fpadovani/jpn-jpan-100mb-ppt-Dp-100mb_seed455
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
cdf488cb36a1e1f72eaddcd332614217d5fdbb5d
jpn-jpan-100mb-ppt-Dp-100mb…/checkpoint-1500/trainer_state.json

3035 lines
81 KiB
JSON
Raw Normal View History

初始化项目，由ModelHub XC社区提供模型 Model: fpadovani/jpn-jpan-100mb-ppt-Dp-100mb_seed455 Source: Original Platform
2026-11-03 16:30:21 +08:00
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.1145431636821809,
"eval_steps": 500,
"global_step": 1500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.69164161682129,
"epoch": 0.00038181054560726966,
"grad_norm": 12.625,
"learning_rate": 2e-06,
"loss": 10.8218,
"mean_token_accuracy": 8.658008882775903e-05,
"num_tokens": 12361.0,
"step": 5
},
{
"entropy": 10.691694355010986,
"epoch": 0.0007636210912145393,
"grad_norm": 12.125,
"learning_rate": 4.5e-06,
"loss": 10.7677,
"mean_token_accuracy": 7.163324044086039e-05,
"num_tokens": 24868.0,
"step": 10
},
{
"entropy": 10.691064453125,
"epoch": 0.001145431636821809,
"grad_norm": 8.625,
"learning_rate": 7e-06,
"loss": 10.5505,
"mean_token_accuracy": 0.025445129600120707,
"num_tokens": 37349.0,
"step": 15
},
{
"entropy": 10.675649642944336,
"epoch": 0.0015272421824290786,
"grad_norm": 5.84375,
"learning_rate": 9.5e-06,
"loss": 10.2462,
"mean_token_accuracy": 0.04803365767002106,
"num_tokens": 50032.0,
"step": 20
},
{
"entropy": 10.536279582977295,
"epoch": 0.0019090527280363483,
"grad_norm": 3.9375,
"learning_rate": 1.2e-05,
"loss": 9.9405,
"mean_token_accuracy": 0.05315057449042797,
"num_tokens": 61773.0,
"step": 25
},
{
"entropy": 10.429403972625732,
"epoch": 0.002290863273643618,
"grad_norm": 2.96875,
"learning_rate": 1.4500000000000002e-05,
"loss": 9.7807,
"mean_token_accuracy": 0.06437831856310368,
"num_tokens": 74017.0,
"step": 30
},
{
"entropy": 10.479450607299805,
"epoch": 0.0026726738192508876,
"grad_norm": 2.859375,
"learning_rate": 1.7000000000000003e-05,
"loss": 9.7981,
"mean_token_accuracy": 0.05096841156482697,
"num_tokens": 87243.0,
"step": 35
},
{
"entropy": 10.484258651733398,
"epoch": 0.0030544843648581572,
"grad_norm": 2.5,
"learning_rate": 1.95e-05,
"loss": 9.7624,
"mean_token_accuracy": 0.05546987727284432,
"num_tokens": 99487.0,
"step": 40
},
{
"entropy": 10.404310131072998,
"epoch": 0.003436294910465427,
"grad_norm": 2.171875,
"learning_rate": 2.2e-05,
"loss": 9.6868,
"mean_token_accuracy": 0.05369072034955025,
"num_tokens": 112020.0,
"step": 45
},
{
"entropy": 10.335991859436035,
"epoch": 0.0038181054560726966,
"grad_norm": 2.5625,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.6778,
"mean_token_accuracy": 0.06317304000258446,
"num_tokens": 124703.0,
"step": 50
},
{
"entropy": 10.376905536651611,
"epoch": 0.004199916001679967,
"grad_norm": 2.421875,
"learning_rate": 2.7e-05,
"loss": 9.6359,
"mean_token_accuracy": 0.06049793548882008,
"num_tokens": 137288.0,
"step": 55
},
{
"entropy": 10.288232707977295,
"epoch": 0.004581726547287236,
"grad_norm": 2.09375,
"learning_rate": 2.95e-05,
"loss": 9.5869,
"mean_token_accuracy": 0.06641441285610199,
"num_tokens": 150954.0,
"step": 60
},
{
"entropy": 10.287444877624512,
"epoch": 0.004963537092894506,
"grad_norm": 2.171875,
"learning_rate": 3.2e-05,
"loss": 9.5108,
"mean_token_accuracy": 0.0745735965669155,
"num_tokens": 163545.0,
"step": 65
},
{
"entropy": 10.178703498840331,
"epoch": 0.005345347638501775,
"grad_norm": 2.703125,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.3584,
"mean_token_accuracy": 0.08311713635921478,
"num_tokens": 175089.0,
"step": 70
},
{
"entropy": 10.129299736022949,
"epoch": 0.005727158184109045,
"grad_norm": 1.8515625,
"learning_rate": 3.7e-05,
"loss": 9.3743,
"mean_token_accuracy": 0.08062887638807296,
"num_tokens": 187355.0,
"step": 75
},
{
"entropy": 10.13417091369629,
"epoch": 0.0061089687297163145,
"grad_norm": 1.9140625,
"learning_rate": 3.95e-05,
"loss": 9.3354,
"mean_token_accuracy": 0.08319172337651252,
"num_tokens": 200925.0,
"step": 80
},
{
"entropy": 10.144186305999757,
"epoch": 0.006490779275323585,
"grad_norm": 1.875,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.2585,
"mean_token_accuracy": 0.08622478619217873,
"num_tokens": 213393.0,
"step": 85
},
{
"entropy": 10.041219711303711,
"epoch": 0.006872589820930854,
"grad_norm": 2.203125,
"learning_rate": 4.45e-05,
"loss": 9.2302,
"mean_token_accuracy": 0.08239276111125945,
"num_tokens": 226261.0,
"step": 90
},
{
"entropy": 10.069394493103028,
"epoch": 0.007254400366538124,
"grad_norm": 1.9375,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.1435,
"mean_token_accuracy": 0.0902495726943016,
"num_tokens": 238277.0,
"step": 95
},
{
"entropy": 9.870719623565673,
"epoch": 0.007636210912145393,
"grad_norm": 2.109375,
"learning_rate": 4.9500000000000004e-05,
"loss": 8.9964,
"mean_token_accuracy": 0.10341855883598328,
"num_tokens": 250664.0,
"step": 100
},
{
"entropy": 10.039793014526367,
"epoch": 0.008018021457752662,
"grad_norm": 1.78125,
"learning_rate": 5.2e-05,
"loss": 9.0284,
"mean_token_accuracy": 0.086080052703619,
"num_tokens": 262403.0,
"step": 105
},
{
"entropy": 9.880093574523926,
"epoch": 0.008399832003359933,
"grad_norm": 1.671875,
"learning_rate": 5.45e-05,
"loss": 8.9692,
"mean_token_accuracy": 0.09072378277778625,
"num_tokens": 276036.0,
"step": 110
},
{
"entropy": 9.872721576690674,
"epoch": 0.008781642548967203,
"grad_norm": 1.5703125,
