.0004992114861861753,
"loss": 5.726,
"mean_token_accuracy": 0.13775497898459435,
"num_tokens": 8054068.0,
"step": 3660
},
{
"entropy": 5.895388793945313,
"epoch": 0.3637356093687971,
"grad_norm": 0.79296875,
"learning_rate": 0.0004992085196870184,
"loss": 5.7041,
"mean_token_accuracy": 0.13520997688174247,
"num_tokens": 8065286.0,
"step": 3665
},
{
"entropy": 5.8350739002227785,
"epoch": 0.36423183803096465,
"grad_norm": 0.8671875,
"learning_rate": 0.0004992055476279734,
"loss": 5.7644,
"mean_token_accuracy": 0.12864286303520203,
"num_tokens": 8077122.0,
"step": 3670
},
{
"entropy": 5.914077806472778,
"epoch": 0.3647280666931322,
"grad_norm": 0.875,
"learning_rate": 0.0004992025700091139,
"loss": 5.7053,
"mean_token_accuracy": 0.1394345611333847,
"num_tokens": 8087036.0,
"step": 3675
},
{
"entropy": 5.887608432769776,
"epoch": 0.3652242953552997,
"grad_norm": 0.796875,
"learning_rate": 0.0004991995868305138,
"loss": 5.7439,
"mean_token_accuracy": 0.13455166965723037,
"num_tokens": 8098517.0,
"step": 3680
},
{
"entropy": 5.778636121749878,
"epoch": 0.36572052401746724,
"grad_norm": 0.8984375,
"learning_rate": 0.0004991965980922471,
"loss": 5.6182,
"mean_token_accuracy": 0.13923544213175773,
"num_tokens": 8107475.0,
"step": 3685
},
{
"entropy": 5.82491021156311,
"epoch": 0.36621675267963477,
"grad_norm": 0.80859375,
"learning_rate": 0.0004991936037943878,
"loss": 5.7266,
"mean_token_accuracy": 0.1303709253668785,
"num_tokens": 8118529.0,
"step": 3690
},
{
"entropy": 5.9018360614776615,
"epoch": 0.3667129813418023,
"grad_norm": 0.890625,
"learning_rate": 0.0004991906039370104,
"loss": 5.6292,
"mean_token_accuracy": 0.14039468988776208,
"num_tokens": 8128951.0,
"step": 3695
},
{
"entropy": 5.746214437484741,
"epoch": 0.36720921000396983,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991875985201889,
"loss": 5.6492,
"mean_token_accuracy": 0.1410698860883713,
"num_tokens": 8139385.0,
"step": 3700
},
{
"entropy": 5.918550968170166,
"epoch": 0.36770543866613736,
"grad_norm": 0.859375,
"learning_rate": 0.000499184587543998,
"loss": 5.6843,
"mean_token_accuracy": 0.1389869749546051,
"num_tokens": 8149376.0,
"step": 3705
},
{
"entropy": 5.8605245590209964,
"epoch": 0.3682016673283049,
"grad_norm": 0.81640625,
"learning_rate": 0.0004991815710085126,
"loss": 5.7092,
"mean_token_accuracy": 0.13653566539287568,
"num_tokens": 8160687.0,
"step": 3710
},
{
"entropy": 5.872888708114624,
"epoch": 0.3686978959904724,
"grad_norm": 0.81640625,
"learning_rate": 0.0004991785489138072,
"loss": 5.7141,
"mean_token_accuracy": 0.1377060107886791,
"num_tokens": 8171762.0,
"step": 3715
},
{
"entropy": 5.724028301239014,
"epoch": 0.36919412465263995,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991755212599569,
"loss": 5.4791,
"mean_token_accuracy": 0.14507829397916794,
"num_tokens": 8181643.0,
"step": 3720
},
{
"entropy": 5.697611093521118,
"epoch": 0.3696903533148075,
"grad_norm": 0.86328125,
"learning_rate": 0.0004991724880470366,
"loss": 5.7306,
"mean_token_accuracy": 0.13918372690677644,
"num_tokens": 8192692.0,
"step": 3725
},
{
"entropy": 5.91280164718628,
"epoch": 0.370186581976975,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991694492751217,
"loss": 5.678,
"mean_token_accuracy": 0.13002298176288604,
"num_tokens": 8202941.0,
"step": 3730
},
{
"entropy": 5.799401473999024,
"epoch": 0.3706828106391425,
"grad_norm": 0.83203125,
"learning_rate": 0.0004991664049442876,
"loss": 5.6749,
"mean_token_accuracy": 0.13614187836647035,
"num_tokens": 8214206.0,
"step": 3735
},
{
"entropy": 5.792055654525757,
"epoch": 0.37117903930131,
"grad_norm": 0.98046875,
"learning_rate": 0.0004991633550546097,
"loss": 5.5088,
"mean_token_accuracy": 0.15157946944236755,
"num_tokens": 8223672.0,
"step": 3740
},
{
"entropy": 5.7840086936950685,
"epoch": 0.37167526796347755,
"grad_norm": 0.8984375,
"learning_rate": 0.0004991602996061635,
"loss": 5.6511,
"mean_token_accuracy": 0.14170872122049333,
"num_tokens": 8234189.0,
"step": 3745
},
{
"entropy": 5.839711046218872,
"epoch": 0.3721714966256451,
"grad_norm": 0.8515625,
"learning_rate": 0.0004991572385990249,
"loss": 5.6868,
"mean_token_accuracy": 0.139129176735878,
"num_tokens": 8245382.0,
"step": 3750
},
{
"entropy": 5.895852756500244,
"epoch": 0.3726677252878126,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991541720332698,
"loss": 5.6406,
"mean_token_accuracy": 0.13922024741768838,
"num_tokens": 8254530.0,
"step": 3755
},
{
"entropy": 5.741476011276245,
"epoch": 0.37316395394998014,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991510999089743,
"loss": 5.6856,
"mean_token_accuracy": 0.13898830264806747,
"num_tokens": 8264030.0,
"step": 3760
},
{
"entropy": 5.792221021652222,
"epoch": 0.3736601826121477,
"grad_norm": 0.87109375,
"learning_rate": 0.0004991480222262145,
"loss": 5.6503,
"mean_token_accuracy": 0.14368472024798393,
"num_tokens": 8273537.0,
"step": 3765
},
{
"entropy": 5.88636155128479,
"epoch": 0.3741564112743152,
"grad_norm": 0.70703125,
"learning_rate": 0.0004991449389850668,
"loss": 5.6834,
"mean_token_accuracy": 0.13585606068372727,
"num_tokens": 8286123.0,
"step": 3770
},
{
"entropy": 5.796380996704102,
"epoch": 0.37465263993648273,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991418501856076,
"loss": 5.7091,
"mean_token_accuracy": 0.13685489520430566,
"num_tokens": 8298115.0,
"step": 3775
},
{
"entropy": 5.87733302116394,
"epoch": 0.37514886859865026,
"grad_norm": 0.94140625,
"learning_rate": 0.0004991387558279134,
"loss": 5.7145,
"mean_token_accuracy": 0.1364159256219864,
"num_tokens": 8309376.0,
"step": 3780
},
{
"entropy": 5.805009841918945,
"epoch": 0.3756450972608178,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991356559120612,
"loss": 5.6778,
"mean_token_accuracy": 0.13979559317231177,
"num_tokens": 8319272.0,
"step": 3785
},
{
"entropy": 5.818359422683716,
"epoch": 0.3761413259229853,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991325504381275,
"loss": 5.6691,
"mean_token_accuracy": 0.13450096398591996,
"num_tokens": 8330042.0,
"step": 3790
},
{
"entropy": 5.800448036193847,
"epoch": 0.37663755458515286,
"grad_norm": 0.80859375,
"learning_rate": 0.0004991294394061896,
"loss": 5.622,
"mean_token_accuracy": 0.1370707221329212,
"num_tokens": 8340394.0,
"step": 3795
},
{
"entropy": 5.775977277755738,
"epoch": 0.3771337832473204,
"grad_norm": 0.80078125,
"learning_rate": 0.0004991263228163246,
"loss": 5.6841,
"mean_token_accuracy": 0.1365865007042885,
"num_tokens": 8351103.0,
"step": 3800
},
{
"entropy": 5.865830612182617,
"epoch": 0.3776300119094879,
"grad_norm": 0.78515625,
"learning_rate": 0.0004991232006686096,
"loss": 5.6929,
"mean_token_accuracy": 0.13774350062012672,
"num_tokens": 8363215.0,
"step": 3805
},
{
"entropy": 5.919228649139404,
"epoch": 0.3781262405716554,
"grad_norm": 0.828125,
"learning_rate": 0.0004991200729631223,
"loss": 5.7373,
"mean_token_accuracy": 0.12794058844447137,
"num_tokens": 8374602.0,
"step": 3810
},
{
"entropy": 5.787239599227905,
"epoch": 0.3786224692338229,
"grad_norm": 0.859375,
"learning_rate": 0.0004991169396999401,
"loss": 5.6555,
"mean_token_accuracy": 0.13108878657221795,
"num_tokens": 8386470.0,
"step": 3815
},
{
"entropy": 5.842096328735352,
"epoch": 0.37911869789599045,
"grad_norm": 0.97265625,
"learning_rate": 0.0004991138008791408,
"loss": 5.6799,
"mean_token_accuracy": 0.13508746549487113,
"num_tokens": 8396284.0,
"step": 3820
},
{
"entropy": 5.8682530403137205,
"epoch": 0.379614926558158,
"grad_norm": 0.82421875,
"learning_rate": 0.0004991106565008021,
"loss": 5.7021,
"mean_token_accuracy": 0.13416668847203256,
"num_tokens": 8407856.0,
"step": 3825
},
{
"entropy": 5.822223329544068,
"epoch": 0.3801111552203255,
"grad_norm": 0.83203125,
"learning_rate": 0.0004991075065650019,
"loss": 5.6105,
"mean_token_accuracy": 0.1422433525323868,
"num_tokens": 8419560.0,
"step": 3830
},
{
"entropy": 5.871718502044677,
"epoch": 0.38060738388249304,
"grad_norm": 0.875,
"learning_rate": 0.0004991043510718186,
"loss": 5.6177,
"mean_token_accuracy": 0.1404113858938217,
"num_tokens": 8430594.0,
"step": 3835
},
{
"entropy": 5.7367877006530765,
"epoch": 0.3811036125446606,
"grad_norm": 0.86328125,
"learning_rate": 0.0004991011900213303,
"loss": 5.588,
"mean_token_accuracy": 0.1445833697915077,
"num_tokens": 8440549.0,
"step": 3840
},
{
"entropy": 5.7166365623474125,
"epoch": 0.3815998412068281,
"grad_norm": 0.84375,
"learning_rate": 0.0004990980234136152,
"loss": 5.6232,
"mean_token_accuracy": 0.13667350262403488,
"num_tokens": 8451553.0,
"step": 3845
},
{
"entropy": 5.814205455780029,
"epoch": 0.38209606986899564,
"grad_norm": 0.84375,
"learning_rate": 0.0004990948512487522,
"loss": 5.6252,
"mean_token_accuracy": 0.1421022191643715,
"num_tokens": 8462899.0,
"step": 3850
},
{
"entropy": 5.8078104019165036,
"epoch": 0.38259229853116317,
"grad_norm": 0.93359375,
"learning_rate": 0.0004990916735268198,
"loss": 5.6796,
"mean_token_accuracy": 0.13645819798111916,
"num_tokens": 8474566.0,
"step": 3855
},
{
"entropy": 5.869540357589722,
"epoch": 0.3830885271933307,
"grad_norm": 0.921875,
"learning_rate": 0.0004990884902478966,
"loss": 5.6865,
"mean_token_accuracy": 0.14038145467638968,
"num_tokens": 8484786.0,
"step": 3860
},
{
"entropy": 5.750705623626709,
"epoch": 0.3835847558554982,
"grad_norm": 0.83984375,
"learning_rate": 0.0004990853014120619,
"loss": 5.625,
"mean_token_accuracy": 0.14226635843515395,
"num_tokens": 8496416.0,
"step": 3865
},
{
"entropy": 5.798192834854126,
"epoch": 0.38408098451766576,
"grad_norm": 0.85546875,
"learning_rate": 0.0004990821070193944,
"loss": 5.6561,
"mean_token_accuracy": 0.136373158544302,
"num_tokens": 8507287.0,
"step": 3870
},
{
"entropy": 5.865294790267944,
"epoch": 0.3845772131798333,
"grad_norm": 0.82421875,
"learning_rate": 0.0004990789070699736,
"loss": 5.6948,
"mean_token_accuracy": 0.13455733358860017,
"num_tokens": 8517922.0,
"step": 3875
},
{
"entropy": 5.765826034545898,
"epoch": 0.3850734418420008,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990757015638787,
"loss": 5.631,
"mean_token_accuracy": 0.14302095845341684,
"num_tokens": 8529532.0,
"step": 3880
},
{
"entropy": 5.816843843460083,
"epoch": 0.3855696705041683,
"grad_norm": 0.8828125,
"learning_rate": 0.0004990724905011894,
"loss": 5.6033,
"mean_token_accuracy": 0.14065930545330046,
"num_tokens": 8540207.0,
"step": 3885
},
{
"entropy": 5.801877880096436,
"epoch": 0.3860658991663358,
"grad_norm": 0.9296875,
"learning_rate": 0.000499069273881985,
"loss": 5.6383,
"mean_token_accuracy": 0.13733408302068711,
"num_tokens": 8551634.0,
"step": 3890
},
{
"entropy": 5.8024838924407955,
"epoch": 0.38656212782850335,
"grad_norm": 0.80859375,
"learning_rate": 0.0004990660517063455,
"loss": 5.7033,
"mean_token_accuracy": 0.1306481033563614,
"num_tokens": 8563256.0,
"step": 3895
},
{
"entropy": 5.81333646774292,
"epoch": 0.3870583564906709,
"grad_norm": 0.875,
"learning_rate": 0.0004990628239743508,
"loss": 5.7163,
"mean_token_accuracy": 0.13748448565602303,
"num_tokens": 8573833.0,
"step": 3900
},
{
"entropy": 5.79175386428833,
"epoch": 0.3875545851528384,
"grad_norm": 0.89453125,
"learning_rate": 0.0004990595906860807,
"loss": 5.6222,
"mean_token_accuracy": 0.1423626087605953,
"num_tokens": 8584283.0,
"step": 3905
},
{
"entropy": 5.789517545700074,
"epoch": 0.38805081381500595,
"grad_norm": 0.90234375,
"learning_rate": 0.0004990563518416157,
"loss": 5.6789,
"mean_token_accuracy": 0.1358125925064087,
"num_tokens": 8595344.0,
"step": 3910
},
{
"entropy": 5.814813470840454,
"epoch": 0.3885470424771735,
"grad_norm": 0.94140625,
"learning_rate": 0.0004990531074410358,
"loss": 5.6036,
"mean_token_accuracy": 0.137834145873785,
"num_tokens": 8605625.0,
"step": 3915
},
{
"entropy": 5.7738450050354,
"epoch": 0.389043271139341,
"grad_norm": 0.84765625,
"learning_rate": 0.0004990498574844218,
"loss": 5.5894,
"mean_token_accuracy": 0.14018410369753836,
"num_tokens": 8616259.0,
"step": 3920
},
{
"entropy": 5.82464485168457,
"epoch": 0.38953949980150854,
"grad_norm": 0.90234375,
"learning_rate": 0.0004990466019718542,
"loss": 5.6034,
"mean_token_accuracy": 0.1412269026041031,
"num_tokens": 8627179.0,
"step": 3925
},
{
"entropy": 5.831127595901489,
"epoch": 0.39003572846367607,
"grad_norm": 0.84765625,
"learning_rate": 0.0004990433409034134,
"loss": 5.71,
"mean_token_accuracy": 0.14471403062343596,
"num_tokens": 8637448.0,
"step": 3930
},
{
"entropy": 5.725909614562989,
"epoch": 0.3905319571258436,
"grad_norm": 0.921875,
"learning_rate": 0.0004990400742791806,
"loss": 5.6293,
"mean_token_accuracy": 0.1356988586485386,
"num_tokens": 8649168.0,
"step": 3935
},
{
"entropy": 5.88536319732666,
"epoch": 0.39102818578801113,
"grad_norm": 0.796875,
"learning_rate": 0.0004990368020992368,
"loss": 5.7237,
"mean_token_accuracy": 0.13518148735165597,
"num_tokens": 8660653.0,
"step": 3940
},
{
"entropy": 5.793624401092529,
"epoch": 0.39152441445017866,
"grad_norm": 0.84765625,
"learning_rate": 0.000499033524363663,
"loss": 5.6971,
"mean_token_accuracy": 0.1437949612736702,
"num_tokens": 8672000.0,
"step": 3945
},
{
"entropy": 5.733256244659424,
"epoch": 0.3920206431123462,
"grad_norm": 0.890625,
"learning_rate": 0.0004990302410725405,
"loss": 5.6268,
"mean_token_accuracy": 0.14762071818113326,
"num_tokens": 8682455.0,
"step": 3950
},
{
"entropy": 5.769017505645752,
"epoch": 0.3925168717745137,