"learning_rate": 5.7e-05,
"loss": 8.8476,
"mean_token_accuracy": 0.0959131620824337,
"num_tokens": 287556.0,
"step": 115
},
{
"entropy": 9.759550285339355,
"epoch": 0.009163453094574472,
"grad_norm": 2.03125,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.8335,
"mean_token_accuracy": 0.09087418392300606,
"num_tokens": 299285.0,
"step": 120
},
{
"entropy": 9.744350814819336,
"epoch": 0.009545263640181741,
"grad_norm": 1.609375,
"learning_rate": 6.2e-05,
"loss": 8.7962,
"mean_token_accuracy": 0.08946920111775399,
"num_tokens": 312172.0,
"step": 125
},
{
"entropy": 9.738001823425293,
"epoch": 0.009927074185789012,
"grad_norm": 1.5390625,
"learning_rate": 6.450000000000001e-05,
"loss": 8.707,
"mean_token_accuracy": 0.09405562952160836,
"num_tokens": 325689.0,
"step": 130
},
{
"entropy": 9.49710578918457,
"epoch": 0.010308884731396281,
"grad_norm": 1.34375,
"learning_rate": 6.7e-05,
"loss": 8.645,
"mean_token_accuracy": 0.08777745962142944,
"num_tokens": 337893.0,
"step": 135
},
{
"entropy": 9.492643737792969,
"epoch": 0.01069069527700355,
"grad_norm": 1.6015625,
"learning_rate": 6.950000000000001e-05,
"loss": 8.576,
"mean_token_accuracy": 0.09355450496077537,
"num_tokens": 349733.0,
"step": 140
},
{
"entropy": 9.370576190948487,
"epoch": 0.011072505822610821,
"grad_norm": 1.4375,
"learning_rate": 7.2e-05,
"loss": 8.5334,
"mean_token_accuracy": 0.0905280888080597,
"num_tokens": 362624.0,
"step": 145
},
{
"entropy": 9.226265716552735,
"epoch": 0.01145431636821809,
"grad_norm": 1.234375,
"learning_rate": 7.45e-05,
"loss": 8.4527,
"mean_token_accuracy": 0.0901684120297432,
"num_tokens": 374564.0,
"step": 150
},
{
"entropy": 9.136695861816406,
"epoch": 0.01183612691382536,
"grad_norm": 1.9296875,
"learning_rate": 7.7e-05,
"loss": 8.4298,
"mean_token_accuracy": 0.09475162327289581,
"num_tokens": 387813.0,
"step": 155
},
{
"entropy": 9.067316818237305,
"epoch": 0.012217937459432629,
"grad_norm": 1.421875,
"learning_rate": 7.950000000000001e-05,
"loss": 8.4477,
"mean_token_accuracy": 0.09096779003739357,
"num_tokens": 400037.0,
"step": 160
},
{
"entropy": 8.901337909698487,
"epoch": 0.0125997480050399,
"grad_norm": 1.3515625,
"learning_rate": 8.2e-05,
"loss": 8.3423,
"mean_token_accuracy": 0.09573070406913757,
"num_tokens": 412753.0,
"step": 165
},
{
"entropy": 8.815200519561767,
"epoch": 0.01298155855064717,
"grad_norm": 1.1640625,
"learning_rate": 8.450000000000001e-05,
"loss": 8.2726,
"mean_token_accuracy": 0.09817055985331535,
"num_tokens": 425067.0,
"step": 170
},
{
"entropy": 8.797827339172363,
"epoch": 0.013363369096254438,
"grad_norm": 1.2734375,
"learning_rate": 8.7e-05,
"loss": 8.2692,
"mean_token_accuracy": 0.09679182544350624,
"num_tokens": 437866.0,
"step": 175
},
{
"entropy": 8.725690650939942,
"epoch": 0.013745179641861708,
"grad_norm": 1.3125,
"learning_rate": 8.95e-05,
"loss": 8.2826,
"mean_token_accuracy": 0.09829293191432953,
"num_tokens": 449242.0,
"step": 180
},
{
"entropy": 8.697874736785888,
"epoch": 0.014126990187468979,
"grad_norm": 1.140625,
"learning_rate": 9.2e-05,
"loss": 8.3005,
"mean_token_accuracy": 0.09152775704860687,
"num_tokens": 461696.0,
"step": 185
},
{
"entropy": 8.68546266555786,
"epoch": 0.014508800733076248,
"grad_norm": 1.171875,
"learning_rate": 9.45e-05,
"loss": 8.3056,
"mean_token_accuracy": 0.09098555892705917,
"num_tokens": 473875.0,
"step": 190
},
{
"entropy": 8.710109996795655,
"epoch": 0.014890611278683517,
"grad_norm": 1.125,
"learning_rate": 9.7e-05,
"loss": 8.2518,
"mean_token_accuracy": 0.09681897088885308,
"num_tokens": 486124.0,
"step": 195
},
{
"entropy": 8.575210094451904,
"epoch": 0.015272421824290786,
"grad_norm": 1.2109375,
"learning_rate": 9.95e-05,
"loss": 8.1633,
"mean_token_accuracy": 0.09988361150026322,
"num_tokens": 498782.0,
"step": 200
},
{
"entropy": 8.50928020477295,
"epoch": 0.015654232369898057,
"grad_norm": 1.0625,
"learning_rate": 0.000102,
"loss": 8.2107,
"mean_token_accuracy": 0.09708656743168831,
"num_tokens": 511164.0,
"step": 205
},
{
"entropy": 8.605385684967041,
"epoch": 0.016036042915505325,
"grad_norm": 1.0703125,
"learning_rate": 0.00010449999999999999,
"loss": 8.1851,
"mean_token_accuracy": 0.09802236407995224,
"num_tokens": 522323.0,
"step": 210
},
{
"entropy": 8.443045234680175,
"epoch": 0.016417853461112596,
"grad_norm": 1.2265625,
"learning_rate": 0.000107,
"loss": 8.1946,
"mean_token_accuracy": 0.09736634492874145,
"num_tokens": 535007.0,
"step": 215
},
{
"entropy": 8.477715015411377,
"epoch": 0.016799664006719867,
"grad_norm": 1.078125,
"learning_rate": 0.0001095,
"loss": 8.1695,
"mean_token_accuracy": 0.09888660684227943,
"num_tokens": 547715.0,
"step": 220
},
{
"entropy": 8.503206348419189,
"epoch": 0.017181474552327134,
"grad_norm": 1.1484375,
"learning_rate": 0.000112,
"loss": 8.1855,
"mean_token_accuracy": 0.09887165278196335,
"num_tokens": 559301.0,
"step": 225
},
{
"entropy": 8.340103912353516,
"epoch": 0.017563285097934405,
"grad_norm": 1.140625,
"learning_rate": 0.0001145,
"loss": 8.0996,
"mean_token_accuracy": 0.10817644372582436,
"num_tokens": 570741.0,
"step": 230
},
{
"entropy": 8.480874156951904,
"epoch": 0.017945095643541676,
"grad_norm": 1.3203125,
"learning_rate": 0.00011700000000000001,
"loss": 8.1212,
"mean_token_accuracy": 0.10553076937794685,
"num_tokens": 582263.0,
"step": 235
},
{
"entropy": 8.344198608398438,
"epoch": 0.018326906189148943,
"grad_norm": 1.0859375,
"learning_rate": 0.00011949999999999999,
"loss": 8.106,
"mean_token_accuracy": 0.10434014797210693,
"num_tokens": 593081.0,
"step": 240
},
{
"entropy": 8.449238491058349,
"epoch": 0.018708716734756214,
"grad_norm": 1.21875,
"learning_rate": 0.000122,
"loss": 8.2145,
"mean_token_accuracy": 0.09799340069293976,
"num_tokens": 605725.0,
"step": 245
},
{
"entropy": 8.343080520629883,
"epoch": 0.019090527280363482,
"grad_norm": 1.3125,
"learning_rate": 0.0001245,
"loss": 8.1114,
"mean_token_accuracy": 0.10767603665590286,
"num_tokens": 617740.0,
"step": 250
},
{
"entropy": 8.363426685333252,
"epoch": 0.019472337825970753,
"grad_norm": 1.1875,
"learning_rate": 0.000127,
"loss": 8.0848,
"mean_token_accuracy": 0.10610114932060241,
"num_tokens": 630186.0,
"step": 255
},
{
"entropy": 8.39581470489502,
"epoch": 0.019854148371578024,
"grad_norm": 1.0234375,
"learning_rate": 0.0001295,
"loss": 8.0586,
"mean_token_accuracy": 0.10069277137517929,
"num_tokens": 642304.0,
"step": 260
},
{
"entropy": 8.270940399169922,
"epoch": 0.02023595891718529,
"grad_norm": 1.03125,
"learning_rate": 0.000132,
"loss": 8.1002,
"mean_token_accuracy": 0.10599680617451668,
"num_tokens": 655029.0,
"step": 265
},
{
"entropy": 8.393358612060547,
"epoch": 0.020617769462792562,
"grad_norm": 1.453125,
"learning_rate": 0.00013450000000000002,
"loss": 8.1602,
"mean_token_accuracy": 0.10279513671994209,
"num_tokens": 667193.0,
"step": 270
},
{
"entropy": 8.328470611572266,
"epoch": 0.020999580008399833,
"grad_norm": 1.125,
"learning_rate": 0.00013700000000000002,
"loss": 8.0464,
"mean_token_accuracy": 0.11250988617539406,
"num_tokens": 681382.0,
"step": 275
},
{
"entropy": 8.321224308013916,
"epoch": 0.0213813905540071,
"grad_norm": 1.0390625,
"learning_rate": 0.0001395,
"loss": 8.155,
"mean_token_accuracy": 0.1049920804798603,
"num_tokens": 694280.0,
"step": 280
},
{
"entropy": 8.34145975112915,
"epoch": 0.02176320109961437,
"grad_norm": 1.28125,
"learning_rate": 0.00014199999999999998,
"loss": 7.9959,
"mean_token_accuracy": 0.11552856862545013,
"num_tokens": 707682.0,
"step": 285
},
{
"entropy": 8.31129264831543,
"epoch": 0.022145011645221643,
"grad_norm": 1.2734375,
"learning_rate": 0.0001445,
"loss": 8.1054,
"mean_token_accuracy": 0.10616647526621818,
"num_tokens": 720237.0,
"step": 290
},
{
"entropy": 8.282443714141845,
"epoch": 0.02252682219082891,
"grad_norm": 1.234375,
"learning_rate": 0.000147,
"loss": 8.0555,
"mean_token_accuracy": 0.11427121162414551,
"num_tokens": 732816.0,
"step": 295
},
{
"entropy": 8.297718620300293,
"epoch": 0.02290863273643618,
"grad_norm": 1.453125,
"learning_rate": 0.0001495,
"loss": 8.0507,
"mean_token_accuracy": 0.10974556654691696,
"num_tokens": 744107.0,
"step": 300
},
{
"entropy": 8.396364116668702,
"epoch": 0.02329044328204345,
"grad_norm": 1.1640625,
"learning_rate": 0.000152,
"loss": 8.0398,
"mean_token_accuracy": 0.10700830891728401,
"num_tokens": 757498.0,
"step": 305
},
{
"entropy": 8.064451313018798,
"epoch": 0.02367225382765072,
"grad_norm": 2.09375,
"learning_rate": 0.00015450000000000001,
"loss": 7.9454,
"mean_token_accuracy": 0.12401409074664116,
"num_tokens": 769429.0,
"step": 310
},
{
"entropy": 8.283315277099609,
"epoch": 0.02405406437325799,
"grad_norm": 1.0703125,
"learning_rate": 0.000157,
"loss": 7.9851,
"mean_token_accuracy": 0.113057342171669,
"num_tokens": 781982.0,
"step": 315
},
{
"entropy": 8.28858814239502,
"epoch": 0.024435874918865258,
"grad_norm": 1.1484375,
"learning_rate": 0.0001595,
"loss": 8.0389,
"mean_token_accuracy": 0.11170414686203003,
"num_tokens": 795936.0,
"step": 320
},
{
"entropy": 8.194317626953126,
"epoch": 0.02481768546447253,
"grad_norm": 1.25,
"learning_rate": 0.000162,
"loss": 7.9885,
"mean_token_accuracy": 0.11449503526091576,
"num_tokens": 808509.0,
"step": 325
},
{
"entropy": 8.179093265533448,
"epoch": 0.0251994960100798,
"grad_norm": 1.296875,
"learning_rate": 0.00016450000000000001,
"loss": 7.9899,
"mean_token_accuracy": 0.11678558364510536,
"num_tokens": 822457.0,
"step": 330
},
{
"entropy": 8.29484577178955,
"epoch": 0.025581306555687067,
"grad_norm": 1.234375,
"learning_rate": 0.00016700000000000002,
"loss": 7.9877,
"mean_token_accuracy": 0.11315491572022437,
"num_tokens": 835539.0,
"step": 335
},
{
"entropy": 8.240567207336426,
"epoch": 0.02596311710129434,
"grad_norm": 1.21875,
"learning_rate": 0.00016950000000000003,
"loss": 7.9996,
"mean_token_accuracy": 0.11061925888061523,
"num_tokens": 848255.0,
"step": 340
},
{
"entropy": 8.150354766845703,
"epoch": 0.026344927646901606,
"grad_norm": 0.9375,
"learning_rate": 0.00017199999999999998,
"loss": 7.9971,
"mean_token_accuracy": 0.11511600390076637,
"num_tokens": 860940.0,
"step": 345
},
{
"entropy": 8.30101490020752,
"epoch": 0.026726738192508877,
"grad_norm": 1.1953125,
"learning_rate": 0.00017449999999999999,
"loss": 8.0025,
"mean_token_accuracy": 0.11156911998987198,
"num_tokens": 873757.0,
"step": 350
},
{
"entropy": 8.16147975921631,
"epoch": 0.027108548738116148,
"grad_norm": 1.21875,
"learning_rate": 0.000177,
"loss": 7.9879,
"mean_token_accuracy": 0.11351457163691521,
"num_tokens": 885751.0,
"step": 355
},
{
"entropy": 8.138264703750611,
"epoch": 0.027490359283723415,
"grad_norm": 1.4609375,
"learning_rate": 0.0001795,
"loss": 7.8642,
"mean_token_accuracy": 0.11878235787153243,
"num_tokens": 899203.0,