"grad_norm": 0.83203125,
"learning_rate": 0.0004990269522259509,
"loss": 5.6153,
"mean_token_accuracy": 0.1415529504418373,
"num_tokens": 8692975.0,
"step": 3955
},
{
"entropy": 5.729673624038696,
"epoch": 0.3930131004366812,
"grad_norm": 0.83203125,
"learning_rate": 0.0004990236578239755,
"loss": 5.5754,
"mean_token_accuracy": 0.14106195271015168,
"num_tokens": 8703569.0,
"step": 3960
},
{
"entropy": 5.820702600479126,
"epoch": 0.3935093290988487,
"grad_norm": 0.87890625,
"learning_rate": 0.0004990203578666962,
"loss": 5.6878,
"mean_token_accuracy": 0.13368871957063674,
"num_tokens": 8713734.0,
"step": 3965
},
{
"entropy": 5.783185625076294,
"epoch": 0.39400555776101626,
"grad_norm": 0.91796875,
"learning_rate": 0.0004990170523541948,
"loss": 5.5848,
"mean_token_accuracy": 0.14641537964344026,
"num_tokens": 8723852.0,
"step": 3970
},
{
"entropy": 5.793915796279907,
"epoch": 0.3945017864231838,
"grad_norm": 0.8515625,
"learning_rate": 0.0004990137412865531,
"loss": 5.747,
"mean_token_accuracy": 0.13510857298970222,
"num_tokens": 8734300.0,
"step": 3975
},
{
"entropy": 5.898438024520874,
"epoch": 0.3949980150853513,
"grad_norm": 1.0390625,
"learning_rate": 0.0004990104246638533,
"loss": 5.7356,
"mean_token_accuracy": 0.1344385825097561,
"num_tokens": 8745022.0,
"step": 3980
},
{
"entropy": 5.785170221328736,
"epoch": 0.39549424374751885,
"grad_norm": 0.8671875,
"learning_rate": 0.0004990071024861778,
"loss": 5.6298,
"mean_token_accuracy": 0.13691987618803977,
"num_tokens": 8754870.0,
"step": 3985
},
{
"entropy": 5.872950315475464,
"epoch": 0.3959904724096864,
"grad_norm": 0.85546875,
"learning_rate": 0.0004990037747536087,
"loss": 5.6865,
"mean_token_accuracy": 0.1410525433719158,
"num_tokens": 8765536.0,
"step": 3990
},
{
"entropy": 5.788661527633667,
"epoch": 0.3964867010718539,
"grad_norm": 1.03125,
"learning_rate": 0.0004990004414662288,
"loss": 5.6444,
"mean_token_accuracy": 0.13316506817936896,
"num_tokens": 8776898.0,
"step": 3995
},
{
"entropy": 5.795246744155884,
"epoch": 0.39698292973402144,
"grad_norm": 0.84765625,
"learning_rate": 0.0004989971026241205,
"loss": 5.7092,
"mean_token_accuracy": 0.1341576285660267,
"num_tokens": 8788176.0,
"step": 4000
},
{
"entropy": 5.745181846618652,
"epoch": 0.39747915839618897,
"grad_norm": 0.94921875,
"learning_rate": 0.0004989937582273668,
"loss": 5.6423,
"mean_token_accuracy": 0.13551107943058013,
"num_tokens": 8798939.0,
"step": 4005
},
{
"entropy": 5.790009880065918,
"epoch": 0.3979753870583565,
"grad_norm": 0.8671875,
"learning_rate": 0.0004989904082760506,
"loss": 5.6695,
"mean_token_accuracy": 0.14157403707504274,
"num_tokens": 8809092.0,
"step": 4010
},
{
"entropy": 5.820794916152954,
"epoch": 0.39847161572052403,
"grad_norm": 0.85546875,
"learning_rate": 0.000498987052770255,
"loss": 5.5906,
"mean_token_accuracy": 0.14081012904644014,
"num_tokens": 8819352.0,
"step": 4015
},
{
"entropy": 5.824670314788818,
"epoch": 0.39896784438269156,
"grad_norm": 0.87109375,
"learning_rate": 0.000498983691710063,
"loss": 5.7098,
"mean_token_accuracy": 0.1329125739634037,
"num_tokens": 8829878.0,
"step": 4020
},
{
"entropy": 5.778005743026734,
"epoch": 0.3994640730448591,
"grad_norm": 0.8984375,
"learning_rate": 0.000498980325095558,
"loss": 5.5798,
"mean_token_accuracy": 0.1393500655889511,
"num_tokens": 8840364.0,
"step": 4025
},
{
"entropy": 5.7912201404571535,
"epoch": 0.3999603017070266,
"grad_norm": 0.9140625,
"learning_rate": 0.0004989769529268238,
"loss": 5.6914,
"mean_token_accuracy": 0.1342613399028778,
"num_tokens": 8851340.0,
"step": 4030
},
{
"entropy": 5.750426435470581,
"epoch": 0.4004565303691941,
"grad_norm": 0.92578125,
"learning_rate": 0.0004989735752039436,
"loss": 5.6583,
"mean_token_accuracy": 0.13800134211778642,
"num_tokens": 8862571.0,
"step": 4035
},
{
"entropy": 5.796105480194091,
"epoch": 0.4009527590313616,
"grad_norm": 0.84765625,
"learning_rate": 0.0004989701919270015,
"loss": 5.6605,
"mean_token_accuracy": 0.1415472649037838,
"num_tokens": 8873927.0,
"step": 4040
},
{
"entropy": 5.866248321533203,
"epoch": 0.40144898769352916,
"grad_norm": 0.9296875,
"learning_rate": 0.0004989668030960812,
"loss": 5.8541,
"mean_token_accuracy": 0.12870026156306266,
"num_tokens": 8886542.0,
"step": 4045
},
{
"entropy": 5.886906385421753,
"epoch": 0.4019452163556967,
"grad_norm": 0.88671875,
"learning_rate": 0.0004989634087112667,
"loss": 5.6018,
"mean_token_accuracy": 0.1396520234644413,
"num_tokens": 8896114.0,
"step": 4050
},
{
"entropy": 5.802579689025879,
"epoch": 0.4024414450178642,
"grad_norm": 0.98828125,
"learning_rate": 0.0004989600087726425,
"loss": 5.6846,
"mean_token_accuracy": 0.1316719926893711,
"num_tokens": 8906303.0,
"step": 4055
},
{
"entropy": 5.829805946350097,
"epoch": 0.40293767368003175,
"grad_norm": 0.83203125,
"learning_rate": 0.0004989566032802924,
"loss": 5.7132,
"mean_token_accuracy": 0.13243434056639672,
"num_tokens": 8916961.0,
"step": 4060
},
{
"entropy": 5.879288005828857,
"epoch": 0.4034339023421993,
"grad_norm": 0.859375,
"learning_rate": 0.0004989531922343013,
"loss": 5.805,
"mean_token_accuracy": 0.13305630013346673,
"num_tokens": 8927700.0,
"step": 4065
},
{
"entropy": 5.819786024093628,
"epoch": 0.4039301310043668,
"grad_norm": 0.8046875,
"learning_rate": 0.0004989497756347534,
"loss": 5.6316,
"mean_token_accuracy": 0.13920575231313706,
"num_tokens": 8939655.0,
"step": 4070
},
{
"entropy": 5.780047941207886,
"epoch": 0.40442635966653434,
"grad_norm": 0.8359375,
"learning_rate": 0.0004989463534817339,
"loss": 5.5705,
"mean_token_accuracy": 0.14523619860410691,
"num_tokens": 8951238.0,
"step": 4075
},
{
"entropy": 5.802511358261109,
"epoch": 0.40492258832870187,
"grad_norm": 0.85546875,
"learning_rate": 0.0004989429257753273,
"loss": 5.6305,
"mean_token_accuracy": 0.14453877732157708,
"num_tokens": 8962019.0,
"step": 4080
},
{
"entropy": 5.7828209400177,
"epoch": 0.4054188169908694,
"grad_norm": 0.8984375,
"learning_rate": 0.0004989394925156187,
"loss": 5.6351,
"mean_token_accuracy": 0.13892287462949754,
"num_tokens": 8973861.0,
"step": 4085
},
{
"entropy": 5.741808176040649,
"epoch": 0.40591504565303693,
"grad_norm": 0.8125,
"learning_rate": 0.0004989360537026932,
"loss": 5.6414,
"mean_token_accuracy": 0.14035039395093918,
"num_tokens": 8985981.0,
"step": 4090
},
{
"entropy": 5.7918037414550785,
"epoch": 0.40641127431520446,
"grad_norm": 0.8671875,
"learning_rate": 0.000498932609336636,
"loss": 5.5418,
"mean_token_accuracy": 0.14218456000089646,
"num_tokens": 8995318.0,
"step": 4095
},
{
"entropy": 5.7594521045684814,
"epoch": 0.406907502977372,
"grad_norm": 0.84375,
"learning_rate": 0.0004989291594175327,
"loss": 5.5765,
"mean_token_accuracy": 0.1396835669875145,
"num_tokens": 9004679.0,
"step": 4100
},
{
"entropy": 5.91703233718872,
"epoch": 0.4074037316395395,
"grad_norm": 0.859375,
"learning_rate": 0.0004989257039454689,
"loss": 5.7059,
"mean_token_accuracy": 0.1379046157002449,
"num_tokens": 9016422.0,
"step": 4105
},
{
"entropy": 5.724780511856079,
"epoch": 0.40789996030170705,
"grad_norm": 0.84765625,
"learning_rate": 0.00049892224292053,
"loss": 5.6229,
"mean_token_accuracy": 0.14078758656978607,
"num_tokens": 9027914.0,
"step": 4110
},
{
"entropy": 5.761678218841553,
"epoch": 0.40839618896387453,
"grad_norm": 0.80859375,
"learning_rate": 0.0004989187763428021,
"loss": 5.5918,
"mean_token_accuracy": 0.144223216176033,
"num_tokens": 9038701.0,
"step": 4115
},
{
"entropy": 5.768312168121338,
"epoch": 0.40889241762604206,
"grad_norm": 0.90234375,
"learning_rate": 0.000498915304212371,
"loss": 5.6561,
"mean_token_accuracy": 0.13807908743619918,
"num_tokens": 9050620.0,
"step": 4120
},
{
"entropy": 5.846501874923706,
"epoch": 0.4093886462882096,
"grad_norm": 0.85546875,
"learning_rate": 0.0004989118265293228,
"loss": 5.6829,
"mean_token_accuracy": 0.14339982569217682,
"num_tokens": 9060900.0,
"step": 4125
},
{
"entropy": 5.792847585678101,
"epoch": 0.4098848749503771,
"grad_norm": 0.828125,
"learning_rate": 0.0004989083432937439,
"loss": 5.5866,
"mean_token_accuracy": 0.13959128633141518,
"num_tokens": 9072122.0,
"step": 4130
},
{
"entropy": 5.749606561660767,
"epoch": 0.41038110361254465,
"grad_norm": 0.84765625,
"learning_rate": 0.0004989048545057204,
"loss": 5.617,
"mean_token_accuracy": 0.13930945098400116,
"num_tokens": 9082780.0,
"step": 4135
},
{
"entropy": 5.782565975189209,
"epoch": 0.4108773322747122,
"grad_norm": 0.8984375,
"learning_rate": 0.0004989013601653391,
"loss": 5.6658,
"mean_token_accuracy": 0.1379122644662857,
"num_tokens": 9093518.0,
"step": 4140
},
{
"entropy": 5.767605924606324,
"epoch": 0.4113735609368797,
"grad_norm": 0.80859375,
"learning_rate": 0.0004988978602726866,
"loss": 5.6779,
"mean_token_accuracy": 0.1437633991241455,
"num_tokens": 9105035.0,
"step": 4145
},
{
"entropy": 5.812577152252198,
"epoch": 0.41186978959904724,
"grad_norm": 0.8359375,
"learning_rate": 0.0004988943548278494,
"loss": 5.587,
"mean_token_accuracy": 0.14110235944390298,
"num_tokens": 9117154.0,
"step": 4150
},
{
"entropy": 5.77940182685852,
"epoch": 0.4123660182612148,
"grad_norm": 0.87890625,
"learning_rate": 0.0004988908438309149,
"loss": 5.6667,
"mean_token_accuracy": 0.13823975771665573,
"num_tokens": 9128818.0,
"step": 4155
},
{
"entropy": 5.8112109184265135,
"epoch": 0.4128622469233823,
"grad_norm": 0.80859375,
"learning_rate": 0.0004988873272819697,
"loss": 5.67,
"mean_token_accuracy": 0.1413659855723381,
"num_tokens": 9140188.0,
"step": 4160
},
{
"entropy": 5.94907431602478,
"epoch": 0.41335847558554983,
"grad_norm": 0.9765625,
"learning_rate": 0.0004988838051811014,
"loss": 5.7523,
"mean_token_accuracy": 0.134553412348032,
"num_tokens": 9150565.0,
"step": 4165
},
{
"entropy": 5.801510667800903,
"epoch": 0.41385470424771736,
"grad_norm": 0.8984375,
"learning_rate": 0.000498880277528397,
"loss": 5.6054,
"mean_token_accuracy": 0.13975557759404184,
"num_tokens": 9160130.0,
"step": 4170
},
{
"entropy": 5.799229383468628,
"epoch": 0.4143509329098849,
"grad_norm": 0.859375,
"learning_rate": 0.0004988767443239443,
"loss": 5.6117,
"mean_token_accuracy": 0.1398836188018322,
"num_tokens": 9171740.0,
"step": 4175
},
{
"entropy": 5.76237120628357,
"epoch": 0.4148471615720524,
"grad_norm": 0.9140625,
"learning_rate": 0.0004988732055678307,
"loss": 5.5544,
"mean_token_accuracy": 0.15382059216499328,
"num_tokens": 9182459.0,
"step": 4180
},
{
"entropy": 5.805677366256714,
"epoch": 0.41534339023421996,
"grad_norm": 0.83984375,
"learning_rate": 0.000498869661260144,
"loss": 5.6322,
"mean_token_accuracy": 0.14206244200468063,
"num_tokens": 9192284.0,
"step": 4185
},
{
"entropy": 5.780238819122315,
"epoch": 0.41583961889638743,
"grad_norm": 0.9453125,
"learning_rate": 0.000498866111400972,
"loss": 5.6115,
"mean_token_accuracy": 0.1399123787879944,
"num_tokens": 9202700.0,
"step": 4190
},
{
"entropy": 5.7953057289123535,
"epoch": 0.41633584755855496,
"grad_norm": 0.90234375,
"learning_rate": 0.0004988625559904029,
"loss": 5.6243,
"mean_token_accuracy": 0.13381586745381355,
"num_tokens": 9213013.0,
"step": 4195
},
{
"entropy": 5.83180456161499,
"epoch": 0.4168320762207225,
"grad_norm": 0.7421875,
"learning_rate": 0.0004988589950285248,
"loss": 5.6966,
"mean_token_accuracy": 0.13726072534918785,
"num_tokens": 9224568.0,
"step": 4200
},
{
"entropy": 5.76538610458374,
"epoch": 0.41732830488289,
"grad_norm": 0.85546875,
"learning_rate": 0.0004988554285154259,
"loss": 5.5748,
"mean_token_accuracy": 0.13923351615667343,
"num_tokens": 9233791.0,
"step": 4205
},
{
"entropy": 5.792706155776978,
"epoch": 0.41782453354505755,
"grad_norm": 0.8515625,
"learning_rate": 0.0004988518564511948,
"loss": 5.6178,
"mean_token_accuracy": 0.13635936975479127,
"num_tokens": 9245048.0,
"step": 4210
},
{
"entropy": 5.860803031921387,
"epoch": 0.4183207622072251,
"grad_norm": 0.83203125,
"learning_rate": 0.0004988482788359202,
"loss": 5.6396,
"mean_token_accuracy": 0.14628660529851914,
"num_tokens": 9256508.0,
"step": 4215
},
{
"entropy": 5.803260707855225,
"epoch": 0.4188169908693926,
"grad_norm": 0.87890625,
"learning_rate": 0.0004988446956696904,
"loss": 5.668,
"mean_token_accuracy": 0.14071763530373574,
"num_tokens": 9267871.0,
"step": 4220
},
{
"entropy": 5.85851092338562,
"epoch": 0.41931321953156014,
"grad_norm": 0.84765625,
"learning_rate": 0.0004988411069525947,
"loss": 5.7168,
"mean_token_accuracy": 0.13775873184204102,
"num_tokens": 9279287.0,
"step": 4225
},
{
"entropy": 5.862564992904663,
"epoch": 0.4198094481937277,
"grad_norm": 0.90234375,
"learning_rate": 0.0004988375126847216,
"loss": 5.7249,
"mean_token_accuracy": 0.13378578275442124,
"num_tokens": 9291257.0,
"step": 4230
},
{
"entropy": 5.840371370315552,
"epoch": 0.4203056768558952,
"grad_norm": 1.0078125,
"learning_rate": 0.0004988339128661607,
"loss": 5.7021,
"mean_token_accuracy": 0.1337619088590145,
"num_tokens": 9302590.0,
"step": 4235
},
{
"entropy": 5.699577569961548,
"epoch": 0.42080190551806274,
"grad_norm": 0.80859375,
"learning_rate": 0.000498830307497001,
"loss": 5.5512,
"mean_token_accuracy": 0.14313922077417374,
"num_tokens": 9313174.0,
"step": 4240
},
{
"entropy": 5.794048738479614,
"epoch": 0.42129813418023027,
"grad_norm": 0.88671875,
"learning_rate": 0.000498826696577332,
"loss": 5.6396,
"mean_token_accuracy": 0.1417992301285267,
"num_tokens": 9325034.0,
"step": 4245
},
{
"entropy": 5.806614542007447,
"epoch": 0.4217943628423978,
"grad_norm": 0.86328125,
"learning_rate": 0.0004988230801072432,
"loss": 5.6206,
"mean_token_accuracy": 0.13834859877824784,
"num_tokens": 9336834.0,
"step": 4250
},
{
"entropy": 5.737400197982788,
"epoch": 0.4222905915045653,
"grad_norm": 0.8671875,
"learning_rate": 0.0004988194580868244,
"loss": 5.6237,
"mean_token_accuracy": 0.13811139315366744,
"num_tokens": 9347747.0,
"step": 4255
},
{
"entropy": 5.771370553970337,
"epoch": 0.42278682016673286,
"grad_norm": 0.82421875,
"learning_rate": 0.0004988158305161654,
"loss": 5.6391,
"mean_token_accuracy": 0.1432290256023407,
"num_tokens": 9359186.0,
"step": 4260
},
{
"entropy": 5.782405090332031,
"epoch": 0.42328304882890033,
"grad_norm": 0.86328125,
"learning_rate": 0.000498812197395356,
"loss": 5.5621,
"mean_token_accuracy": 0.14198907613754272,
"num_tokens": 9369485.0,
"step": 4265
},
{
"entropy": 5.782733201980591,
"epoch": 0.42377927749106786,
"grad_norm": 0.96875,
"learning_rate": 0.0004988085587244863,
"loss": 5.6577,
"mean_token_accuracy": 0.13524223864078522,
"num_tokens": 9382658.0,
"step": 4270
},
{
"entropy": 5.74179515838623,
"epoch": 0.4242755061532354,
"grad_norm": 0.81640625,
"learning_rate": 0.0004988049145036466,
"loss": 5.6175,
"mean_token_accuracy": 0.1354674480855465,
"num_tokens": 9394151.0,
"step": 4275
},
{
"entropy": 5.861121129989624,
"epoch": 0.4247717348154029,
"grad_norm": 0.8203125,
"learning_rate": 0.0004988012647329273,
"loss": 5.7112,
"mean_token_accuracy": 0.13729124367237092,
"num_tokens": 9405956.0,
"step": 4280
},
{
"entropy": 5.821110439300537,
"epoch": 0.42526796347757045,
"grad_norm": 0.90625,
"learning_rate": 0.000498797609412419,
"loss": 5.6038,
"mean_token_accuracy": 0.13974196910858155,
"num_tokens": 9416889.0,
"step": 4285
},
{
"entropy": 5.7748674869537355,
"epoch": 0.425764192139738,
"grad_norm": 0.83203125,
"learning_rate": 0.0004987939485422122,
"loss": 5.7055,
"mean_token_accuracy": 0.136608724296093,
"num_tokens": 9427602.0,
"step": 4290
},
{
"entropy": 5.823736667633057,
"epoch": 0.4262604208019055,
"grad_norm": 0.7578125,
"learning_rate": 0.0004987902821223976,
"loss": 5.629,
"mean_token_accuracy": 0.14252188354730605,
"num_tokens": 9438557.0,
"step": 4295
},
{
"entropy": 5.777904653549195,
"epoch": 0.42675664946407305,
"grad_norm": 0.8125,
"learning_rate": 0.0004987866101530663,
"loss": 5.667,
"mean_token_accuracy": 0.13800618946552276,
"num_tokens": 9449598.0,
"step": 4300
},
{
"entropy": 5.8158505916595455,
"epoch": 0.4272528781262406,
"grad_norm": 0.84765625,
"learning_rate": 0.0004987829326343092,
"loss": 5.6076,
"mean_token_accuracy": 0.14416556805372238,
"num_tokens": 9459740.0,
"step": 4305
},
{
"entropy": 5.753842973709107,
"epoch": 0.4277491067884081,
"grad_norm": 0.88671875,
"learning_rate": 0.0004987792495662177,
"loss": 5.6073,
"mean_token_accuracy": 0.14074130430817605,
"num_tokens": 9469835.0,
"step": 4310
},
{
"entropy": 5.818010139465332,
"epoch": 0.42824533545057564,
"grad_norm": 0.9296875,
"learning_rate": 0.000498775560948883,
"loss": 5.6917,
"mean_token_accuracy": 0.13898994252085686,
"num_tokens": 9480735.0,
"step": 4315
},
{
"entropy": 5.785709524154663,
"epoch": 0.42874156411274317,
"grad_norm": 0.91015625,
"learning_rate": 0.0004987718667823965,
"loss": 5.5887,
"mean_token_accuracy": 0.14314745515584945,
"num_tokens": 9491321.0,
"step": 4320
},
{
"entropy": 5.817282676696777,
"epoch": 0.4292377927749107,
"grad_norm": 0.78125,
"learning_rate": 0.00049876816706685,
"loss": 5.721,
"mean_token_accuracy": 0.13546350300312043,
"num_tokens": 9503507.0,
"step": 4325
},
{
"entropy": 5.835805463790893,
"epoch": 0.42973402143707823,
"grad_norm": 1.6953125,
"learning_rate": 0.000498764461802335,
"loss": 5.7267,
"mean_token_accuracy": 0.14347420930862426,
"num_tokens": 9514404.0,
"step": 4330
},
{
"entropy": 5.736133670806884,
"epoch": 0.43023025009924576,
"grad_norm": 0.921875,
"learning_rate": 0.0004987607509889437,
"loss": 5.5945,
"mean_token_accuracy": 0.13721355572342872,
"num_tokens": 9525555.0,
"step": 4335
},
{
"entropy": 5.844026803970337,
"epoch": 0.43072647876141323,
"grad_norm": 0.87890625,
"learning_rate": 0.0004987570346267677,
"loss": 5.593,
"mean_token_accuracy": 0.14471945762634278,
"num_tokens": 9535084.0,
"step": 4340
},
{
"entropy": 5.716549587249756,
"epoch": 0.43122270742358076,
"grad_norm": 0.81640625,
"learning_rate": 0.0004987533127158996,
"loss": 5.5996,
"mean_token_accuracy": 0.1444016769528389,
"num_tokens": 9546170.0,
"step": 4345
},
{
"entropy": 5.811110973358154,
"epoch": 0.4317189360857483,
"grad_norm": 0.84375,
"learning_rate": 0.0004987495852564314,
"loss": 5.6225,
"mean_token_accuracy": 0.14201515763998032,
"num_tokens": 9556793.0,
"step": 4350
},
{
"entropy": 5.784374189376831,
"epoch": 0.4322151647479158,
"grad_norm": 0.94140625,
"learning_rate": 0.0004987458522484556,
"loss": 5.566,
"mean_token_accuracy": 0.14575942158699035,
"num_tokens": 9568114.0,
"step": 4355
},
{
"entropy": 5.740352964401245,
"epoch": 0.43271139341008336,
"grad_norm": 0.95703125,
"learning_rate": 0.0004987421136920648,
"loss": 5.5771,
"mean_token_accuracy": 0.13610847890377045,
"num_tokens": 9579119.0,
"step": 4360
},
{
"entropy": 5.722053813934326,
"epoch": 0.4332076220722509,
"grad_norm": 0.9296875,
"learning_rate": 0.0004987383695873517,
"loss": 5.6147,
"mean_token_accuracy": 0.14022180810570717,
"num_tokens": 9590259.0,
"step": 4365
},
{
"entropy": 5.802224493026733,
"epoch": 0.4337038507344184,
"grad_norm": 0.875,
"learning_rate": 0.0004987346199344092,
"loss": 5.6512,
"mean_token_accuracy": 0.13241124674677848,
"num_tokens": 9600431.0,
"step": 4370
},
{
"entropy": 5.747055196762085,
"epoch": 0.43420007939658595,
"grad_norm": 0.79296875,
"learning_rate": 0.0004987308647333303,
"loss": 5.6298,
"mean_token_accuracy": 0.14093587100505828,
"num_tokens": 9610286.0,
"step": 4375
},
{
"entropy": 5.844609880447388,
"epoch": 0.4346963080587535,
"grad_norm": 0.98828125,
"learning_rate": 0.0004987271039842079,
"loss": 5.6392,
"mean_token_accuracy": 0.14039682298898698,
"num_tokens": 9621148.0,
"step": 4380
},
{
"entropy": 5.768031358718872,
"epoch": 0.435192536720921,
"grad_norm": 0.83203125,
"learning_rate": 0.0004987233376871355,
"loss": 5.5234,
"mean_token_accuracy": 0.1461140587925911,
"num_tokens": 9631300.0,
"step": 4385
},
{
"entropy": 5.78350191116333,
"epoch": 0.43568876538308854,
"grad_norm": 0.8359375,
"learning_rate": 0.0004987195658422063,
"loss": 5.6269,
"mean_token_accuracy": 0.13842125460505486,
"num_tokens": 9641986.0,
"step": 4390
},
{
"entropy": 5.781039094924926,
"epoch": 0.43618499404525607,
"grad_norm": 0.8046875,
"learning_rate": 0.0004987157884495141,
"loss": 5.6395,
"mean_token_accuracy": 0.1402071826159954,
"num_tokens": 9652962.0,
"step": 4395
},
{
"entropy": 5.757225465774536,
"epoch": 0.4366812227074236,
"grad_norm": 0.82421875,
"learning_rate": 0.0004987120055091523,
"loss": 5.6788,
"mean_token_accuracy": 0.13589339554309846,
"num_tokens": 9664084.0,
"step": 4400
},
{
"entropy": 5.754967546463012,
"epoch": 0.43717745136959113,
"grad_norm": 0.8203125,
"learning_rate": 0.0004987082170212149,
"loss": 5.5724,
"mean_token_accuracy": 0.1397831603884697,
"num_tokens": 9675821.0,
"step": 4405
},
{
"entropy": 5.774995422363281,
"epoch": 0.43767368003175866,
"grad_norm": 0.9140625,
"learning_rate": 0.0004987044229857957,
"loss": 5.6874,
"mean_token_accuracy": 0.1395527347922325,
"num_tokens": 9687964.0,
"step": 4410
},
{
"entropy": 5.788907766342163,
"epoch": 0.43816990869392614,
"grad_norm": 0.796875,
"learning_rate": 0.0004987006234029889,
"loss": 5.6718,
"mean_token_accuracy": 0.139390766620636,
"num_tokens": 9699653.0,
"step": 4415
},
{
"entropy": 5.857262229919433,
"epoch": 0.43866613735609367,
"grad_norm": 0.91796875,
"learning_rate": 0.0004986968182728885,
"loss": 5.6781,
"mean_token_accuracy": 0.13918035179376603,
"num_tokens": 9709872.0,
"step": 4420
},
{
"entropy": 5.817832374572754,
"epoch": 0.4391623660182612,
"grad_norm": 1.0390625,
"learning_rate": 0.0004986930075955892,
"loss": 5.601,
"mean_token_accuracy": 0.13685947135090828,
"num_tokens": 9720589.0,
"step": 4425
},
{
"entropy": 5.700802898406982,
"epoch": 0.4396585946804287,
"grad_norm": 0.87109375,
"learning_rate": 0.0004986891913711853,
"loss": 5.5351,
"mean_token_accuracy": 0.14478714317083358,
"num_tokens": 9731862.0,
"step": 4430
},
{
"entropy": 5.73977108001709,
"epoch": 0.44015482334259626,
"grad_norm": 0.83203125,
"learning_rate": 0.0004986853695997715,
"loss": 5.6872,
"mean_token_accuracy": 0.14369284883141517,
"num_tokens": 9743710.0,
"step": 4435
},
{
"entropy": 5.799801588058472,
"epoch": 0.4406510520047638,
"grad_norm": 0.921875,
"learning_rate": 0.0004986815422814425,
"loss": 5.6418,
"mean_token_accuracy": 0.13150426223874093,
"num_tokens": 9753855.0,
"step": 4440
},
{
"entropy": 5.875456953048706,
"epoch": 0.4411472806669313,
"grad_norm": 0.90234375,
"learning_rate": 0.0004986777094162932,
"loss": 5.6379,
"mean_token_accuracy": 0.14705183878540992,
"num_tokens": 9763883.0,
"step": 4445
},
{
"entropy": 5.748485708236695,
"epoch": 0.44164350932909885,
"grad_norm": 0.83203125,
"learning_rate": 0.0004986738710044187,
"loss": 5.5989,
"mean_token_accuracy": 0.14280258789658545,
"num_tokens": 9774890.0,
"step": 4450
},
{
"entropy": 5.732300472259522,
"epoch": 0.4421397379912664,
"grad_norm": 0.84375,
"learning_rate": 0.0004986700270459142,
"loss": 5.5783,
"mean_token_accuracy": 0.1434408903121948,
"num_tokens": 9785249.0,
"step": 4455
},
{
"entropy": 5.791567468643189,
"epoch": 0.4426359666534339,
"grad_norm": 0.8984375,
"learning_rate": 0.0004986661775408749,
"loss": 5.5913,
"mean_token_accuracy": 0.14472425431013108,
"num_tokens": 9795723.0,
"step": 4460
},
{
"entropy": 5.7759202003479,
"epoch": 0.44313219531560144,
"grad_norm": 0.79296875,
"learning_rate": 0.0004986623224893965,
"loss": 5.5394,
"mean_token_accuracy": 0.1415581427514553,
"num_tokens": 9807637.0,
"step": 4465
},
{
"entropy": 5.784197425842285,
"epoch": 0.44362842397776897,
"grad_norm": 0.84375,
"learning_rate": 0.0004986584618915744,
"loss": 5.648,
"mean_token_accuracy": 0.1421603761613369,
"num_tokens": 9820298.0,
"step": 4470
},
{
"entropy": 5.709079837799072,
"epoch": 0.4441246526399365,
"grad_norm": 0.95703125,
"learning_rate": 0.0004986545957475045,
"loss": 5.5383,
"mean_token_accuracy": 0.14711108356714248,
"num_tokens": 9831251.0,
"step": 4475
},
{
"entropy": 5.734958457946777,
"epoch": 0.44462088130210403,
"grad_norm": 0.8203125,
"learning_rate": 0.0004986507240572825,
"loss": 5.596,
"mean_token_accuracy": 0.14287913516163825,
"num_tokens": 9842884.0,
"step": 4480
},
{
"entropy": 5.72746171951294,
"epoch": 0.44511710996427156,
"grad_norm": 0.83984375,
"learning_rate": 0.0004986468468210042,
"loss": 5.4933,
"mean_token_accuracy": 0.14691198095679284,
"num_tokens": 9854413.0,
"step": 4485
},
{
"entropy": 5.793558883666992,
"epoch": 0.44561333862643904,
"grad_norm": 0.8671875,
"learning_rate": 0.0004986429640387663,
"loss": 5.6508,
"mean_token_accuracy": 0.1404290720820427,
"num_tokens": 9865771.0,
"step": 4490
},
{
"entropy": 5.766204071044922,
"epoch": 0.44610956728860657,
"grad_norm": 0.7734375,
"learning_rate": 0.0004986390757106648,
"loss": 5.5707,
"mean_token_accuracy": 0.1469174288213253,
"num_tokens": 9876717.0,
"step": 4495
},
{
"entropy": 5.730377721786499,
"epoch": 0.4466057959507741,
"grad_norm": 0.79296875,
"learning_rate": 0.0004986351818367959,
"loss": 5.6486,
"mean_token_accuracy": 0.140429724752903,
"num_tokens": 9887616.0,
"step": 4500
},
{
"entropy": 5.818638896942138,
"epoch": 0.44710202461294163,
"grad_norm": 0.8984375,
"learning_rate": 0.0004986312824172566,
"loss": 5.595,
"mean_token_accuracy": 0.14335185587406157,
"num_tokens": 9898979.0,
"step": 4505
},
{
"entropy": 5.797679376602173,
"epoch": 0.44759825327510916,
"grad_norm": 0.7890625,
"learning_rate": 0.0004986273774521433,
"loss": 5.6306,
"mean_token_accuracy": 0.13533177375793456,
"num_tokens": 9912158.0,
"step": 4510
},
{
"entropy": 5.7367393493652346,
"epoch": 0.4480944819372767,
"grad_norm": 0.90234375,
"learning_rate": 0.0004986234669415529,
"loss": 5.6312,
"mean_token_accuracy": 0.13736921548843384,
"num_tokens": 9923886.0,
"step": 4515
},
{
"entropy": 5.776280212402344,
"epoch": 0.4485907105994442,
"grad_norm": 0.79296875,
"learning_rate": 0.0004986195508855824,
"loss": 5.6023,
"mean_token_accuracy": 0.14297038465738296,
"num_tokens": 9934979.0,
"step": 4520
},
{
"entropy": 5.79187970161438,
"epoch": 0.44908693926161175,
"grad_norm": 0.8828125,
"learning_rate": 0.000498615629284329,
"loss": 5.7071,
"mean_token_accuracy": 0.13817890509963035,
"num_tokens": 9944715.0,
"step": 4525
},
{
"entropy": 5.775436878204346,
"epoch": 0.4495831679237793,
"grad_norm": 0.82421875,
"learning_rate": 0.0004986117021378896,
"loss": 5.5741,
"mean_token_accuracy": 0.14012717232108116,
"num_tokens": 9954736.0,
"step": 4530
},
{
"entropy": 5.79448766708374,
"epoch": 0.4500793965859468,
"grad_norm": 0.75,
"learning_rate": 0.000498607769446362,
"loss": 5.6594,
"mean_token_accuracy": 0.1388283774256706,
"num_tokens": 9965952.0,
"step": 4535
},
{
"entropy": 5.7624434471130375,
"epoch": 0.45057562524811434,
"grad_norm": 0.94140625,
"learning_rate": 0.0004986038312098435,
"loss": 5.604,
"mean_token_accuracy": 0.14314566776156426,
"num_tokens": 9976428.0,
"step": 4540
},
{
"entropy": 5.75066294670105,
"epoch": 0.4510718539102819,