"step": 360
},
{
"entropy": 8.131235504150391,
"epoch": 0.027872169829330686,
"grad_norm": 1.3984375,
"learning_rate": 0.000182,
"loss": 7.8713,
"mean_token_accuracy": 0.12657914236187934,
"num_tokens": 911384.0,
"step": 365
},
{
"entropy": 8.151749897003175,
"epoch": 0.028253980374937957,
"grad_norm": 1.2109375,
"learning_rate": 0.0001845,
"loss": 7.9322,
"mean_token_accuracy": 0.11955902799963951,
"num_tokens": 923558.0,
"step": 370
},
{
"entropy": 8.194352340698241,
"epoch": 0.028635790920545225,
"grad_norm": 1.0234375,
"learning_rate": 0.000187,
"loss": 7.9254,
"mean_token_accuracy": 0.11569129005074501,
"num_tokens": 936309.0,
"step": 375
},
{
"entropy": 8.09564666748047,
"epoch": 0.029017601466152496,
"grad_norm": 1.1875,
"learning_rate": 0.0001895,
"loss": 7.9299,
"mean_token_accuracy": 0.11621968820691109,
"num_tokens": 947882.0,
"step": 380
},
{
"entropy": 8.115512371063232,
"epoch": 0.029399412011759767,
"grad_norm": 1.1875,
"learning_rate": 0.000192,
"loss": 7.9385,
"mean_token_accuracy": 0.11436433717608452,
"num_tokens": 960242.0,
"step": 385
},
{
"entropy": 8.190027332305908,
"epoch": 0.029781222557367034,
"grad_norm": 0.98046875,
"learning_rate": 0.0001945,
"loss": 7.9643,
"mean_token_accuracy": 0.11476908028125762,
"num_tokens": 973814.0,
"step": 390
},
{
"entropy": 8.06707820892334,
"epoch": 0.030163033102974305,
"grad_norm": 1.0859375,
"learning_rate": 0.00019700000000000002,
"loss": 7.8704,
"mean_token_accuracy": 0.12641484588384627,
"num_tokens": 985935.0,
"step": 395
},
{
"entropy": 8.11848578453064,
"epoch": 0.030544843648581572,
"grad_norm": 1.125,
"learning_rate": 0.00019950000000000002,
"loss": 7.9411,
"mean_token_accuracy": 0.11482023000717163,
"num_tokens": 997618.0,
"step": 400
},
{
"entropy": 8.1826322555542,
"epoch": 0.030926654194188843,
"grad_norm": 0.91015625,
"learning_rate": 0.000202,
"loss": 7.8844,
"mean_token_accuracy": 0.12198737263679504,
"num_tokens": 1009745.0,
"step": 405
},
{
"entropy": 8.074276161193847,
"epoch": 0.031308464739796114,
"grad_norm": 1.46875,
"learning_rate": 0.00020449999999999998,
"loss": 7.8455,
"mean_token_accuracy": 0.12325679659843444,
"num_tokens": 1022401.0,
"step": 410
},
{
"entropy": 8.10068097114563,
"epoch": 0.031690275285403385,
"grad_norm": 1.1640625,
"learning_rate": 0.000207,
"loss": 7.8965,
"mean_token_accuracy": 0.11033826768398285,
"num_tokens": 1034844.0,
"step": 415
},
{
"entropy": 7.962525510787964,
"epoch": 0.03207208583101065,
"grad_norm": 1.203125,
"learning_rate": 0.0002095,
"loss": 7.8646,
"mean_token_accuracy": 0.12236305102705955,
"num_tokens": 1046733.0,
"step": 420
},
{
"entropy": 8.19951171875,
"epoch": 0.03245389637661792,
"grad_norm": 1.2421875,
"learning_rate": 0.000212,
"loss": 7.9132,
"mean_token_accuracy": 0.11312553584575653,
"num_tokens": 1060150.0,
"step": 425
},
{
"entropy": 8.00361738204956,
"epoch": 0.03283570692222519,
"grad_norm": 1.046875,
"learning_rate": 0.0002145,
"loss": 7.8368,
"mean_token_accuracy": 0.1230610378086567,
"num_tokens": 1072178.0,
"step": 430
},
{
"entropy": 8.090557527542114,
"epoch": 0.03321751746783246,
"grad_norm": 1.15625,
"learning_rate": 0.00021700000000000002,
"loss": 7.8982,
"mean_token_accuracy": 0.11123135685920715,
"num_tokens": 1084404.0,
"step": 435
},
{
"entropy": 7.951789474487304,
"epoch": 0.03359932801343973,
"grad_norm": 1.2265625,
"learning_rate": 0.0002195,
"loss": 7.8519,
"mean_token_accuracy": 0.11904852241277694,
"num_tokens": 1096211.0,
"step": 440
},
{
"entropy": 8.07854413986206,
"epoch": 0.033981138559047004,
"grad_norm": 1.1640625,
"learning_rate": 0.000222,
"loss": 7.8994,
"mean_token_accuracy": 0.12089689522981643,
"num_tokens": 1109118.0,
"step": 445
},
{
"entropy": 8.024729442596435,
"epoch": 0.03436294910465427,
"grad_norm": 1.1015625,
"learning_rate": 0.0002245,
"loss": 7.8407,
"mean_token_accuracy": 0.11579521223902703,
"num_tokens": 1121785.0,
"step": 450
},
{
"entropy": 8.080478191375732,
"epoch": 0.03474475965026154,
"grad_norm": 1.125,
"learning_rate": 0.00022700000000000002,
"loss": 7.8267,
"mean_token_accuracy": 0.11973192021250725,
"num_tokens": 1134888.0,
"step": 455
},
{
"entropy": 7.96352367401123,
"epoch": 0.03512657019586881,
"grad_norm": 1.25,
"learning_rate": 0.00022950000000000002,
"loss": 7.7474,
"mean_token_accuracy": 0.1206622526049614,
"num_tokens": 1146777.0,
"step": 460
},
{
"entropy": 8.051364040374756,
"epoch": 0.03550838074147608,
"grad_norm": 1.0625,
"learning_rate": 0.00023200000000000003,
"loss": 7.8404,
"mean_token_accuracy": 0.11428562551736832,
"num_tokens": 1159489.0,
"step": 465
},
{
"entropy": 7.962832498550415,
"epoch": 0.03589019128708335,
"grad_norm": 1.0390625,
"learning_rate": 0.00023449999999999998,
"loss": 7.7844,
"mean_token_accuracy": 0.12907169461250306,
"num_tokens": 1171746.0,
"step": 470
},
{
"entropy": 7.957472324371338,
"epoch": 0.036272001832690616,
"grad_norm": 1.1640625,
"learning_rate": 0.000237,
"loss": 7.8531,
"mean_token_accuracy": 0.1179780401289463,
"num_tokens": 1184965.0,
"step": 475
},
{
"entropy": 8.081069564819336,
"epoch": 0.03665381237829789,
"grad_norm": 1.0625,
"learning_rate": 0.0002395,
"loss": 7.8668,
"mean_token_accuracy": 0.1181311696767807,
"num_tokens": 1198080.0,
"step": 480
},
{
"entropy": 7.988467025756836,
"epoch": 0.03703562292390516,
"grad_norm": 1.0,
"learning_rate": 0.000242,
"loss": 7.8406,
"mean_token_accuracy": 0.12035353481769562,
"num_tokens": 1210484.0,
"step": 485
},
{
"entropy": 7.958967494964599,
"epoch": 0.03741743346951243,
"grad_norm": 1.0859375,
"learning_rate": 0.0002445,
"loss": 7.7806,
"mean_token_accuracy": 0.12781597301363945,
"num_tokens": 1223524.0,
"step": 490
},
{
"entropy": 7.937625741958618,
"epoch": 0.0377992440151197,
"grad_norm": 1.078125,
"learning_rate": 0.000247,
"loss": 7.7644,
"mean_token_accuracy": 0.12598755583167076,
"num_tokens": 1234946.0,
"step": 495
},
{
"entropy": 7.959400177001953,
"epoch": 0.038181054560726964,
"grad_norm": 1.15625,
"learning_rate": 0.0002495,
"loss": 7.7763,
"mean_token_accuracy": 0.12367670983076096,
"num_tokens": 1247187.0,
"step": 500
},
{
"entropy": 7.950307416915893,
"epoch": 0.038562865106334235,
"grad_norm": 1.03125,
"learning_rate": 0.000252,
"loss": 7.8293,
"mean_token_accuracy": 0.12162172868847847,
"num_tokens": 1260218.0,
"step": 505
},
{
"entropy": 7.809165287017822,
"epoch": 0.038944675651941506,
"grad_norm": 1.109375,
"learning_rate": 0.0002545,
"loss": 7.6739,
"mean_token_accuracy": 0.13806419745087622,
"num_tokens": 1273855.0,
"step": 510
},
{
"entropy": 7.908237648010254,
"epoch": 0.03932648619754878,
"grad_norm": 1.3515625,
"learning_rate": 0.000257,
"loss": 7.7707,
"mean_token_accuracy": 0.12450363859534264,
"num_tokens": 1285909.0,
"step": 515
},
{
"entropy": 7.994627666473389,
"epoch": 0.03970829674315605,
"grad_norm": 1.3203125,
"learning_rate": 0.0002595,
"loss": 7.7752,
"mean_token_accuracy": 0.11765163615345955,
"num_tokens": 1297879.0,
"step": 520
},
{
"entropy": 7.9173308372497555,
"epoch": 0.04009010728876332,
"grad_norm": 1.1953125,
"learning_rate": 0.000262,
"loss": 7.7913,
"mean_token_accuracy": 0.12168432697653771,
"num_tokens": 1311307.0,
"step": 525
},
{
"entropy": 8.005512809753418,
"epoch": 0.04047191783437058,
"grad_norm": 1.0859375,
"learning_rate": 0.00026450000000000003,
"loss": 7.7981,
"mean_token_accuracy": 0.11987643837928771,
"num_tokens": 1324113.0,
"step": 530
},
{
"entropy": 7.827685785293579,
"epoch": 0.040853728379977854,
"grad_norm": 1.0,
"learning_rate": 0.00026700000000000004,
"loss": 7.7048,
"mean_token_accuracy": 0.11903792023658752,
"num_tokens": 1336052.0,
"step": 535
},
{
"entropy": 7.9320954322814945,
"epoch": 0.041235538925585125,
"grad_norm": 1.046875,
"learning_rate": 0.00026950000000000005,
"loss": 7.7457,
"mean_token_accuracy": 0.11790703907608986,
"num_tokens": 1347661.0,
"step": 540
},
{
"entropy": 7.904759407043457,
"epoch": 0.041617349471192396,
"grad_norm": 1.3046875,
"learning_rate": 0.00027200000000000005,
"loss": 7.7597,
"mean_token_accuracy": 0.1271845705807209,
"num_tokens": 1360956.0,
"step": 545
},
{
"entropy": 7.884268856048584,
"epoch": 0.041999160016799666,
"grad_norm": 1.21875,
"learning_rate": 0.0002745,
"loss": 7.6907,
"mean_token_accuracy": 0.12421448156237602,
"num_tokens": 1373593.0,
"step": 550
},
{
"entropy": 7.827823638916016,
"epoch": 0.04238097056240693,
"grad_norm": 1.046875,
"learning_rate": 0.000277,
"loss": 7.7223,
"mean_token_accuracy": 0.1241831436753273,
"num_tokens": 1385340.0,
"step": 555
},
{
"entropy": 7.889366149902344,
"epoch": 0.0427627811080142,
"grad_norm": 0.9296875,
"learning_rate": 0.0002795,
"loss": 7.7836,
"mean_token_accuracy": 0.12166346386075019,
"num_tokens": 1398419.0,
"step": 560
},
{
"entropy": 7.861316871643067,
"epoch": 0.04314459165362147,
"grad_norm": 1.25,
"learning_rate": 0.00028199999999999997,
"loss": 7.7168,
"mean_token_accuracy": 0.12731389701366425,
"num_tokens": 1411542.0,
"step": 565
},
{
"entropy": 7.757087516784668,
"epoch": 0.04352640219922874,
"grad_norm": 1.125,
"learning_rate": 0.0002845,
"loss": 7.7696,
"mean_token_accuracy": 0.12272902503609658,
"num_tokens": 1425100.0,
"step": 570
},
{
"entropy": 7.878741073608398,
"epoch": 0.043908212744836014,
"grad_norm": 1.1015625,
"learning_rate": 0.000287,
"loss": 7.6386,
"mean_token_accuracy": 0.13037696853280067,
"num_tokens": 1436652.0,
"step": 575
},
{
"entropy": 7.814185047149659,
"epoch": 0.044290023290443285,
"grad_norm": 1.0234375,
"learning_rate": 0.0002895,
"loss": 7.7382,
"mean_token_accuracy": 0.11856948360800743,
"num_tokens": 1449190.0,
"step": 580
},
{
"entropy": 7.764650344848633,
"epoch": 0.04467183383605055,
"grad_norm": 1.1484375,
"learning_rate": 0.000292,
"loss": 7.6459,
"mean_token_accuracy": 0.12288401573896408,
"num_tokens": 1461938.0,
"step": 585
},
{
"entropy": 7.873277187347412,
"epoch": 0.04505364438165782,
"grad_norm": 1.171875,
"learning_rate": 0.0002945,
"loss": 7.686,
"mean_token_accuracy": 0.12823309153318405,
"num_tokens": 1474736.0,
"step": 590
},
{
"entropy": 7.764197444915771,
"epoch": 0.04543545492726509,
"grad_norm": 1.1640625,
"learning_rate": 0.000297,
"loss": 7.761,
"mean_token_accuracy": 0.12043945714831353,
"num_tokens": 1488776.0,
"step": 595
},
{
"entropy": 7.895608949661255,
"epoch": 0.04581726547287236,
"grad_norm": 1.15625,
"learning_rate": 0.0002995,
"loss": 7.6935,
"mean_token_accuracy": 0.12775885909795762,
"num_tokens": 1501355.0,
"step": 600
},
{
"entropy": 7.791248989105225,
"epoch": 0.04619907601847963,
"grad_norm": 1.078125,
"learning_rate": 0.000302,
"loss": 7.7223,
"mean_token_accuracy": 0.1267842173576355,
"num_tokens": 1512481.0,
"step": 605
},
{
"entropy": 7.843862771987915,
"epoch": 0.0465808865640869,
"grad_norm": 1.1484375,
"learning_rate": 0.0003045,