"grad_norm": 0.84765625,
"learning_rate": 0.0004985998874284319,
"loss": 5.546,
"mean_token_accuracy": 0.1484223261475563,
"num_tokens": 9986100.0,
"step": 4545
},
{
"entropy": 5.655505847930908,
"epoch": 0.4515680825724494,
"grad_norm": 0.83203125,
"learning_rate": 0.0004985959381022248,
"loss": 5.5458,
"mean_token_accuracy": 0.14303347319364548,
"num_tokens": 9996584.0,
"step": 4550
},
{
"entropy": 5.739107084274292,
"epoch": 0.45206431123461693,
"grad_norm": 0.94140625,
"learning_rate": 0.0004985919832313203,
"loss": 5.557,
"mean_token_accuracy": 0.14798323214054107,
"num_tokens": 10007792.0,
"step": 4555
},
{
"entropy": 5.851394081115723,
"epoch": 0.45256053989678446,
"grad_norm": 0.828125,
"learning_rate": 0.0004985880228158164,
"loss": 5.756,
"mean_token_accuracy": 0.13254465758800507,
"num_tokens": 10019615.0,
"step": 4560
},
{
"entropy": 5.697079610824585,
"epoch": 0.45305676855895194,
"grad_norm": 0.91796875,
"learning_rate": 0.0004985840568558112,
"loss": 5.5814,
"mean_token_accuracy": 0.14153416082262993,
"num_tokens": 10030223.0,
"step": 4565
},
{
"entropy": 5.775788831710815,
"epoch": 0.45355299722111947,
"grad_norm": 0.8203125,
"learning_rate": 0.0004985800853514033,
"loss": 5.5967,
"mean_token_accuracy": 0.137832148373127,
"num_tokens": 10041522.0,
"step": 4570
},
{
"entropy": 5.785793733596802,
"epoch": 0.454049225883287,
"grad_norm": 0.8828125,
"learning_rate": 0.0004985761083026911,
"loss": 5.5212,
"mean_token_accuracy": 0.15029784739017488,
"num_tokens": 10052847.0,
"step": 4575
},
{
"entropy": 5.655645513534546,
"epoch": 0.45454545454545453,
"grad_norm": 0.875,
"learning_rate": 0.0004985721257097731,
"loss": 5.5434,
"mean_token_accuracy": 0.14386311173439026,
"num_tokens": 10062259.0,
"step": 4580
},
{
"entropy": 5.820441770553589,
"epoch": 0.45504168320762206,
"grad_norm": 0.87890625,
"learning_rate": 0.0004985681375727482,
"loss": 5.7118,
"mean_token_accuracy": 0.13377906680107116,
"num_tokens": 10073341.0,
"step": 4585
},
{
"entropy": 5.822879838943481,
"epoch": 0.4555379118697896,
"grad_norm": 0.7890625,
"learning_rate": 0.0004985641438917151,
"loss": 5.5629,
"mean_token_accuracy": 0.14327463507652283,
"num_tokens": 10083992.0,
"step": 4590
},
{
"entropy": 5.765067672729492,
"epoch": 0.4560341405319571,
"grad_norm": 0.84765625,
"learning_rate": 0.0004985601446667731,
"loss": 5.5608,
"mean_token_accuracy": 0.14663187712430953,
"num_tokens": 10094245.0,
"step": 4595
},
{
"entropy": 5.7631641864776615,
"epoch": 0.45653036919412465,
"grad_norm": 0.83984375,
"learning_rate": 0.0004985561398980213,
"loss": 5.5366,
"mean_token_accuracy": 0.14789687395095824,
"num_tokens": 10105976.0,
"step": 4600
},
{
"entropy": 5.672163963317871,
"epoch": 0.4570265978562922,
"grad_norm": 0.859375,
"learning_rate": 0.0004985521295855588,
"loss": 5.6168,
"mean_token_accuracy": 0.14210953339934349,
"num_tokens": 10117373.0,
"step": 4605
},
{
"entropy": 5.754079294204712,
"epoch": 0.4575228265184597,
"grad_norm": 0.82421875,
"learning_rate": 0.0004985481137294853,
"loss": 5.5685,
"mean_token_accuracy": 0.13775459825992584,
"num_tokens": 10128470.0,
"step": 4610
},
{
"entropy": 5.714517211914062,
"epoch": 0.45801905518062724,
"grad_norm": 0.8125,
"learning_rate": 0.0004985440923299002,
"loss": 5.5662,
"mean_token_accuracy": 0.1423081412911415,
"num_tokens": 10140217.0,
"step": 4615
},
{
"entropy": 5.710232543945312,
"epoch": 0.4585152838427948,
"grad_norm": 0.88671875,
"learning_rate": 0.0004985400653869034,
"loss": 5.633,
"mean_token_accuracy": 0.14135775342583656,
"num_tokens": 10152308.0,
"step": 4620
},
{
"entropy": 5.8700785636901855,
"epoch": 0.4590115125049623,
"grad_norm": 0.875,
"learning_rate": 0.0004985360329005945,
"loss": 5.6582,
"mean_token_accuracy": 0.14096900820732117,
"num_tokens": 10164234.0,
"step": 4625
},
{
"entropy": 5.874130821228027,
"epoch": 0.45950774116712984,
"grad_norm": 0.84765625,
"learning_rate": 0.0004985319948710738,
"loss": 5.7163,
"mean_token_accuracy": 0.13750708624720573,
"num_tokens": 10175287.0,
"step": 4630
},
{
"entropy": 5.801472568511963,
"epoch": 0.46000396982929737,
"grad_norm": 1.0,
"learning_rate": 0.0004985279512984413,
"loss": 5.5875,
"mean_token_accuracy": 0.14970208927989007,
"num_tokens": 10184658.0,
"step": 4635
},
{
"entropy": 5.737123918533325,
"epoch": 0.46050019849146484,
"grad_norm": 0.93359375,
"learning_rate": 0.0004985239021827972,
"loss": 5.5849,
"mean_token_accuracy": 0.1390370771288872,
"num_tokens": 10194593.0,
"step": 4640
},
{
"entropy": 5.701331233978271,
"epoch": 0.46099642715363237,
"grad_norm": 0.79296875,
"learning_rate": 0.000498519847524242,
"loss": 5.5092,
"mean_token_accuracy": 0.1458827719092369,
"num_tokens": 10205200.0,
"step": 4645
},
{
"entropy": 5.724951648712159,
"epoch": 0.4614926558157999,
"grad_norm": 0.83984375,
"learning_rate": 0.0004985157873228762,
"loss": 5.5194,
"mean_token_accuracy": 0.14664195030927657,
"num_tokens": 10216848.0,
"step": 4650
},
{
"entropy": 5.757262468338013,
"epoch": 0.46198888447796743,
"grad_norm": 0.8359375,
"learning_rate": 0.0004985117215788005,
"loss": 5.5703,
"mean_token_accuracy": 0.14250332042574881,
"num_tokens": 10228188.0,
"step": 4655
},
{
"entropy": 5.751273918151855,
"epoch": 0.46248511314013496,
"grad_norm": 0.875,
"learning_rate": 0.0004985076502921158,
"loss": 5.659,
"mean_token_accuracy": 0.1386226199567318,
"num_tokens": 10238253.0,
"step": 4660
},
{
"entropy": 5.799253034591675,
"epoch": 0.4629813418023025,
"grad_norm": 0.94921875,
"learning_rate": 0.000498503573462923,
"loss": 5.604,
"mean_token_accuracy": 0.14217669069766997,
"num_tokens": 10248169.0,
"step": 4665
},
{
"entropy": 5.719962024688721,
"epoch": 0.46347757046447,
"grad_norm": 0.82421875,
"learning_rate": 0.0004984994910913232,
"loss": 5.4916,
"mean_token_accuracy": 0.14536735266447068,
"num_tokens": 10258962.0,
"step": 4670
},
{
"entropy": 5.754688787460327,
"epoch": 0.46397379912663755,
"grad_norm": 0.765625,
"learning_rate": 0.0004984954031774175,
"loss": 5.5963,
"mean_token_accuracy": 0.14113237708806992,
"num_tokens": 10270303.0,
"step": 4675
},
{
"entropy": 5.649554014205933,
"epoch": 0.4644700277888051,
"grad_norm": 0.7421875,
"learning_rate": 0.0004984913097213075,
"loss": 5.5111,
"mean_token_accuracy": 0.1448366239666939,
"num_tokens": 10281782.0,
"step": 4680
},
{
"entropy": 5.7514485836029055,
"epoch": 0.4649662564509726,
"grad_norm": 0.9140625,
"learning_rate": 0.0004984872107230945,
"loss": 5.5542,
"mean_token_accuracy": 0.14501627534627914,
"num_tokens": 10292937.0,
"step": 4685
},
{
"entropy": 5.758346080780029,
"epoch": 0.46546248511314015,
"grad_norm": 0.796875,
"learning_rate": 0.0004984831061828803,
"loss": 5.606,
"mean_token_accuracy": 0.13853585198521615,
"num_tokens": 10305443.0,
"step": 4690
},
{
"entropy": 5.717559385299682,
"epoch": 0.4659587137753077,
"grad_norm": 0.83203125,
"learning_rate": 0.0004984789961007667,
"loss": 5.5427,
"mean_token_accuracy": 0.14828034415841101,
"num_tokens": 10316489.0,
"step": 4695
},
{
"entropy": 5.701294040679931,
"epoch": 0.4664549424374752,
"grad_norm": 0.91796875,
"learning_rate": 0.0004984748804768554,
"loss": 5.5884,
"mean_token_accuracy": 0.1406767785549164,
"num_tokens": 10326690.0,
"step": 4700
},
{
"entropy": 5.7987219333648685,
"epoch": 0.46695117109964274,
"grad_norm": 0.90234375,
"learning_rate": 0.0004984707593112487,
"loss": 5.5973,
"mean_token_accuracy": 0.13755468428134918,
"num_tokens": 10337059.0,
"step": 4705
},
{
"entropy": 5.678113794326782,
"epoch": 0.46744739976181027,
"grad_norm": 0.84375,
"learning_rate": 0.0004984666326040487,
"loss": 5.4636,
"mean_token_accuracy": 0.15040105506777762,
"num_tokens": 10347035.0,
"step": 4710
},
{
"entropy": 5.685402202606201,
"epoch": 0.46794362842397774,
"grad_norm": 0.890625,
"learning_rate": 0.0004984625003553578,
"loss": 5.5931,
"mean_token_accuracy": 0.14260579720139505,
"num_tokens": 10359492.0,
"step": 4715
},
{
"entropy": 5.814010429382324,
"epoch": 0.4684398570861453,
"grad_norm": 0.890625,
"learning_rate": 0.0004984583625652783,
"loss": 5.573,
"mean_token_accuracy": 0.1426217183470726,
"num_tokens": 10370128.0,
"step": 4720
},
{
"entropy": 5.729959917068482,
"epoch": 0.4689360857483128,
"grad_norm": 0.8984375,
"learning_rate": 0.000498454219233913,
"loss": 5.5762,
"mean_token_accuracy": 0.13988015651702881,
"num_tokens": 10380276.0,
"step": 4725
},
{
"entropy": 5.6514264106750485,
"epoch": 0.46943231441048033,
"grad_norm": 0.89453125,
"learning_rate": 0.0004984500703613645,
"loss": 5.5602,
"mean_token_accuracy": 0.1456937611103058,
"num_tokens": 10392489.0,
"step": 4730
},
{
"entropy": 5.83931975364685,
"epoch": 0.46992854307264786,
"grad_norm": 0.7421875,
"learning_rate": 0.0004984459159477358,
"loss": 5.6432,
"mean_token_accuracy": 0.13943550512194633,
"num_tokens": 10404259.0,
"step": 4735
},
{
"entropy": 5.690709781646729,
"epoch": 0.4704247717348154,
"grad_norm": 0.8515625,
"learning_rate": 0.0004984417559931298,
"loss": 5.5047,
"mean_token_accuracy": 0.14229437559843064,
"num_tokens": 10415000.0,
"step": 4740
},
{
"entropy": 5.74327278137207,
"epoch": 0.4709210003969829,
"grad_norm": 0.93359375,
"learning_rate": 0.0004984375904976498,
"loss": 5.6375,
"mean_token_accuracy": 0.13753386437892914,
"num_tokens": 10426078.0,
"step": 4745
},
{
"entropy": 5.792974853515625,
"epoch": 0.47141722905915046,
"grad_norm": 0.78125,
"learning_rate": 0.0004984334194613989,
"loss": 5.5616,
"mean_token_accuracy": 0.14064555689692498,
"num_tokens": 10436838.0,
"step": 4750
},
{
"entropy": 5.712934160232544,
"epoch": 0.471913457721318,
"grad_norm": 1.0234375,
"learning_rate": 0.0004984292428844809,
"loss": 5.5534,
"mean_token_accuracy": 0.1431037314236164,
"num_tokens": 10448544.0,
"step": 4755
},
{
"entropy": 5.819026613235474,
"epoch": 0.4724096863834855,
"grad_norm": 0.83984375,
"learning_rate": 0.0004984250607669989,
"loss": 5.6213,
"mean_token_accuracy": 0.1387665741145611,
"num_tokens": 10459730.0,
"step": 4760
},
{
"entropy": 5.8150371551513675,
"epoch": 0.47290591504565305,
"grad_norm": 0.890625,
"learning_rate": 0.0004984208731090569,
"loss": 5.6797,
"mean_token_accuracy": 0.14077720865607263,
"num_tokens": 10470257.0,
"step": 4765
},
{
"entropy": 5.776945304870606,
"epoch": 0.4734021437078206,
"grad_norm": 0.8359375,
"learning_rate": 0.0004984166799107587,
"loss": 5.6371,
"mean_token_accuracy": 0.14257957339286803,
"num_tokens": 10480835.0,
"step": 4770
},
{
"entropy": 5.759775972366333,
"epoch": 0.4738983723699881,
"grad_norm": 0.93359375,
"learning_rate": 0.0004984124811722081,
"loss": 5.6488,
"mean_token_accuracy": 0.14650474786758422,
"num_tokens": 10491720.0,
"step": 4775
},
{
"entropy": 5.753606271743775,
"epoch": 0.47439460103215564,
"grad_norm": 0.85546875,
"learning_rate": 0.0004984082768935096,
"loss": 5.6651,
"mean_token_accuracy": 0.14373267889022828,
"num_tokens": 10502313.0,
"step": 4780
},
{
"entropy": 5.78562560081482,
"epoch": 0.47489082969432317,
"grad_norm": 0.86328125,
"learning_rate": 0.0004984040670747672,
"loss": 5.54,
"mean_token_accuracy": 0.14707302302122116,
"num_tokens": 10513333.0,
"step": 4785
},
{
"entropy": 5.711219787597656,
"epoch": 0.47538705835649064,
"grad_norm": 0.92578125,
"learning_rate": 0.0004983998517160852,
"loss": 5.5278,
"mean_token_accuracy": 0.15109342858195304,
"num_tokens": 10522766.0,
"step": 4790
},
{
"entropy": 5.755279207229615,
"epoch": 0.4758832870186582,
"grad_norm": 0.8359375,
"learning_rate": 0.0004983956308175684,
"loss": 5.5675,
"mean_token_accuracy": 0.14042461812496185,
"num_tokens": 10534362.0,
"step": 4795
},
{
"entropy": 5.7386070728302006,
"epoch": 0.4763795156808257,
"grad_norm": 0.83984375,
"learning_rate": 0.0004983914043793212,
"loss": 5.5811,
"mean_token_accuracy": 0.14227042123675346,
"num_tokens": 10545072.0,
"step": 4800
},
{
"entropy": 5.673999500274658,
"epoch": 0.47687574434299324,
"grad_norm": 0.796875,
"learning_rate": 0.0004983871724014487,
"loss": 5.5169,
"mean_token_accuracy": 0.14101085662841797,
"num_tokens": 10555953.0,
"step": 4805
},
{
"entropy": 5.778817129135132,
"epoch": 0.47737197300516077,
"grad_norm": 0.77734375,
"learning_rate": 0.0004983829348840556,
"loss": 5.5675,
"mean_token_accuracy": 0.14799523502588272,
"num_tokens": 10567757.0,
"step": 4810
},
{
"entropy": 5.668074464797973,
"epoch": 0.4778682016673283,
"grad_norm": 0.82421875,
"learning_rate": 0.000498378691827247,
"loss": 5.4701,
"mean_token_accuracy": 0.1479513630270958,
"num_tokens": 10579390.0,
"step": 4815
},
{
"entropy": 5.742642593383789,
"epoch": 0.4783644303294958,
"grad_norm": 0.90234375,
"learning_rate": 0.0004983744432311282,
"loss": 5.6669,
"mean_token_accuracy": 0.13707347735762596,
"num_tokens": 10589624.0,
"step": 4820
},
{
"entropy": 5.829803085327148,
"epoch": 0.47886065899166336,
"grad_norm": 0.796875,
"learning_rate": 0.0004983701890958046,
"loss": 5.6361,
"mean_token_accuracy": 0.13275278136134147,
"num_tokens": 10601927.0,
"step": 4825
},
{
"entropy": 5.667952632904052,
"epoch": 0.4793568876538309,
"grad_norm": 0.8359375,
"learning_rate": 0.0004983659294213815,
"loss": 5.5147,
"mean_token_accuracy": 0.1479553312063217,
"num_tokens": 10614073.0,
"step": 4830
},
{
"entropy": 5.720580911636352,
"epoch": 0.4798531163159984,
"grad_norm": 0.8984375,
"learning_rate": 0.0004983616642079649,
"loss": 5.5511,
"mean_token_accuracy": 0.1447890058159828,
"num_tokens": 10625535.0,