"loss": 7.6304,
"mean_token_accuracy": 0.12487628906965256,
"num_tokens": 1524113.0,
"step": 610
},
{
"entropy": 7.761060047149658,
"epoch": 0.04696269710969417,
"grad_norm": 0.94921875,
"learning_rate": 0.000307,
"loss": 7.6886,
"mean_token_accuracy": 0.12064972445368767,
"num_tokens": 1537217.0,
"step": 615
},
{
"entropy": 7.829102945327759,
"epoch": 0.04734450765530144,
"grad_norm": 0.97265625,
"learning_rate": 0.0003095,
"loss": 7.7095,
"mean_token_accuracy": 0.12352586463093758,
"num_tokens": 1549195.0,
"step": 620
},
{
"entropy": 7.797214317321777,
"epoch": 0.04772631820090871,
"grad_norm": 1.09375,
"learning_rate": 0.000312,
"loss": 7.6685,
"mean_token_accuracy": 0.1338474079966545,
"num_tokens": 1561415.0,
"step": 625
},
{
"entropy": 7.677298212051392,
"epoch": 0.04810812874651598,
"grad_norm": 1.125,
"learning_rate": 0.0003145,
"loss": 7.6156,
"mean_token_accuracy": 0.12719253152608873,
"num_tokens": 1572907.0,
"step": 630
},
{
"entropy": 7.779621505737305,
"epoch": 0.048489939292123245,
"grad_norm": 1.171875,
"learning_rate": 0.000317,
"loss": 7.6295,
"mean_token_accuracy": 0.1251477211713791,
"num_tokens": 1585750.0,
"step": 635
},
{
"entropy": 7.745410346984864,
"epoch": 0.048871749837730516,
"grad_norm": 1.3125,
"learning_rate": 0.0003195,
"loss": 7.6397,
"mean_token_accuracy": 0.1262856461107731,
"num_tokens": 1598663.0,
"step": 640
},
{
"entropy": 7.74163031578064,
"epoch": 0.04925356038333779,
"grad_norm": 1.0859375,
"learning_rate": 0.000322,
"loss": 7.6479,
"mean_token_accuracy": 0.12824388071894646,
"num_tokens": 1612892.0,
"step": 645
},
{
"entropy": 7.63819317817688,
"epoch": 0.04963537092894506,
"grad_norm": 1.1328125,
"learning_rate": 0.00032450000000000003,
"loss": 7.5937,
"mean_token_accuracy": 0.13200653344392776,
"num_tokens": 1625309.0,
"step": 650
},
{
"entropy": 7.840318441390991,
"epoch": 0.05001718147455233,
"grad_norm": 1.0625,
"learning_rate": 0.00032700000000000003,
"loss": 7.6778,
"mean_token_accuracy": 0.12029165998101235,
"num_tokens": 1637517.0,
"step": 655
},
{
"entropy": 7.758690071105957,
"epoch": 0.0503989920201596,
"grad_norm": 1.1171875,
"learning_rate": 0.00032950000000000004,
"loss": 7.645,
"mean_token_accuracy": 0.12100816667079925,
"num_tokens": 1649920.0,
"step": 660
},
{
"entropy": 7.688795471191407,
"epoch": 0.050780802565766864,
"grad_norm": 1.0390625,
"learning_rate": 0.00033200000000000005,
"loss": 7.5524,
"mean_token_accuracy": 0.1325525127351284,
"num_tokens": 1661522.0,
"step": 665
},
{
"entropy": 7.659743690490723,
"epoch": 0.051162613111374135,
"grad_norm": 0.953125,
"learning_rate": 0.00033450000000000005,
"loss": 7.6219,
"mean_token_accuracy": 0.12344450354576111,
"num_tokens": 1674237.0,
"step": 670
},
{
"entropy": 7.706264543533325,
"epoch": 0.051544423656981406,
"grad_norm": 1.0390625,
"learning_rate": 0.000337,
"loss": 7.5183,
"mean_token_accuracy": 0.13378539383411409,
"num_tokens": 1685816.0,
"step": 675
},
{
"entropy": 7.726587629318237,
"epoch": 0.05192623420258868,
"grad_norm": 1.171875,
"learning_rate": 0.0003395,
"loss": 7.6156,
"mean_token_accuracy": 0.12633788362145423,
"num_tokens": 1698047.0,
"step": 680
},
{
"entropy": 7.485805225372315,
"epoch": 0.05230804474819595,
"grad_norm": 0.9375,
"learning_rate": 0.000342,
"loss": 7.5241,
"mean_token_accuracy": 0.13966493904590607,
"num_tokens": 1711726.0,
"step": 685
},
{
"entropy": 7.724063396453857,
"epoch": 0.05268985529380321,
"grad_norm": 1.0234375,
"learning_rate": 0.00034449999999999997,
"loss": 7.631,
"mean_token_accuracy": 0.12105451375246049,
"num_tokens": 1724867.0,
"step": 690
},
{
"entropy": 7.6883948802947994,
"epoch": 0.05307166583941048,
"grad_norm": 1.03125,
"learning_rate": 0.000347,
"loss": 7.6295,
"mean_token_accuracy": 0.12337101623415947,
"num_tokens": 1737366.0,
"step": 695
},
{
"entropy": 7.677063703536987,
"epoch": 0.053453476385017754,
"grad_norm": 1.09375,
"learning_rate": 0.0003495,
"loss": 7.5468,
"mean_token_accuracy": 0.12484264373779297,
"num_tokens": 1749338.0,
"step": 700
},
{
"entropy": 7.653280639648438,
"epoch": 0.053835286930625025,
"grad_norm": 1.0,
"learning_rate": 0.000352,
"loss": 7.5636,
"mean_token_accuracy": 0.12633164450526238,
"num_tokens": 1761957.0,
"step": 705
},
{
"entropy": 7.726439523696899,
"epoch": 0.054217097476232295,
"grad_norm": 0.96875,
"learning_rate": 0.0003545,
"loss": 7.5609,
"mean_token_accuracy": 0.12474608942866325,
"num_tokens": 1774092.0,
"step": 710
},
{
"entropy": 7.627110290527344,
"epoch": 0.054598908021839566,
"grad_norm": 1.0234375,
"learning_rate": 0.000357,
"loss": 7.5663,
"mean_token_accuracy": 0.12494908422231674,
"num_tokens": 1787396.0,
"step": 715
},
{
"entropy": 7.668584871292114,
"epoch": 0.05498071856744683,
"grad_norm": 1.1640625,
"learning_rate": 0.0003595,
"loss": 7.5086,
"mean_token_accuracy": 0.13190939873456956,
"num_tokens": 1798719.0,
"step": 720
},
{
"entropy": 7.653689813613892,
"epoch": 0.0553625291130541,
"grad_norm": 1.015625,
"learning_rate": 0.000362,
"loss": 7.5602,
"mean_token_accuracy": 0.12464875876903533,
"num_tokens": 1811746.0,
"step": 725
},
{
"entropy": 7.689782953262329,
"epoch": 0.05574433965866137,
"grad_norm": 0.96875,
"learning_rate": 0.0003645,
"loss": 7.6788,
"mean_token_accuracy": 0.12057624980807305,
"num_tokens": 1825931.0,
"step": 730
},
{
"entropy": 7.587789821624756,
"epoch": 0.05612615020426864,