"step": 4835
},
{
"entropy": 5.812899208068847,
"epoch": 0.48034934497816595,
"grad_norm": 0.91796875,
"learning_rate": 0.0004983573934556601,
"loss": 5.5241,
"mean_token_accuracy": 0.1474653646349907,
"num_tokens": 10637160.0,
"step": 4840
},
{
"entropy": 5.780987215042114,
"epoch": 0.4808455736403335,
"grad_norm": 0.88671875,
"learning_rate": 0.0004983531171645734,
"loss": 5.6279,
"mean_token_accuracy": 0.14129644334316255,
"num_tokens": 10649061.0,
"step": 4845
},
{
"entropy": 5.638265609741211,
"epoch": 0.481341802302501,
"grad_norm": 0.859375,
"learning_rate": 0.0004983488353348106,
"loss": 5.4829,
"mean_token_accuracy": 0.14804023057222365,
"num_tokens": 10661130.0,
"step": 4850
},
{
"entropy": 5.721083116531372,
"epoch": 0.48183803096466854,
"grad_norm": 0.828125,
"learning_rate": 0.0004983445479664779,
"loss": 5.5476,
"mean_token_accuracy": 0.1464819297194481,
"num_tokens": 10673017.0,
"step": 4855
},
{
"entropy": 5.7977289199829105,
"epoch": 0.48233425962683607,
"grad_norm": 0.8828125,
"learning_rate": 0.0004983402550596817,
"loss": 5.6084,
"mean_token_accuracy": 0.14504522681236268,
"num_tokens": 10683268.0,
"step": 4860
},
{
"entropy": 5.833132791519165,
"epoch": 0.48283048828900355,
"grad_norm": 0.85546875,
"learning_rate": 0.0004983359566145285,
"loss": 5.6222,
"mean_token_accuracy": 0.14250127971172333,
"num_tokens": 10694220.0,
"step": 4865
},
{
"entropy": 5.747547292709351,
"epoch": 0.4833267169511711,
"grad_norm": 0.8671875,
"learning_rate": 0.0004983316526311248,
"loss": 5.5401,
"mean_token_accuracy": 0.14009479209780692,
"num_tokens": 10704829.0,
"step": 4870
},
{
"entropy": 5.74469518661499,
"epoch": 0.4838229456133386,
"grad_norm": 0.95703125,
"learning_rate": 0.0004983273431095773,
"loss": 5.7219,
"mean_token_accuracy": 0.13392212018370628,
"num_tokens": 10717160.0,
"step": 4875
},
{
"entropy": 5.78354058265686,
"epoch": 0.48431917427550614,
"grad_norm": 0.9375,
"learning_rate": 0.0004983230280499929,
"loss": 5.5425,
"mean_token_accuracy": 0.14195659756660461,
"num_tokens": 10728638.0,
"step": 4880
},
{
"entropy": 5.737115001678466,
"epoch": 0.48481540293767367,
"grad_norm": 0.8515625,
"learning_rate": 0.0004983187074524787,
"loss": 5.5888,
"mean_token_accuracy": 0.13973118513822555,
"num_tokens": 10740122.0,
"step": 4885
},
{
"entropy": 5.766622018814087,
"epoch": 0.4853116315998412,
"grad_norm": 0.82421875,
"learning_rate": 0.0004983143813171418,
"loss": 5.5563,
"mean_token_accuracy": 0.1492931678891182,
"num_tokens": 10751103.0,
"step": 4890
},
{
"entropy": 5.706625509262085,
"epoch": 0.48580786026200873,
"grad_norm": 0.859375,
"learning_rate": 0.0004983100496440895,
"loss": 5.5063,
"mean_token_accuracy": 0.1490258678793907,
"num_tokens": 10761937.0,
"step": 4895
},
{
"entropy": 5.796269845962525,
"epoch": 0.48630408892417626,
"grad_norm": 0.87109375,
"learning_rate": 0.000498305712433429,
"loss": 5.7346,
"mean_token_accuracy": 0.14223510399460793,
"num_tokens": 10773527.0,
"step": 4900
},
{
"entropy": 5.765747499465943,
"epoch": 0.4868003175863438,
"grad_norm": 0.828125,
"learning_rate": 0.000498301369685268,
"loss": 5.5754,
"mean_token_accuracy": 0.14397116079926492,
"num_tokens": 10783873.0,
"step": 4905
},
{
"entropy": 5.6565389156341555,
"epoch": 0.4872965462485113,
"grad_norm": 0.84765625,
"learning_rate": 0.0004982970213997142,
"loss": 5.4238,
"mean_token_accuracy": 0.15091024041175843,
"num_tokens": 10796124.0,
"step": 4910
},
{
"entropy": 5.614575958251953,
"epoch": 0.48779277491067885,
"grad_norm": 0.82421875,
"learning_rate": 0.0004982926675768756,
"loss": 5.5215,
"mean_token_accuracy": 0.14468540474772454,
"num_tokens": 10806468.0,
"step": 4915
},
{
"entropy": 5.737712860107422,
"epoch": 0.4882890035728464,
"grad_norm": 0.82421875,
"learning_rate": 0.0004982883082168599,
"loss": 5.5778,
"mean_token_accuracy": 0.14341849982738494,
"num_tokens": 10817126.0,
"step": 4920
},
{
"entropy": 5.780298519134521,
"epoch": 0.4887852322350139,
"grad_norm": 0.87109375,
"learning_rate": 0.0004982839433197752,
"loss": 5.597,
"mean_token_accuracy": 0.13673522770404817,
"num_tokens": 10828985.0,
"step": 4925
},
{
"entropy": 5.686007690429688,
"epoch": 0.48928146089718144,
"grad_norm": 0.8125,
"learning_rate": 0.00049827957288573,
"loss": 5.5045,
"mean_token_accuracy": 0.14484182596206666,
"num_tokens": 10840528.0,
"step": 4930
},
{
"entropy": 5.708341026306153,
"epoch": 0.48977768955934897,
"grad_norm": 0.88671875,
"learning_rate": 0.0004982751969148324,
"loss": 5.5301,
"mean_token_accuracy": 0.14575338512659072,
"num_tokens": 10851312.0,
"step": 4935
},
{
"entropy": 5.67660961151123,
"epoch": 0.49027391822151645,
"grad_norm": 0.85546875,
"learning_rate": 0.0004982708154071912,
"loss": 5.4893,
"mean_token_accuracy": 0.1472529575228691,
"num_tokens": 10861882.0,
"step": 4940
},
{
"entropy": 5.73410964012146,
"epoch": 0.490770146883684,
"grad_norm": 0.8671875,
"learning_rate": 0.0004982664283629147,
"loss": 5.6418,
"mean_token_accuracy": 0.135673226416111,
"num_tokens": 10873187.0,
"step": 4945
},
{
"entropy": 5.750768995285034,
"epoch": 0.4912663755458515,
"grad_norm": 0.8984375,
"learning_rate": 0.0004982620357821118,
"loss": 5.529,
"mean_token_accuracy": 0.1487978346645832,
"num_tokens": 10884254.0,
"step": 4950
},
{
"entropy": 5.62707486152649,
"epoch": 0.49176260420801904,
"grad_norm": 0.84765625,
"learning_rate": 0.0004982576376648918,
"loss": 5.5165,
"mean_token_accuracy": 0.14257091730833055,
"num_tokens": 10895770.0,
"step": 4955
},
{
"entropy": 5.730557155609131,
"epoch": 0.49225883287018657,
"grad_norm": 0.79296875,
"learning_rate": 0.0004982532340113633,
"loss": 5.5434,
"mean_token_accuracy": 0.14722897857427597,
"num_tokens": 10906436.0,
"step": 4960
},
{
"entropy": 5.758408737182617,
"epoch": 0.4927550615323541,
"grad_norm": 0.83984375,
"learning_rate": 0.0004982488248216357,
"loss": 5.4945,
"mean_token_accuracy": 0.14825854152441026,
"num_tokens": 10917655.0,
"step": 4965
},
{
"entropy": 5.69063892364502,
"epoch": 0.49325129019452163,
"grad_norm": 0.87890625,
"learning_rate": 0.0004982444100958182,
"loss": 5.515,
"mean_token_accuracy": 0.1498719871044159,
"num_tokens": 10928359.0,
"step": 4970
},
{
"entropy": 5.748080778121948,
"epoch": 0.49374751885668916,
"grad_norm": 0.8125,
"learning_rate": 0.0004982399898340205,
"loss": 5.6798,
"mean_token_accuracy": 0.1363233283162117,
"num_tokens": 10940494.0,
"step": 4975
},
{
"entropy": 5.799513626098633,
"epoch": 0.4942437475188567,
"grad_norm": 0.828125,
"learning_rate": 0.000498235564036352,
"loss": 5.6694,
"mean_token_accuracy": 0.13602429032325744,
"num_tokens": 10951298.0,
"step": 4980
},
{
"entropy": 5.7167809963226315,
"epoch": 0.4947399761810242,
"grad_norm": 0.90234375,
"learning_rate": 0.0004982311327029226,
"loss": 5.565,
"mean_token_accuracy": 0.14564179703593255,
"num_tokens": 10961680.0,
"step": 4985
},
{
"entropy": 5.820360136032105,
"epoch": 0.49523620484319175,
"grad_norm": 1.1875,
"learning_rate": 0.0004982266958338421,
"loss": 5.6719,
"mean_token_accuracy": 0.14173018112778663,
"num_tokens": 10974968.0,
"step": 4990
},
{
"entropy": 5.712174558639527,
"epoch": 0.4957324335053593,
"grad_norm": 0.82421875,
"learning_rate": 0.0004982222534292205,
"loss": 5.6018,
"mean_token_accuracy": 0.13866912946105003,
"num_tokens": 10986383.0,
"step": 4995
},
{
"entropy": 5.779634046554565,
"epoch": 0.4962286621675268,
"grad_norm": 0.83203125,
"learning_rate": 0.0004982178054891681,
"loss": 5.6601,
"mean_token_accuracy": 0.14250716120004653,
"num_tokens": 10998384.0,
"step": 5000
},
{
"entropy": 5.722892570495605,
"epoch": 0.49672489082969434,
"grad_norm": 0.87109375,
"learning_rate": 0.0004982133520137951,
"loss": 5.4276,
"mean_token_accuracy": 0.15285419523715973,
"num_tokens": 11009420.0,
"step": 5005
},
{
"entropy": 5.763294076919555,
"epoch": 0.4972211194918619,
"grad_norm": 0.828125,
"learning_rate": 0.0004982088930032119,
"loss": 5.5968,
"mean_token_accuracy": 0.1417885646224022,
"num_tokens": 11019601.0,
"step": 5010
},
{
"entropy": 5.721580982208252,
"epoch": 0.4977173481540294,
"grad_norm": 0.8515625,
"learning_rate": 0.0004982044284575293,
"loss": 5.5333,
"mean_token_accuracy": 0.1445724494755268,
"num_tokens": 11031138.0,
"step": 5015
},
{
"entropy": 5.723393726348877,
"epoch": 0.4982135768161969,
"grad_norm": 0.8359375,
"learning_rate": 0.0004981999583768576,
"loss": 5.5418,
"mean_token_accuracy": 0.14509216398000718,
"num_tokens": 11041643.0,
"step": 5020
},
{
"entropy": 5.676524686813354,
"epoch": 0.4987098054783644,
"grad_norm": 0.84765625,
"learning_rate": 0.000498195482761308,
"loss": 5.5537,
"mean_token_accuracy": 0.13784084841609,
"num_tokens": 11052116.0,
"step": 5025
},
{
"entropy": 5.821243333816528,
"epoch": 0.49920603414053194,
"grad_norm": 0.79296875,
"learning_rate": 0.0004981910016109914,
"loss": 5.734,
"mean_token_accuracy": 0.136544182151556,
"num_tokens": 11064796.0,
"step": 5030
},
{
"entropy": 5.8484121799469,
"epoch": 0.49970226280269947,
"grad_norm": 0.890625,
"learning_rate": 0.0004981865149260189,
"loss": 5.5833,
"mean_token_accuracy": 0.14325116872787474,
"num_tokens": 11075029.0,
"step": 5035
},
{
"entropy": 5.731780004501343,
"epoch": 0.5001984914648671,
"grad_norm": 0.8046875,
"learning_rate": 0.0004981820227065016,
"loss": 5.6085,
"mean_token_accuracy": 0.14171847850084304,
"num_tokens": 11086702.0,
"step": 5040
},
{
"entropy": 5.700930786132813,
"epoch": 0.5006947201270345,
"grad_norm": 0.890625,
"learning_rate": 0.0004981775249525513,
"loss": 5.5816,
"mean_token_accuracy": 0.1423800431191921,
"num_tokens": 11097417.0,
"step": 5045
},
{
"entropy": 5.67112922668457,
"epoch": 0.5011909487892021,
"grad_norm": 0.8828125,
"learning_rate": 0.0004981730216642791,
"loss": 5.4835,
"mean_token_accuracy": 0.14785366728901864,
"num_tokens": 11108458.0,
"step": 5050
},
{
"entropy": 5.775296688079834,
"epoch": 0.5016871774513696,
"grad_norm": 0.78125,
"learning_rate": 0.0004981685128417969,
"loss": 5.6521,
"mean_token_accuracy": 0.1389369361102581,
"num_tokens": 11121466.0,
"step": 5055
},
{
"entropy": 5.7190674304962155,
"epoch": 0.5021834061135371,
"grad_norm": 0.87109375,
"learning_rate": 0.0004981639984852165,
"loss": 5.5956,
"mean_token_accuracy": 0.13504938408732414,
"num_tokens": 11132489.0,
"step": 5060
},
{
"entropy": 5.695435190200806,
"epoch": 0.5026796347757047,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981594785946498,
"loss": 5.5532,
"mean_token_accuracy": 0.14500434398651124,
"num_tokens": 11143087.0,
"step": 5065
},
{
"entropy": 5.7991992950439455,
"epoch": 0.5031758634378721,
"grad_norm": 0.83984375,
"learning_rate": 0.0004981549531702089,
"loss": 5.5835,
"mean_token_accuracy": 0.1371206223964691,
"num_tokens": 11154229.0,
"step": 5070
},
{
"entropy": 5.748044872283936,
"epoch": 0.5036720921000397,
"grad_norm": 0.8046875,
"learning_rate": 0.000498150422212006,
"loss": 5.5454,
"mean_token_accuracy": 0.143742111325264,
"num_tokens": 11165698.0,
"step": 5075
},
{
"entropy": 5.725100755691528,
"epoch": 0.5041683207622072,
"grad_norm": 0.859375,
"learning_rate": 0.0004981458857201534,
"loss": 5.5462,
"mean_token_accuracy": 0.14240316525101662,
"num_tokens": 11177025.0,
"step": 5080
},
{
"entropy": 5.78506989479065,
"epoch": 0.5046645494243748,
"grad_norm": 1.0546875,
"learning_rate": 0.0004981413436947639,
"loss": 5.6209,
"mean_token_accuracy": 0.14323396608233452,
"num_tokens": 11188766.0,
"step": 5085
},
{
"entropy": 5.720196533203125,
"epoch": 0.5051607780865423,
"grad_norm": 0.8203125,
"learning_rate": 0.0004981367961359496,
"loss": 5.6546,
"mean_token_accuracy": 0.13432737216353416,
"num_tokens": 11201626.0,
"step": 5090
},
{
"entropy": 5.732619857788086,
"epoch": 0.5056570067487098,
"grad_norm": 0.8671875,
"learning_rate": 0.0004981322430438239,
"loss": 5.564,
"mean_token_accuracy": 0.13936000987887381,
"num_tokens": 11213975.0,
"step": 5095
},
{
"entropy": 5.688735437393189,
"epoch": 0.5061532354108773,
"grad_norm": 0.90234375,
"learning_rate": 0.0004981276844184991,
"loss": 5.5301,
"mean_token_accuracy": 0.14741144627332686,
"num_tokens": 11223636.0,
"step": 5100
},
{
"entropy": 5.6913628578186035,
"epoch": 0.5066494640730449,
"grad_norm": 0.96484375,
"learning_rate": 0.0004981231202600886,
"loss": 5.497,
"mean_token_accuracy": 0.14770336896181108,
"num_tokens": 11233549.0,
"step": 5105
},
{
"entropy": 5.639012718200684,
"epoch": 0.5071456927352124,
"grad_norm": 0.859375,
"learning_rate": 0.0004981185505687056,
"loss": 5.5018,
"mean_token_accuracy": 0.14809941947460176,
"num_tokens": 11242915.0,
"step": 5110
},
{
"entropy": 5.788972234725952,
"epoch": 0.50764192139738,
"grad_norm": 0.80078125,
"learning_rate": 0.0004981139753444632,
"loss": 5.6213,
"mean_token_accuracy": 0.13717583566904068,
"num_tokens": 11254370.0,
"step": 5115
},
{
"entropy": 5.745554065704345,
"epoch": 0.5081381500595474,
"grad_norm": 0.984375,
"learning_rate": 0.0004981093945874752,
"loss": 5.5478,
"mean_token_accuracy": 0.15008534342050553,
"num_tokens": 11264226.0,
"step": 5120
},
{
"entropy": 5.689509296417237,
"epoch": 0.508634378721715,
"grad_norm": 0.88671875,
"learning_rate": 0.0004981048082978547,
"loss": 5.4536,
"mean_token_accuracy": 0.1530928671360016,
"num_tokens": 11275304.0,
"step": 5125
},
{
"entropy": 5.705717992782593,
"epoch": 0.5091306073838825,
"grad_norm": 0.8515625,
"learning_rate": 0.0004981002164757158,
"loss": 5.5157,