"grad_norm": 1.0625,
"learning_rate": 0.000367,
"loss": 7.5349,
"mean_token_accuracy": 0.1323995038866997,
"num_tokens": 1837780.0,
"step": 735
},
{
"entropy": 7.646715497970581,
"epoch": 0.056507960749875914,
"grad_norm": 1.09375,
"learning_rate": 0.0003695,
"loss": 7.5447,
"mean_token_accuracy": 0.1258799411356449,
"num_tokens": 1850313.0,
"step": 740
},
{
"entropy": 7.690730428695678,
"epoch": 0.05688977129548318,
"grad_norm": 1.2109375,
"learning_rate": 0.000372,
"loss": 7.5674,
"mean_token_accuracy": 0.12429766207933426,
"num_tokens": 1862401.0,
"step": 745
},
{
"entropy": 7.547059154510498,
"epoch": 0.05727158184109045,
"grad_norm": 1.1015625,
"learning_rate": 0.0003745,
"loss": 7.4847,
"mean_token_accuracy": 0.1276401810348034,
"num_tokens": 1875725.0,
"step": 750
},
{
"entropy": 7.501403713226319,
"epoch": 0.05765339238669772,
"grad_norm": 1.0390625,
"learning_rate": 0.000377,
"loss": 7.5628,
"mean_token_accuracy": 0.12886957228183746,
"num_tokens": 1888468.0,
"step": 755
},
{
"entropy": 7.6124814510345455,
"epoch": 0.05803520293230499,
"grad_norm": 1.0234375,
"learning_rate": 0.0003795,
"loss": 7.4506,
"mean_token_accuracy": 0.12971155121922492,
"num_tokens": 1901050.0,
"step": 760
},
{
"entropy": 7.724295949935913,
"epoch": 0.05841701347791226,
"grad_norm": 0.91796875,
"learning_rate": 0.000382,
"loss": 7.5653,
"mean_token_accuracy": 0.12600777819752693,
"num_tokens": 1913757.0,
"step": 765
},
{
"entropy": 7.493232536315918,
"epoch": 0.05879882402351953,
"grad_norm": 1.015625,
"learning_rate": 0.0003845,
"loss": 7.4896,
"mean_token_accuracy": 0.12974480390548707,
"num_tokens": 1924991.0,
"step": 770
},
{
"entropy": 7.642123174667359,
"epoch": 0.0591806345691268,
"grad_norm": 0.9765625,
"learning_rate": 0.00038700000000000003,
"loss": 7.5038,
"mean_token_accuracy": 0.13034126982092858,
"num_tokens": 1937141.0,
"step": 775
},
{
"entropy": 7.6067742824554445,
"epoch": 0.05956244511473407,
"grad_norm": 1.09375,
"learning_rate": 0.00038950000000000003,
"loss": 7.5088,
"mean_token_accuracy": 0.12796926721930504,
"num_tokens": 1949162.0,
"step": 780
},
{
"entropy": 7.511234521865845,
"epoch": 0.05994425566034134,
"grad_norm": 1.1015625,
"learning_rate": 0.00039200000000000004,
"loss": 7.5461,
"mean_token_accuracy": 0.12637099325656892,
"num_tokens": 1961819.0,
"step": 785
},
{
"entropy": 7.51845498085022,
"epoch": 0.06032606620594861,
"grad_norm": 1.1875,
"learning_rate": 0.00039450000000000005,
"loss": 7.479,
"mean_token_accuracy": 0.1301673784852028,
"num_tokens": 1973986.0,
"step": 790
},
{
"entropy": 7.56482458114624,
"epoch": 0.06070787675155588,
"grad_norm": 1.015625,
"learning_rate": 0.00039700000000000005,
"loss": 7.4703,
"mean_token_accuracy": 0.12951676845550536,
"num_tokens": 1986346.0,
"step": 795
},
{
"entropy": 7.587308406829834,
"epoch": 0.061089687297163145,
"grad_norm": 0.99609375,
"learning_rate": 0.0003995,
"loss": 7.4923,
"mean_token_accuracy": 0.13167140632867813,
"num_tokens": 1998482.0,
"step": 800
},
{
"entropy": 7.487084817886353,
"epoch": 0.061471497842770416,
"grad_norm": 1.21875,
"learning_rate": 0.000402,
"loss": 7.4136,
"mean_token_accuracy": 0.1325934983789921,
"num_tokens": 2009862.0,
"step": 805
},
{
"entropy": 7.508486318588257,
"epoch": 0.06185330838837769,
"grad_norm": 1.0,
"learning_rate": 0.0004045,
"loss": 7.4456,
"mean_token_accuracy": 0.13007148131728172,
"num_tokens": 2022395.0,
"step": 810
},
{
"entropy": 7.574994611740112,
"epoch": 0.06223511893398496,
"grad_norm": 1.0859375,
"learning_rate": 0.00040699999999999997,
"loss": 7.4549,
"mean_token_accuracy": 0.12644017487764359,
"num_tokens": 2034241.0,
"step": 815
},
{
"entropy": 7.48778190612793,
"epoch": 0.06261692947959223,
"grad_norm": 0.9609375,
"learning_rate": 0.0004095,
"loss": 7.4462,
"mean_token_accuracy": 0.12410394623875617,
"num_tokens": 2046461.0,
"step": 820
},
{
"entropy": 7.518596124649048,
"epoch": 0.06299874002519949,
"grad_norm": 1.0703125,
"learning_rate": 0.000412,
"loss": 7.4244,
"mean_token_accuracy": 0.12908011823892593,
"num_tokens": 2058354.0,
"step": 825
},
{
"entropy": 7.530619955062866,
"epoch": 0.06338055057080677,
"grad_norm": 0.97265625,
"learning_rate": 0.0004145,
"loss": 7.548,
"mean_token_accuracy": 0.12275430709123611,
"num_tokens": 2071544.0,
"step": 830
},
{
"entropy": 7.5850340843200685,
"epoch": 0.06376236111641403,
"grad_norm": 0.94921875,
"learning_rate": 0.000417,
"loss": 7.4707,
"mean_token_accuracy": 0.1265728145837784,
"num_tokens": 2085523.0,
"step": 835
},
{
"entropy": 7.422417163848877,
"epoch": 0.0641441716620213,
"grad_norm": 1.171875,
"learning_rate": 0.0004195,
"loss": 7.4665,
"mean_token_accuracy": 0.12771449163556098,
"num_tokens": 2098386.0,
"step": 840
},
{
"entropy": 7.5254480838775635,
"epoch": 0.06452598220762858,
"grad_norm": 0.99609375,
"learning_rate": 0.000422,
"loss": 7.4525,
"mean_token_accuracy": 0.12055407911539077,
"num_tokens": 2111610.0,
"step": 845
},
{
"entropy": 7.508930778503418,
"epoch": 0.06490779275323584,
"grad_norm": 1.0703125,
"learning_rate": 0.0004245,
"loss": 7.4697,
"mean_token_accuracy": 0.12965044155716896,
"num_tokens": 2124102.0,
"step": 850
},
{
"entropy": 7.527687978744507,
"epoch": 0.06528960329884312,
"grad_norm": 1.0859375,
"learning_rate": 0.000427,
"loss": 7.4583,
"mean_token_accuracy": 0.13157543540000916,
"num_tokens": 2135817.0,
"step": 855
},
{