"mean_token_accuracy": 0.14436992928385733,
"num_tokens": 11286179.0,
"step": 5130
},
{
"entropy": 5.761962938308716,
"epoch": 0.50962683604605,
"grad_norm": 0.90625,
"learning_rate": 0.0004980956191211723,
"loss": 5.5448,
"mean_token_accuracy": 0.14527795910835267,
"num_tokens": 11297244.0,
"step": 5135
},
{
"entropy": 5.739287281036377,
"epoch": 0.5101230647082176,
"grad_norm": 0.859375,
"learning_rate": 0.0004980910162343382,
"loss": 5.5531,
"mean_token_accuracy": 0.14658074975013732,
"num_tokens": 11307251.0,
"step": 5140
},
{
"entropy": 5.726897239685059,
"epoch": 0.510619293370385,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980864078153275,
"loss": 5.5294,
"mean_token_accuracy": 0.1411370761692524,
"num_tokens": 11318441.0,
"step": 5145
},
{
"entropy": 5.605728054046631,
"epoch": 0.5111155220325526,
"grad_norm": 0.93359375,
"learning_rate": 0.0004980817938642548,
"loss": 5.497,
"mean_token_accuracy": 0.1480163112282753,
"num_tokens": 11328743.0,
"step": 5150
},
{
"entropy": 5.608948802947998,
"epoch": 0.5116117506947201,
"grad_norm": 0.890625,
"learning_rate": 0.000498077174381234,
"loss": 5.4985,
"mean_token_accuracy": 0.148304795473814,
"num_tokens": 11340745.0,
"step": 5155
},
{
"entropy": 5.767784595489502,
"epoch": 0.5121079793568877,
"grad_norm": 0.9453125,
"learning_rate": 0.0004980725493663802,
"loss": 5.5726,
"mean_token_accuracy": 0.14685158357024192,
"num_tokens": 11352060.0,
"step": 5160
},
{
"entropy": 5.702456951141357,
"epoch": 0.5126042080190552,
"grad_norm": 0.9453125,
"learning_rate": 0.0004980679188198078,
"loss": 5.5702,
"mean_token_accuracy": 0.14360618591308594,
"num_tokens": 11363097.0,
"step": 5165
},
{
"entropy": 5.721812438964844,
"epoch": 0.5131004366812227,
"grad_norm": 0.875,
"learning_rate": 0.0004980632827416317,
"loss": 5.602,
"mean_token_accuracy": 0.15401604920625686,
"num_tokens": 11373410.0,
"step": 5170
},
{
"entropy": 5.758921146392822,
"epoch": 0.5135966653433902,
"grad_norm": 0.859375,
"learning_rate": 0.0004980586411319669,
"loss": 5.6816,
"mean_token_accuracy": 0.13971229791641235,
"num_tokens": 11385615.0,
"step": 5175
},
{
"entropy": 5.666999673843383,
"epoch": 0.5140928940055578,
"grad_norm": 0.76953125,
"learning_rate": 0.0004980539939909283,
"loss": 5.4783,
"mean_token_accuracy": 0.14054736942052842,
"num_tokens": 11396492.0,
"step": 5180
},
{
"entropy": 5.781865406036377,
"epoch": 0.5145891226677253,
"grad_norm": 0.91796875,
"learning_rate": 0.0004980493413186314,
"loss": 5.5828,
"mean_token_accuracy": 0.13886439129710199,
"num_tokens": 11406377.0,
"step": 5185
},
{
"entropy": 5.740067386627198,
"epoch": 0.5150853513298929,
"grad_norm": 0.91796875,
"learning_rate": 0.0004980446831151914,
"loss": 5.5575,
"mean_token_accuracy": 0.1424478217959404,
"num_tokens": 11416862.0,
"step": 5190
},
{
"entropy": 5.631788444519043,
"epoch": 0.5155815799920603,
"grad_norm": 0.921875,
"learning_rate": 0.0004980400193807239,
"loss": 5.5077,
"mean_token_accuracy": 0.14980673640966416,
"num_tokens": 11428246.0,
"step": 5195
},
{
"entropy": 5.748811864852906,
"epoch": 0.5160778086542279,
"grad_norm": 0.88671875,
"learning_rate": 0.0004980353501153446,
"loss": 5.5369,
"mean_token_accuracy": 0.14516637101769447,
"num_tokens": 11438147.0,
"step": 5200
},
{
"entropy": 5.762879467010498,
"epoch": 0.5165740373163954,
"grad_norm": 0.859375,
"learning_rate": 0.0004980306753191691,
"loss": 5.6576,
"mean_token_accuracy": 0.13611230552196502,
"num_tokens": 11449154.0,
"step": 5205
},
{
"entropy": 5.747657585144043,
"epoch": 0.5170702659785629,
"grad_norm": 0.87109375,
"learning_rate": 0.0004980259949923135,
"loss": 5.5571,
"mean_token_accuracy": 0.14405836015939713,
"num_tokens": 11459262.0,
"step": 5210
},
{
"entropy": 5.719294309616089,
"epoch": 0.5175664946407305,
"grad_norm": 0.8359375,
"learning_rate": 0.0004980213091348938,
"loss": 5.5391,
"mean_token_accuracy": 0.14385731667280197,
"num_tokens": 11471991.0,
"step": 5215
},
{
"entropy": 5.700556945800781,
"epoch": 0.5180627233028979,
"grad_norm": 0.9453125,
"learning_rate": 0.0004980166177470262,
"loss": 5.4846,
"mean_token_accuracy": 0.14262256026268005,
"num_tokens": 11483176.0,
"step": 5220
},
{
"entropy": 5.722855520248413,
"epoch": 0.5185589519650655,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980119208288271,
"loss": 5.5731,
"mean_token_accuracy": 0.14170767962932587,
"num_tokens": 11494015.0,
"step": 5225
},
{
"entropy": 5.6596784591674805,
"epoch": 0.519055180627233,
"grad_norm": 0.7421875,
"learning_rate": 0.0004980072183804128,
"loss": 5.4895,
"mean_token_accuracy": 0.1490239754319191,
"num_tokens": 11506365.0,
"step": 5230
},
{
"entropy": 5.706877708435059,
"epoch": 0.5195514092894006,
"grad_norm": 0.875,
"learning_rate": 0.0004980025104019,
"loss": 5.5856,
"mean_token_accuracy": 0.15004663541913033,
"num_tokens": 11517199.0,
"step": 5235
},
{
"entropy": 5.682269048690796,
"epoch": 0.520047637951568,
"grad_norm": 0.82421875,
"learning_rate": 0.0004979977968934056,
"loss": 5.4337,
"mean_token_accuracy": 0.15386080890893936,
"num_tokens": 11527591.0,
"step": 5240
},
{
"entropy": 5.74504714012146,
"epoch": 0.5205438666137356,
"grad_norm": 0.953125,
"learning_rate": 0.0004979930778550463,
"loss": 5.6126,
"mean_token_accuracy": 0.14512009173631668,
"num_tokens": 11538657.0,
"step": 5245
},
{
"entropy": 5.662412548065186,
"epoch": 0.5210400952759031,
"grad_norm": 0.94921875,
"learning_rate": 0.0004979883532869392,
"loss": 5.4288,
"mean_token_accuracy": 0.16117970496416092,
"num_tokens": 11549513.0,
"step": 5250
},
{
"entropy": 5.721430921554566,
"epoch": 0.5215363239380707,
"grad_norm": 0.88671875,
"learning_rate": 0.0004979836231892013,
"loss": 5.5206,
"mean_token_accuracy": 0.1425759568810463,
"num_tokens": 11560376.0,
"step": 5255
},
{
"entropy": 5.672984027862549,
"epoch": 0.5220325526002382,
"grad_norm": 0.875,
"learning_rate": 0.0004979788875619501,
"loss": 5.4948,
"mean_token_accuracy": 0.14300124421715737,
"num_tokens": 11571218.0,
"step": 5260
},
{
"entropy": 5.7140038967132565,
"epoch": 0.5225287812624058,
"grad_norm": 0.828125,
"learning_rate": 0.000497974146405303,
"loss": 5.5693,
"mean_token_accuracy": 0.14280975982546806,
"num_tokens": 11581981.0,
"step": 5265
},
{
"entropy": 5.675239849090576,
"epoch": 0.5230250099245732,
"grad_norm": 0.8984375,
"learning_rate": 0.0004979693997193776,
"loss": 5.5226,
"mean_token_accuracy": 0.14427190124988556,
"num_tokens": 11592561.0,
"step": 5270
},
{
"entropy": 5.732921123504639,
"epoch": 0.5235212385867408,
"grad_norm": 0.9375,
"learning_rate": 0.0004979646475042914,
"loss": 5.4962,
"mean_token_accuracy": 0.14883433431386947,
"num_tokens": 11601933.0,
"step": 5275
},
{
"entropy": 5.660780525207519,
"epoch": 0.5240174672489083,
"grad_norm": 0.8984375,
"learning_rate": 0.0004979598897601625,
"loss": 5.4274,
"mean_token_accuracy": 0.15457454025745393,
"num_tokens": 11613938.0,
"step": 5280
},
{
"entropy": 5.766283369064331,
"epoch": 0.5245136959110758,
"grad_norm": 0.90625,
"learning_rate": 0.0004979551264871088,
"loss": 5.6035,
"mean_token_accuracy": 0.13649525493383408,
"num_tokens": 11624890.0,
"step": 5285
},
{
"entropy": 5.738007640838623,
"epoch": 0.5250099245732434,
"grad_norm": 0.8359375,
"learning_rate": 0.0004979503576852482,
"loss": 5.6069,
"mean_token_accuracy": 0.14732139855623244,
"num_tokens": 11635080.0,
"step": 5290
},
{
"entropy": 5.656884241104126,
"epoch": 0.5255061532354108,
"grad_norm": 0.90234375,
"learning_rate": 0.0004979455833546994,
"loss": 5.5011,
"mean_token_accuracy": 0.14415451809763907,
"num_tokens": 11645180.0,
"step": 5295
},
{
"entropy": 5.749808835983276,
"epoch": 0.5260023818975784,
"grad_norm": 0.8671875,
"learning_rate": 0.0004979408034955804,
"loss": 5.5431,
"mean_token_accuracy": 0.14449771195650102,
"num_tokens": 11657599.0,
"step": 5300
},
{
"entropy": 5.738276576995849,
"epoch": 0.5264986105597459,
"grad_norm": 0.859375,
"learning_rate": 0.0004979360181080099,
"loss": 5.5817,
"mean_token_accuracy": 0.13297581300139427,
"num_tokens": 11669177.0,
"step": 5305
},
{
"entropy": 5.658540487289429,
"epoch": 0.5269948392219135,
"grad_norm": 0.87890625,
"learning_rate": 0.0004979312271921065,
"loss": 5.4448,
"mean_token_accuracy": 0.14841168075799943,
"num_tokens": 11679756.0,
"step": 5310
},
{
"entropy": 5.681262445449829,
"epoch": 0.527491067884081,
"grad_norm": 0.8359375,
"learning_rate": 0.000497926430747989,
"loss": 5.5043,
"mean_token_accuracy": 0.1484253525733948,
"num_tokens": 11689805.0,
"step": 5315
},
{
"entropy": 5.6945432186126705,
"epoch": 0.5279872965462485,
"grad_norm": 0.8203125,
"learning_rate": 0.0004979216287757764,
"loss": 5.5277,
"mean_token_accuracy": 0.14734025299549103,
"num_tokens": 11701150.0,
"step": 5320
},
{
"entropy": 5.623287963867187,
"epoch": 0.528483525208416,
"grad_norm": 0.83984375,
"learning_rate": 0.000497916821275588,
"loss": 5.4639,
"mean_token_accuracy": 0.1452282652258873,
"num_tokens": 11712256.0,
"step": 5325
},
{
"entropy": 5.703384447097778,
"epoch": 0.5289797538705836,
"grad_norm": 0.83984375,
"learning_rate": 0.0004979120082475424,
"loss": 5.5123,
"mean_token_accuracy": 0.14741147682070732,
"num_tokens": 11722547.0,
"step": 5330
},
{
"entropy": 5.716305112838745,
"epoch": 0.5294759825327511,
"grad_norm": 0.8828125,
"learning_rate": 0.0004979071896917595,
"loss": 5.522,
"mean_token_accuracy": 0.1516739696264267,
"num_tokens": 11732309.0,
"step": 5335
},
{
"entropy": 5.720453691482544,
"epoch": 0.5299722111949187,
"grad_norm": 0.89453125,
"learning_rate": 0.0004979023656083586,
"loss": 5.5843,
"mean_token_accuracy": 0.14252596721053123,
"num_tokens": 11743982.0,
"step": 5340
},
{
"entropy": 5.66952691078186,
"epoch": 0.5304684398570861,
"grad_norm": 0.79296875,
"learning_rate": 0.0004978975359974594,
"loss": 5.5189,
"mean_token_accuracy": 0.14486275911331176,
"num_tokens": 11755506.0,
"step": 5345
},
{
"entropy": 5.751617050170898,
"epoch": 0.5309646685192537,
"grad_norm": 0.8671875,
"learning_rate": 0.0004978927008591816,
"loss": 5.5216,
"mean_token_accuracy": 0.14368319660425186,
"num_tokens": 11767024.0,
"step": 5350
},
{
"entropy": 5.774104166030884,
"epoch": 0.5314608971814212,
"grad_norm": 0.9375,
"learning_rate": 0.0004978878601936451,
"loss": 5.5214,
"mean_token_accuracy": 0.14008311927318573,
"num_tokens": 11777625.0,
"step": 5355
},
{
"entropy": 5.676749420166016,
"epoch": 0.5319571258435887,
"grad_norm": 0.84765625,
"learning_rate": 0.0004978830140009698,
"loss": 5.4998,
"mean_token_accuracy": 0.15135693401098252,
"num_tokens": 11788610.0,
"step": 5360
},
{
"entropy": 5.6927237033844,
"epoch": 0.5324533545057563,
"grad_norm": 0.88671875,
"learning_rate": 0.0004978781622812762,
"loss": 5.5998,
"mean_token_accuracy": 0.14594103246927262,
"num_tokens": 11800828.0,
"step": 5365
},
{
"entropy": 5.745589065551758,
"epoch": 0.5329495831679237,
"grad_norm": 0.78515625,
"learning_rate": 0.0004978733050346844,
"loss": 5.598,
"mean_token_accuracy": 0.14758748263120652,
"num_tokens": 11811729.0,
"step": 5370
},
{
"entropy": 5.737847137451172,
"epoch": 0.5334458118300913,
"grad_norm": 1.0546875,
"learning_rate": 0.0004978684422613149,
"loss": 5.6203,
"mean_token_accuracy": 0.14279971048235893,
"num_tokens": 11822081.0,
"step": 5375
},
{
"entropy": 5.704253196716309,
"epoch": 0.5339420404922588,
"grad_norm": 0.79296875,
"learning_rate": 0.0004978635739612882,
"loss": 5.4959,
"mean_token_accuracy": 0.1478683464229107,
"num_tokens": 11832984.0,
"step": 5380
},
{
"entropy": 5.69680380821228,
"epoch": 0.5344382691544264,
"grad_norm": 0.90625,
"learning_rate": 0.0004978587001347252,
"loss": 5.4676,
"mean_token_accuracy": 0.1478103816509247,
"num_tokens": 11842918.0,
"step": 5385
},
{
"entropy": 5.63281888961792,
"epoch": 0.5349344978165939,
"grad_norm": 0.82421875,
"learning_rate": 0.0004978538207817465,
"loss": 5.501,
"mean_token_accuracy": 0.1504437059164047,
"num_tokens": 11854364.0,
"step": 5390
},
{
"entropy": 5.684811115264893,
"epoch": 0.5354307264787614,
"grad_norm": 0.9609375,
"learning_rate": 0.0004978489359024734,
"loss": 5.5732,
"mean_token_accuracy": 0.1437056079506874,
"num_tokens": 11866400.0,
"step": 5395
},
{
"entropy": 5.774307823181152,
"epoch": 0.5359269551409289,
"grad_norm": 0.87890625,
"learning_rate": 0.0004978440454970268,
"loss": 5.5738,
"mean_token_accuracy": 0.14117259830236434,
"num_tokens": 11877094.0,
"step": 5400
},
{
"entropy": 5.687881755828857,
"epoch": 0.5364231838030965,
"grad_norm": 0.93359375,
"learning_rate": 0.0004978391495655281,
"loss": 5.5245,
"mean_token_accuracy": 0.14514778330922126,
"num_tokens": 11887289.0,
"step": 5405
},
{
"entropy": 5.739621353149414,
"epoch": 0.536919412465264,
"grad_norm": 0.9453125,
"learning_rate": 0.0004978342481080987,
"loss": 5.5079,
"mean_token_accuracy": 0.14592916518449783,
"num_tokens": 11897700.0,
"step": 5410
},
{
"entropy": 5.5940704345703125,
"epoch": 0.5374156411274316,
"grad_norm": 0.8984375,
"learning_rate": 0.0004978293411248601,
"loss": 5.4367,
"mean_token_accuracy": 0.15131850391626359,
"num_tokens": 11908623.0,
"step": 5415
},
{
"entropy": 5.679359340667725,
"epoch": 0.537911869789599,
"grad_norm": 0.93359375,
"learning_rate": 0.0004978244286159339,
"loss": 5.4525,
"mean_token_accuracy": 0.15489538311958312,
"num_tokens": 11917883.0,
"step": 5420
},
{
"entropy": 5.658584928512573,
"epoch": 0.5384080984517666,
"grad_norm": 0.71875,
"learning_rate": 0.0004978195105814421,
"loss": 5.4932,
"mean_token_accuracy": 0.1469782829284668,
"num_tokens": 11929145.0,
"step": 5425
},
{
"entropy": 5.692590570449829,
"epoch": 0.5389043271139341,
"grad_norm": 0.80859375,
"learning_rate": 0.0004978145870215065,
"loss": 5.4835,
"mean_token_accuracy": 0.1445659376680851,
"num_tokens": 11940073.0,
"step": 5430
},
{
"entropy": 5.716174697875976,
"epoch": 0.5394005557761016,
"grad_norm": 0.91015625,
"learning_rate": 0.0004978096579362493,
"loss": 5.6122,
"mean_token_accuracy": 0.14428975731134414,
"num_tokens": 11953140.0,
"step": 5435
},
{
"entropy": 5.725712585449219,
"epoch": 0.5398967844382692,
"grad_norm": 0.98046875,
"learning_rate": 0.0004978047233257927,
"loss": 5.5355,
"mean_token_accuracy": 0.1476729243993759,
"num_tokens": 11963853.0,
"step": 5440
},
{
"entropy": 5.696603202819825,
"epoch": 0.5403930131004366,
"grad_norm": 0.7734375,
"learning_rate": 0.0004977997831902591,
"loss": 5.4953,
"mean_token_accuracy": 0.14950083196163177,
"num_tokens": 11975469.0,
"step": 5445
},
{
"entropy": 5.667461633682251,
"epoch": 0.5408892417626042,
"grad_norm": 0.96484375,
"learning_rate": 0.0004977948375297709,
"loss": 5.4805,
"mean_token_accuracy": 0.1535550147294998,
"num_tokens": 11985227.0,
"step": 5450
},
{
"entropy": 5.554087400436401,
"epoch": 0.5413854704247717,
"grad_norm": 0.83203125,
"learning_rate": 0.0004977898863444509,
"loss": 5.4905,
"mean_token_accuracy": 0.1473379760980606,
"num_tokens": 11995459.0,
"step": 5455
},
{
"entropy": 5.694110012054443,
"epoch": 0.5418816990869393,
"grad_norm": 0.8671875,
"learning_rate": 0.0004977849296344217,
"loss": 5.5509,
"mean_token_accuracy": 0.14240046888589858,
"num_tokens": 12007705.0,
"step": 5460
},
{
"entropy": 5.6833288192749025,
"epoch": 0.5423779277491068,
"grad_norm": 0.78125,
"learning_rate": 0.0004977799673998063,
"loss": 5.4428,
"mean_token_accuracy": 0.14829470664262773,
"num_tokens": 12019729.0,
"step": 5465
},
{
"entropy": 5.694163846969604,
"epoch": 0.5428741564112743,
"grad_norm": 0.8515625,
"learning_rate": 0.0004977749996407278,
"loss": 5.5204,
"mean_token_accuracy": 0.14065904766321183,
"num_tokens": 12031009.0,
"step": 5470
},
{
"entropy": 5.71015944480896,
"epoch": 0.5433703850734418,
"grad_norm": 0.9296875,
"learning_rate": 0.0004977700263573094,
"loss": 5.4757,
"mean_token_accuracy": 0.149942646920681,
"num_tokens": 12041396.0,
"step": 5475
},
{
"entropy": 5.714011335372925,
"epoch": 0.5438666137356094,
"grad_norm": 0.8046875,
"learning_rate": 0.0004977650475496743,
"loss": 5.6327,
"mean_token_accuracy": 0.13937563449144363,
"num_tokens": 12053112.0,
"step": 5480
},
{
"entropy": 5.701058769226075,
"epoch": 0.5443628423977769,
"grad_norm": 0.76953125,
"learning_rate": 0.0004977600632179459,
"loss": 5.4853,
"mean_token_accuracy": 0.1432749643921852,
"num_tokens": 12064823.0,
"step": 5485
},
{
"entropy": 5.741170930862427,
"epoch": 0.5448590710599445,
"grad_norm": 0.82421875,
"learning_rate": 0.0004977550733622482,
"loss": 5.611,
"mean_token_accuracy": 0.1431275188922882,
"num_tokens": 12076656.0,
"step": 5490
},
{
"entropy": 5.702113914489746,
"epoch": 0.5453552997221119,
"grad_norm": 0.8984375,
"learning_rate": 0.0004977500779827045,
"loss": 5.417,
"mean_token_accuracy": 0.14641434401273729,
"num_tokens": 12086324.0,
"step": 5495
},
{
"entropy": 5.584799289703369,
"epoch": 0.5458515283842795,
"grad_norm": 0.78125,
"learning_rate": 0.0004977450770794389,
"loss": 5.4573,
"mean_token_accuracy": 0.14669612646102906,
"num_tokens": 12098076.0,
"step": 5500
},
{
"entropy": 5.658282661437989,
"epoch": 0.546347757046447,
"grad_norm": 0.734375,
"learning_rate": 0.0004977400706525755,
"loss": 5.5176,
"mean_token_accuracy": 0.14338865503668785,
"num_tokens": 12111282.0,
"step": 5505
},
{
"entropy": 5.67042498588562,
"epoch": 0.5468439857086145,
"grad_norm": 0.8671875,
"learning_rate": 0.0004977350587022381,
"loss": 5.4646,
"mean_token_accuracy": 0.15267725735902787,
"num_tokens": 12121932.0,
"step": 5510
},
{
"entropy": 5.663971471786499,
"epoch": 0.5473402143707821,
"grad_norm": 0.8671875,
"learning_rate": 0.0004977300412285513,
"loss": 5.4506,
"mean_token_accuracy": 0.1547014519572258,
"num_tokens": 12133040.0,
"step": 5515
},
{
"entropy": 5.7008575916290285,
"epoch": 0.5478364430329495,
"grad_norm": 0.87109375,
"learning_rate": 0.0004977250182316394,
"loss": 5.5286,
"mean_token_accuracy": 0.14612772613763808,
"num_tokens": 12142577.0,
"step": 5520
},
{
"entropy": 5.695830965042115,
"epoch": 0.5483326716951171,
"grad_norm": 0.875,
"learning_rate": 0.000497719989711627,
"loss": 5.4969,
"mean_token_accuracy": 0.1528421387076378,
"num_tokens": 12153941.0,
"step": 5525
},
{
"entropy": 5.6863785743713375,
"epoch": 0.5488289003572846,
"grad_norm": 0.84375,
"learning_rate": 0.0004977149556686387,
"loss": 5.5501,
"mean_token_accuracy": 0.14650570154190062,
"num_tokens": 12165347.0,
"step": 5530
},
{
"entropy": 5.599493026733398,
"epoch": 0.5493251290194522,
"grad_norm": 0.8515625,
"learning_rate": 0.0004977099161027996,
"loss": 5.4374,
"mean_token_accuracy": 0.14999375939369203,
"num_tokens": 12175705.0,
"step": 5535
},
{
"entropy": 5.706617259979248,
"epoch": 0.5498213576816197,
"grad_norm": 0.87109375,
"learning_rate": 0.0004977048710142342,
"loss": 5.4914,
"mean_token_accuracy": 0.1500910520553589,
"num_tokens": 12186781.0,
"step": 5540
},
{
"entropy": 5.7512706279754635,
"epoch": 0.5503175863437872,
"grad_norm": 0.91796875,
"learning_rate": 0.0004976998204030681,
"loss": 5.5249,
"mean_token_accuracy": 0.14498836621642114,
"num_tokens": 12197145.0,
"step": 5545
},
{
"entropy": 5.624058055877685,
"epoch": 0.5508138150059547,
"grad_norm": 0.9296875,
"learning_rate": 0.0004976947642694262,
"loss": 5.5192,
"mean_token_accuracy": 0.15011530891060829,
"num_tokens": 12207637.0,
"step": 5550
},
{
"entropy": 5.618358469009399,
"epoch": 0.5513100436681223,
"grad_norm": 0.90625,
"learning_rate": 0.0004976897026134341,
"loss": 5.4405,
"mean_token_accuracy": 0.15141630321741104,
"num_tokens": 12218156.0,
"step": 5555
},
{
"entropy": 5.689433479309082,
"epoch": 0.5518062723302898,
"grad_norm": 0.91015625,
"learning_rate": 0.0004976846354352172,
"loss": 5.528,
"mean_token_accuracy": 0.1447851613163948,
"num_tokens": 12229987.0,
"step": 5560
},
{
"entropy": 5.69376802444458,
"epoch": 0.5523025009924574,
"grad_norm": 0.85546875,
"learning_rate": 0.0004976795627349012,
"loss": 5.5116,
"mean_token_accuracy": 0.1453092396259308,
"num_tokens": 12240508.0,
"step": 5565
},
{
"entropy": 5.652896738052368,
"epoch": 0.5527987296546248,
"grad_norm": 0.85546875,
"learning_rate": 0.0004976744845126118,
"loss": 5.4503,
"mean_token_accuracy": 0.1521064132452011,
"num_tokens": 12250317.0,
"step": 5570
},
{
"entropy": 5.620663976669311,
"epoch": 0.5532949583167924,
"grad_norm": 0.8359375,
"learning_rate": 0.0004976694007684751,
"loss": 5.4953,
"mean_token_accuracy": 0.1509055018424988,
"num_tokens": 12262104.0,
"step": 5575
},
{
"entropy": 5.74803204536438,
"epoch": 0.5537911869789599,
"grad_norm": 0.8359375,
"learning_rate": 0.000497664311502617,
"loss": 5.5393,
"mean_token_accuracy": 0.1466676525771618,
"num_tokens": 12273259.0,
"step": 5580
},
{
"entropy": 5.736172199249268,
"epoch": 0.5542874156411274,
"grad_norm": 0.82421875,
"learning_rate": 0.0004976592167151638,
"loss": 5.5704,
"mean_token_accuracy": 0.13990923017263412,
"num_tokens": 12285504.0,
"step": 5585
},
{
"entropy": 5.692285537719727,
"epoch": 0.554783644303295,
"grad_norm": 0.984375,
"learning_rate": 0.0004976541164062418,
"loss": 5.4988,
"mean_token_accuracy": 0.14217816665768623,
"num_tokens": 12296512.0,
"step": 5590
},
{
"entropy": 5.577828121185303,
"epoch": 0.5552798729654624,
"grad_norm": 0.86328125,
"learning_rate": 0.0004976490105759774,
"loss": 5.3666,
"mean_token_accuracy": 0.15274341851472856,
"num_tokens": 12307948.0,
"step": 5595
},
{
"entropy": 5.6738968849182125,
"epoch": 0.55577610162763,
"grad_norm": 0.9140625,
"learning_rate": 0.0004976438992244976,
"loss": 5.5376,
"mean_token_accuracy": 0.14299441426992415,
"num_tokens": 12320020.0,
"step": 5600
},
{
"entropy": 5.689694452285766,
"epoch": 0.5562723302897975,
"grad_norm": 0.859375,
"learning_rate": 0.0004976387823519286,
"loss": 5.4035,
"mean_token_accuracy": 0.1494772896170616,
"num_tokens": 12331091.0,
"step": 5605
},
{
"entropy": 5.6960502624511715,
"epoch": 0.5567685589519651,
"grad_norm": 0.87890625,
"learning_rate": 0.0004976336599583975,
"loss": 5.5089,
"mean_token_accuracy": 0.14410399943590163,
"num_tokens": 12342621.0,
"step": 5610
},
{
"entropy": 5.788615989685058,
"epoch": 0.5572647876141326,
"grad_norm": 1.4296875,
"learning_rate": 0.0004976285320440315,
"loss": 5.7636,
"mean_token_accuracy": 0.13273412361741066,
"num_tokens": 12353330.0,
"step": 5615
},
{
"entropy": 5.751302766799927,
"epoch": 0.5577610162763001,
"grad_norm": 0.77734375,
"learning_rate": 0.0004976233986089577,
"loss": 5.4999,
"mean_token_accuracy": 0.14577182084321977,
"num_tokens": 12364418.0,
"step": 5620
},
{
"entropy": 5.610360383987427,
"epoch": 0.5582572449384676,
"grad_norm": 0.80078125,
"learning_rate": 0.0004976182596533031,
"loss": 5.4916,
"mean_token_accuracy": 0.1467949479818344,
"num_tokens": 12377022.0,
"step": 5625
},
{
"entropy": 5.739885330200195,
"epoch": 0.5587534736006352,
"grad_norm": 0.85546875,
"learning_rate": 0.0004976131151771955,
"loss": 5.548,
"mean_token_accuracy": 0.14481229856610298,
"num_tokens": 12387799.0,
"step": 5630
},
{
"entropy": 5.670582485198975,
"epoch": 0.5592497022628027,
"grad_norm": 0.86328125,
"learning_rate": 0.0004976079651807623,
"loss": 5.5007,
"mean_token_accuracy": 0.14010098576545715,
"num_tokens": 12397387.0,
"step": 5635
},
{
"entropy": 5.609793567657471,
"epoch": 0.5597459309249703,
"grad_norm": 0.8515625,
"learning_rate": 0.0004976028096641313,
"loss": 5.4277,
"mean_token_accuracy": 0.15353151261806489,
"num_tokens": 12407994.0,
"step": 5640
},
{
"entropy": 5.786797094345093,
"epoch": 0.5602421595871377,
"grad_norm": 0.83984375,
"learning_rate": 0.0004975976486274302,
"loss": 5.6054,
"mean_token_accuracy": 0.14164167046546935,
"num_tokens": 12419124.0,
"step": 5645
},
{
"entropy": 5.701025772094726,
"epoch": 0.5607383882493053,
"grad_norm": 0.86328125,
"learning_rate": 0.0004975924820707872,
"loss": 5.4893,
"mean_token_accuracy": 0.14726104438304902,
"num_tokens": 12428699.0,
"step": 5650
},
{
"entropy": 5.689584064483642,
"epoch": 0.5612346169114728,
"grad_norm": 0.8671875,
"learning_rate": 0.0004975873099943302,
"loss": 5.57,
"mean_token_accuracy": 0.14373983442783356,
"num_tokens": 12439025.0,
"step": 5655
},
{
"entropy": 5.737751007080078,
"epoch": 0.5617308455736403,
"grad_norm": 0.8046875,
"learning_rate": 0.0004975821323981875,
"loss": 5.5221,
"mean_token_accuracy": 0.14871042519807814,
"num_tokens": 12449439.0,
"step": 5660
},
{
"entropy": 5.723925542831421,
"epoch": 0.5622270742358079,
"grad_norm": 0.83984375,
"learning_rate": 0.0004975769492824877,
"loss": 5.4628,
"mean_token_accuracy": 0.14492267221212388,
"num_tokens": 12460066.0,
"step": 5665
},
{
"entropy": 5.664363098144531,
"epoch": 0.5627233028979753,
"grad_norm": 0.90234375,
"learning_rate": 0.0004975717606473589,
"loss": 5.4883,
"mean_token_accuracy": 0.15085864216089248,
"num_tokens": 12471093.0,
"step": 5670
},
{
"entropy": 5.739382266998291,
"epoch": 0.5632195315601429,
"grad_norm": 0.87890625,
"learning_rate": 0.0004975665664929302,
"loss": 5.513,
"mean_token_accuracy": 0.14348138272762298,
"num_tokens": 12481181.0,
"step": 5675
},
{
"entropy": 5.70921950340271,
"epoch": 0.5637157602223104,
"grad_norm": 0.9296875,
"learning_rate": 0.0004975613668193303,
"loss": 5.528,
"mean_token_accuracy": 0.14321721345186234,
"num_tokens": 12493068.0,
"step": 5680
},
{
"entropy": 5.631594800949097,
"epoch": 0.564211988884478,
"grad_norm": 0.921875,
"learning_rate": 0.000497556161626688,
"loss": 5.4839,
"mean_token_accuracy": 0.1489529438316822,
"num_tokens": 12504025.0,
"step": 5685
},
{
"entropy": 5.679368209838867,
"epoch": 0.5647082175466455,
"grad_norm": 0.8203125,
"learning_rate": 0.0004975509509151325,
"loss": 5.4858,
"mean_token_accuracy": 0.1454532712697983,
"num_tokens": 12514567.0,
"step": 5690
},
{
"entropy": 5.785870027542114,
"epoch": 0.565204446208813,
"grad_norm": 0.91015625,
"learning_rate": 0.0004975457346847929,
"loss": 5.5249,
"mean_token_accuracy": 0.1441642612218857,
"num_tokens": 12525088.0,
"step": 5695
},
{
"entropy": 5.677102327346802,
"epoch": 0.5657006748709805,
"grad_norm": 0.79296875,
"learning_rate": 0.0004975405129357986,
"loss": 5.4866,
"mean_token_accuracy": 0.14762923568487168,
"num_tokens": 12536002.0,
"step": 5700
},
{
"entropy": 5.676094245910645,
"epoch": 0.5661969035331481,
"grad_norm": 0.8671875,
"learning_rate": 0.0004975352856682792,
"loss": 5.4974,
"mean_token_accuracy": 0.15148521959781647,
"num_tokens": 12547468.0,
"step": 5705
},
{
"entropy": 5.715325212478637,
"epoch": 0.5666931321953156,
"grad_norm": 0.91015625,
"learning_rate": 0.0004975300528823642,
"loss": 5.4656,
"mean_token_accuracy": 0.1467609941959381,
"num_tokens": 12557670.0,
"step": 5710
},
{
"entropy": 5.710341072082519,
"epoch": 0.5671893608574832,
"grad_norm": 0.89453125,
"learning_rate": 0.0004975248145781834,
"loss": 5.5278,
"mean_token_accuracy": 0.1446359783411026,
"num_tokens": 12568425.0,
"step": 5715
},
{
"entropy": 5.608817863464355,
"epoch": 0.5676855895196506,
"grad_norm": 0.8359375,