"entropy": 7.521559953689575,
"epoch": 0.06567141384445038,
"grad_norm": 1.0546875,
"learning_rate": 0.0004295,
"loss": 7.3957,
"mean_token_accuracy": 0.13194996267557144,
"num_tokens": 2148537.0,
"step": 860
},
{
"entropy": 7.478877019882202,
"epoch": 0.06605322439005765,
"grad_norm": 0.953125,
"learning_rate": 0.000432,
"loss": 7.5064,
"mean_token_accuracy": 0.12444748133420944,
"num_tokens": 2160419.0,
"step": 865
},
{
"entropy": 7.536518478393555,
"epoch": 0.06643503493566492,
"grad_norm": 1.203125,
"learning_rate": 0.0004345,
"loss": 7.4643,
"mean_token_accuracy": 0.12732195481657982,
"num_tokens": 2172778.0,
"step": 870
},
{
"entropy": 7.304726934432983,
"epoch": 0.06681684548127219,
"grad_norm": 1.390625,
"learning_rate": 0.000437,
"loss": 7.3518,
"mean_token_accuracy": 0.13419221788644792,
"num_tokens": 2185479.0,
"step": 875
},
{
"entropy": 7.563678312301636,
"epoch": 0.06719865602687947,
"grad_norm": 1.140625,
"learning_rate": 0.0004395,
"loss": 7.4242,
"mean_token_accuracy": 0.12837793454527854,
"num_tokens": 2196803.0,
"step": 880
},
{
"entropy": 7.466328096389771,
"epoch": 0.06758046657248673,
"grad_norm": 0.921875,
"learning_rate": 0.000442,
"loss": 7.4432,
"mean_token_accuracy": 0.12447937279939651,
"num_tokens": 2209525.0,
"step": 885
},
{
"entropy": 7.418173170089721,
"epoch": 0.06796227711809401,
"grad_norm": 1.0859375,
"learning_rate": 0.0004445,
"loss": 7.4005,
"mean_token_accuracy": 0.13152864798903466,
"num_tokens": 2220571.0,
"step": 890
},
{
"entropy": 7.521706199645996,
"epoch": 0.06834408766370127,
"grad_norm": 1.0703125,
"learning_rate": 0.000447,
"loss": 7.4849,
"mean_token_accuracy": 0.1335865378379822,
"num_tokens": 2233913.0,
"step": 895
},
{
"entropy": 7.453684616088867,
"epoch": 0.06872589820930854,
"grad_norm": 0.95703125,
"learning_rate": 0.00044950000000000003,
"loss": 7.4764,
"mean_token_accuracy": 0.12939356118440629,
"num_tokens": 2247323.0,
"step": 900
},
{
"entropy": 7.3101033687591555,
"epoch": 0.06910770875491581,
"grad_norm": 1.1015625,
"learning_rate": 0.00045200000000000004,
"loss": 7.356,
"mean_token_accuracy": 0.13926377072930335,
"num_tokens": 2258252.0,
"step": 905
},
{
"entropy": 7.4975191116333,
"epoch": 0.06948951930052308,
"grad_norm": 0.86328125,
"learning_rate": 0.00045450000000000004,
"loss": 7.436,
"mean_token_accuracy": 0.13080064281821252,
"num_tokens": 2272626.0,
"step": 910
},
{
"entropy": 7.327397203445434,
"epoch": 0.06987132984613036,
"grad_norm": 1.0234375,
"learning_rate": 0.00045700000000000005,
"loss": 7.393,
"mean_token_accuracy": 0.13220079839229584,
"num_tokens": 2285060.0,
"step": 915
},
{
"entropy": 7.492992210388183,
"epoch": 0.07025314039173762,
"grad_norm": 1.0546875,
"learning_rate": 0.00045950000000000006,
"loss": 7.4141,
"mean_token_accuracy": 0.12240867912769318,
"num_tokens": 2297661.0,
"step": 920
},
{
"entropy": 7.443741703033448,
"epoch": 0.07063495093734488,
"grad_norm": 1.1171875,
"learning_rate": 0.000462,
"loss": 7.426,
"mean_token_accuracy": 0.13318592980504035,
"num_tokens": 2311333.0,
"step": 925
},
{
"entropy": 7.288962745666504,
"epoch": 0.07101676148295216,
"grad_norm": 1.0625,
"learning_rate": 0.0004645,
"loss": 7.3108,
"mean_token_accuracy": 0.14332053065299988,
"num_tokens": 2324081.0,
"step": 930
},
{
"entropy": 7.421329927444458,
"epoch": 0.07139857202855943,
"grad_norm": 0.96484375,
"learning_rate": 0.000467,
"loss": 7.3558,
"mean_token_accuracy": 0.13017595186829567,
"num_tokens": 2337193.0,
"step": 935
},
{
"entropy": 7.282610893249512,
"epoch": 0.0717803825741667,
"grad_norm": 1.0390625,
"learning_rate": 0.0004695,
"loss": 7.4696,
"mean_token_accuracy": 0.12410783097147941,
"num_tokens": 2351352.0,
"step": 940
},
{
"entropy": 7.487767839431763,
"epoch": 0.07216219311977397,
"grad_norm": 0.97265625,
"learning_rate": 0.000472,
"loss": 7.403,
"mean_token_accuracy": 0.12429638803005219,
"num_tokens": 2363855.0,
"step": 945
},
{
"entropy": 7.32493405342102,
"epoch": 0.07254400366538123,
"grad_norm": 1.046875,
"learning_rate": 0.0004745,
"loss": 7.2879,
"mean_token_accuracy": 0.13387082144618034,
"num_tokens": 2374697.0,
"step": 950
},
{
"entropy": 7.315332221984863,
"epoch": 0.07292581421098851,
"grad_norm": 1.0390625,
"learning_rate": 0.000477,
"loss": 7.308,
"mean_token_accuracy": 0.1337247148156166,
"num_tokens": 2386570.0,
"step": 955
},
{
"entropy": 7.331861591339111,
"epoch": 0.07330762475659577,
"grad_norm": 1.1015625,
"learning_rate": 0.0004795,
"loss": 7.281,
"mean_token_accuracy": 0.14531914442777633,
"num_tokens": 2397392.0,
"step": 960
},
{
"entropy": 7.260980463027954,
"epoch": 0.07368943530220305,
"grad_norm": 1.0859375,
"learning_rate": 0.000482,
"loss": 7.241,
"mean_token_accuracy": 0.140303585678339,
"num_tokens": 2409394.0,
"step": 965
},
{
"entropy": 7.264951181411743,
"epoch": 0.07407124584781032,
"grad_norm": 1.0625,
"learning_rate": 0.0004845,
"loss": 7.2926,
"mean_token_accuracy": 0.1356895796954632,
"num_tokens": 2421382.0,
"step": 970
},
{
"entropy": 7.294839763641358,
"epoch": 0.07445305639341758,
"grad_norm": 1.125,
"learning_rate": 0.000487,
"loss": 7.2305,
"mean_token_accuracy": 0.14973011314868928,
"num_tokens": 2433122.0,
"step": 975
},
{
"entropy": 7.3143617630004885,
"epoch": 0.07483486693902486,
"grad_norm": 1.0390625,
"learning_rate": 0.0004895,
"loss": 7.3398,
"mean_token_accuracy": 0.12803151831030846,
"num_tokens": 2444938.0