"learning_rate": 0.0004975195707558667,
"loss": 5.4998,
"mean_token_accuracy": 0.14632719457149507,
"num_tokens": 12579138.0,
"step": 5720
},
{
"entropy": 5.64209623336792,
"epoch": 0.5681818181818182,
"grad_norm": 0.8359375,
"learning_rate": 0.0004975143214155443,
"loss": 5.4958,
"mean_token_accuracy": 0.14967800229787825,
"num_tokens": 12589743.0,
"step": 5725
},
{
"entropy": 5.688083267211914,
"epoch": 0.5686780468439857,
"grad_norm": 0.90625,
"learning_rate": 0.0004975090665573459,
"loss": 5.4851,
"mean_token_accuracy": 0.152118718624115,
"num_tokens": 12600648.0,
"step": 5730
},
{
"entropy": 5.645006799697876,
"epoch": 0.5691742755061532,
"grad_norm": 0.8203125,
"learning_rate": 0.0004975038061814023,
"loss": 5.5031,
"mean_token_accuracy": 0.14906034767627716,
"num_tokens": 12611369.0,
"step": 5735
},
{
"entropy": 5.7086090564727785,
"epoch": 0.5696705041683208,
"grad_norm": 0.78515625,
"learning_rate": 0.0004974985402878437,
"loss": 5.4672,
"mean_token_accuracy": 0.1463977113366127,
"num_tokens": 12621992.0,
"step": 5740
},
{
"entropy": 5.697903108596802,
"epoch": 0.5701667328304882,
"grad_norm": 0.94921875,
"learning_rate": 0.0004974932688768007,
"loss": 5.5454,
"mean_token_accuracy": 0.14656417667865754,
"num_tokens": 12632463.0,
"step": 5745
},
{
"entropy": 5.6192210674285885,
"epoch": 0.5706629614926558,
"grad_norm": 0.91796875,
"learning_rate": 0.0004974879919484042,
"loss": 5.4058,
"mean_token_accuracy": 0.15206122696399688,
"num_tokens": 12642282.0,
"step": 5750
},
{
"entropy": 5.636960077285766,
"epoch": 0.5711591901548233,
"grad_norm": 0.921875,
"learning_rate": 0.0004974827095027848,
"loss": 5.4346,
"mean_token_accuracy": 0.15423172563314438,
"num_tokens": 12653495.0,
"step": 5755
},
{
"entropy": 5.664530849456787,
"epoch": 0.5716554188169909,
"grad_norm": 0.81640625,
"learning_rate": 0.0004974774215400736,
"loss": 5.3785,
"mean_token_accuracy": 0.15366745442152024,
"num_tokens": 12664322.0,
"step": 5760
},
{
"entropy": 5.542517852783203,
"epoch": 0.5721516474791584,
"grad_norm": 0.91796875,
"learning_rate": 0.0004974721280604019,
"loss": 5.4768,
"mean_token_accuracy": 0.14642566740512847,
"num_tokens": 12673981.0,
"step": 5765
},
{
"entropy": 5.599040508270264,
"epoch": 0.572647876141326,
"grad_norm": 0.95703125,
"learning_rate": 0.0004974668290639007,
"loss": 5.4781,
"mean_token_accuracy": 0.14832795709371566,
"num_tokens": 12684778.0,
"step": 5770
},
{
"entropy": 5.705513000488281,
"epoch": 0.5731441048034934,
"grad_norm": 0.8828125,
"learning_rate": 0.0004974615245507014,
"loss": 5.4421,
"mean_token_accuracy": 0.14995291382074355,
"num_tokens": 12695287.0,
"step": 5775
},
{
"entropy": 5.598648929595948,
"epoch": 0.573640333465661,
"grad_norm": 0.875,
"learning_rate": 0.0004974562145209359,
"loss": 5.5041,
"mean_token_accuracy": 0.14448004961013794,
"num_tokens": 12705773.0,
"step": 5780
},
{
"entropy": 5.6014838218688965,
"epoch": 0.5741365621278285,
"grad_norm": 0.86328125,
"learning_rate": 0.0004974508989747354,
"loss": 5.4276,
"mean_token_accuracy": 0.15292005538940429,
"num_tokens": 12716783.0,
"step": 5785
},
{
"entropy": 5.6707727909088135,
"epoch": 0.5746327907899961,
"grad_norm": 0.87109375,
"learning_rate": 0.0004974455779122321,
"loss": 5.4976,
"mean_token_accuracy": 0.14310954585671426,
"num_tokens": 12728146.0,
"step": 5790
},
{
"entropy": 5.705024862289429,
"epoch": 0.5751290194521635,
"grad_norm": 0.88671875,
"learning_rate": 0.0004974402513335577,
"loss": 5.5146,
"mean_token_accuracy": 0.14299591705203057,
"num_tokens": 12739683.0,
"step": 5795
},
{
"entropy": 5.713213539123535,
"epoch": 0.5756252481143311,
"grad_norm": 1.1328125,
"learning_rate": 0.0004974349192388444,
"loss": 5.4891,
"mean_token_accuracy": 0.14356330558657646,
"num_tokens": 12750257.0,
"step": 5800
},
{
"entropy": 5.72646746635437,
"epoch": 0.5761214767764986,
"grad_norm": 0.81640625,
"learning_rate": 0.0004974295816282245,
"loss": 5.5675,
"mean_token_accuracy": 0.14630476087331773,
"num_tokens": 12761055.0,
"step": 5805
},
{
"entropy": 5.67992148399353,
"epoch": 0.5766177054386661,
"grad_norm": 0.94921875,
"learning_rate": 0.0004974242385018303,
"loss": 5.5052,
"mean_token_accuracy": 0.14898856729269028,
"num_tokens": 12771014.0,
"step": 5810
},
{
"entropy": 5.635032606124878,
"epoch": 0.5771139341008337,
"grad_norm": 0.87890625,
"learning_rate": 0.000497418889859794,
"loss": 5.424,
"mean_token_accuracy": 0.15025612711906433,
"num_tokens": 12782623.0,
"step": 5815
},
{
"entropy": 5.713941335678101,
"epoch": 0.5776101627630011,
"grad_norm": 0.92578125,
"learning_rate": 0.0004974135357022488,
"loss": 5.543,
"mean_token_accuracy": 0.14384849816560746,
"num_tokens": 12793533.0,
"step": 5820
},
{
"entropy": 5.593503761291504,
"epoch": 0.5781063914251687,
"grad_norm": 0.87890625,
"learning_rate": 0.0004974081760293269,
"loss": 5.4105,
"mean_token_accuracy": 0.14829638749361038,
"num_tokens": 12804389.0,
"step": 5825
},
{
"entropy": 5.666346645355224,
"epoch": 0.5786026200873362,
"grad_norm": 0.8125,
"learning_rate": 0.0004974028108411617,
"loss": 5.512,
"mean_token_accuracy": 0.15230901390314103,
"num_tokens": 12815672.0,
"step": 5830
},
{
"entropy": 5.732553863525391,
"epoch": 0.5790988487495038,
"grad_norm": 0.875,
"learning_rate": 0.0004973974401378859,
"loss": 5.5116,
"mean_token_accuracy": 0.15112632364034653,
"num_tokens": 12826113.0,
"step": 5835
},
{
"entropy": 5.612425994873047,
"epoch": 0.5795950774116713,
"grad_norm": 0.8984375,
"learning_rate": 0.0004973920639196329,
"loss": 5.4408,
"mean_token_accuracy": 0.1494893580675125,
"num_tokens": 12835941.0,
"step": 5840
},
{
"entropy": 5.639367532730103,
"epoch": 0.5800913060738389,
"grad_norm": 0.8671875,
"learning_rate": 0.0004973866821865358,
"loss": 5.5282,
"mean_token_accuracy": 0.14228470176458358,
"num_tokens": 12846186.0,
"step": 5845
},
{
"entropy": 5.718220901489258,
"epoch": 0.5805875347360063,
"grad_norm": 0.84375,
"learning_rate": 0.0004973812949387283,
"loss": 5.4929,
"mean_token_accuracy": 0.14542728066444396,
"num_tokens": 12856380.0,
"step": 5850
},
{
"entropy": 5.649071073532104,
"epoch": 0.5810837633981739,
"grad_norm": 0.875,
"learning_rate": 0.0004973759021763439,
"loss": 5.5286,
"mean_token_accuracy": 0.14764680862426757,
"num_tokens": 12865999.0,
"step": 5855
},
{
"entropy": 5.68231954574585,
"epoch": 0.5815799920603414,
"grad_norm": 0.890625,
"learning_rate": 0.0004973705038995162,
"loss": 5.4126,
"mean_token_accuracy": 0.14954375773668288,
"num_tokens": 12876388.0,
"step": 5860
},
{
"entropy": 5.625105524063111,
"epoch": 0.582076220722509,
"grad_norm": 0.9140625,
"learning_rate": 0.0004973651001083792,
"loss": 5.4145,
"mean_token_accuracy": 0.14912877306342126,
"num_tokens": 12885862.0,
"step": 5865
},
{
"entropy": 5.605915117263794,
"epoch": 0.5825724493846764,
"grad_norm": 0.88671875,
"learning_rate": 0.0004973596908030669,
"loss": 5.5022,
"mean_token_accuracy": 0.1445961207151413,
"num_tokens": 12897489.0,
"step": 5870
},
{
"entropy": 5.703808736801148,
"epoch": 0.583068678046844,
"grad_norm": 0.90625,
"learning_rate": 0.0004973542759837135,
"loss": 5.5464,
"mean_token_accuracy": 0.14082904383540154,
"num_tokens": 12907395.0,
"step": 5875
},
{
"entropy": 5.62304458618164,
"epoch": 0.5835649067090115,
"grad_norm": 0.87890625,
"learning_rate": 0.000497348855650453,
"loss": 5.4717,
"mean_token_accuracy": 0.1517767146229744,
"num_tokens": 12919439.0,
"step": 5880
},
{
"entropy": 5.69926609992981,
"epoch": 0.584061135371179,
"grad_norm": 0.84375,
"learning_rate": 0.0004973434298034201,
"loss": 5.5616,
"mean_token_accuracy": 0.14480644688010216,
"num_tokens": 12931892.0,
"step": 5885
},
{
"entropy": 5.770834016799927,
"epoch": 0.5845573640333466,
"grad_norm": 0.84375,
"learning_rate": 0.0004973379984427492,
"loss": 5.603,
"mean_token_accuracy": 0.14584836214780808,
"num_tokens": 12943189.0,
"step": 5890
},
{
"entropy": 5.670014476776123,
"epoch": 0.585053592695514,
"grad_norm": 0.875,
"learning_rate": 0.000497332561568575,
"loss": 5.5406,
"mean_token_accuracy": 0.14588731974363328,
"num_tokens": 12954662.0,
"step": 5895
},
{
"entropy": 5.63857250213623,
"epoch": 0.5855498213576816,
"grad_norm": 0.87890625,
"learning_rate": 0.0004973271191810325,
"loss": 5.4833,
"mean_token_accuracy": 0.14281948506832123,
"num_tokens": 12964924.0,
"step": 5900
},
{
"entropy": 5.689583873748779,
"epoch": 0.5860460500198491,
"grad_norm": 0.82421875,
"learning_rate": 0.0004973216712802564,
"loss": 5.5549,
"mean_token_accuracy": 0.14132398664951323,
"num_tokens": 12976483.0,
"step": 5905
},
{
"entropy": 5.654120254516601,
"epoch": 0.5865422786820167,
"grad_norm": 0.83203125,
"learning_rate": 0.000497316217866382,
"loss": 5.4601,
"mean_token_accuracy": 0.15196489095687865,
"num_tokens": 12988054.0,
"step": 5910
},
{
"entropy": 5.69523344039917,
"epoch": 0.5870385073441842,
"grad_norm": 0.87109375,
"learning_rate": 0.0004973107589395444,
"loss": 5.5062,
"mean_token_accuracy": 0.1495279923081398,
"num_tokens": 13000596.0,
"step": 5915
},
{
"entropy": 5.558714103698731,
"epoch": 0.5875347360063518,
"grad_norm": 0.80078125,
"learning_rate": 0.0004973052944998791,
"loss": 5.3601,
"mean_token_accuracy": 0.15159481316804885,
"num_tokens": 13012360.0,
"step": 5920
},
{
"entropy": 5.637037706375122,
"epoch": 0.5880309646685192,
"grad_norm": 0.828125,
"learning_rate": 0.0004972998245475214,
"loss": 5.5061,
"mean_token_accuracy": 0.14521099850535393,
"num_tokens": 13023806.0,
"step": 5925
},
{
"entropy": 5.674616146087646,
"epoch": 0.5885271933306868,
"grad_norm": 0.87890625,
"learning_rate": 0.000497294349082607,
"loss": 5.4444,
"mean_token_accuracy": 0.14743458330631257,
"num_tokens": 13034049.0,
"step": 5930
},
{
"entropy": 5.698005294799804,
"epoch": 0.5890234219928543,
"grad_norm": 0.83984375,
"learning_rate": 0.0004972888681052719,
"loss": 5.5418,
"mean_token_accuracy": 0.14588503241539003,
"num_tokens": 13045006.0,
"step": 5935
},
{
"entropy": 5.713470411300659,
"epoch": 0.5895196506550219,
"grad_norm": 0.8828125,
"learning_rate": 0.0004972833816156518,
"loss": 5.454,
"mean_token_accuracy": 0.151131209731102,
"num_tokens": 13055337.0,
"step": 5940
},
{
"entropy": 5.613440036773682,
"epoch": 0.5900158793171894,
"grad_norm": 0.890625,
"learning_rate": 0.0004972778896138828,
"loss": 5.4414,
"mean_token_accuracy": 0.151805180311203,
"num_tokens": 13065267.0,
"step": 5945
},
{
"entropy": 5.648981380462646,
"epoch": 0.5905121079793569,
"grad_norm": 1.0234375,
"learning_rate": 0.0004972723921001011,
"loss": 5.5115,
"mean_token_accuracy": 0.14798768609762192,
"num_tokens": 13075173.0,
"step": 5950
},
{
"entropy": 5.707030391693115,
"epoch": 0.5910083366415244,
"grad_norm": 0.86328125,
"learning_rate": 0.0004972668890744431,
"loss": 5.4911,
"mean_token_accuracy": 0.14562270641326905,
"num_tokens": 13086848.0,
"step": 5955
},
{
"entropy": 5.640896415710449,
"epoch": 0.5915045653036919,
"grad_norm": 0.83984375,
"learning_rate": 0.0004972613805370452,
"loss": 5.4768,
"mean_token_accuracy": 0.1498607374727726,
"num_tokens": 13096865.0,
"step": 5960
},
{
"entropy": 5.612109804153443,
"epoch": 0.5920007939658595,
"grad_norm": 0.8515625,
"learning_rate": 0.0004972558664880439,
"loss": 5.4543,
"mean_token_accuracy": 0.14806280583143233,
"num_tokens": 13107907.0,
"step": 5965
},
{
"entropy": 5.591845846176147,
"epoch": 0.592497022628027,
"grad_norm": 1.015625,
"learning_rate": 0.000497250346927576,
"loss": 5.373,
"mean_token_accuracy": 0.15522033870220184,
"num_tokens": 13118171.0,
"step": 5970
},
{
"entropy": 5.639474821090698,
"epoch": 0.5929932512901945,
"grad_norm": 1.015625,
"learning_rate": 0.0004972448218557785,
"loss": 5.3608,
"mean_token_accuracy": 0.15381580740213394,
"num_tokens": 13127042.0,
"step": 5975
},
{
"entropy": 5.682769632339477,
"epoch": 0.593489479952362,
"grad_norm": 0.84765625,
"learning_rate": 0.0004972392912727882,
"loss": 5.6267,
"mean_token_accuracy": 0.1483294628560543,
"num_tokens": 13137026.0,
"step": 5980
},
{
"entropy": 5.6737512111663815,
"epoch": 0.5939857086145296,
"grad_norm": 0.83984375,
"learning_rate": 0.0004972337551787425,
"loss": 5.4745,
"mean_token_accuracy": 0.15264537185430527,
"num_tokens": 13146947.0,
"step": 5985
},
{
"entropy": 5.762002801895141,
"epoch": 0.5944819372766971,
"grad_norm": 0.83984375,
"learning_rate": 0.0004972282135737785,
"loss": 5.528,
"mean_token_accuracy": 0.14844981729984283,
"num_tokens": 13158714.0,
"step": 5990
},
{
"entropy": 5.709932327270508,
"epoch": 0.5949781659388647,
"grad_norm": 0.875,
"learning_rate": 0.0004972226664580337,
"loss": 5.5318,
"mean_token_accuracy": 0.14496293887495995,
"num_tokens": 13170156.0,
"step": 5995
},
{
"entropy": 5.593976449966431,
"epoch": 0.5954743946010321,
"grad_norm": 0.90234375,
"learning_rate": 0.0004972171138316456,
"loss": 5.4315,
"mean_token_accuracy": 0.14985745549201965,
"num_tokens": 13179919.0,
"step": 6000
},
{
"epoch": 0.5954743946010321,
"eval_entropy": 5.474441283727434,
"eval_loss": 5.552264213562012,
"eval_mean_token_accuracy": 0.14955780961736417,
"eval_num_tokens": 13179919.0,
"eval_runtime": 20.3484,
"eval_samples_per_second": 1654.624,
"eval_steps_per_second": 206.846,
"step": 6000
}
],
"logging_steps": 5,
"max_steps": 100750,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 3000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3273706621992960.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink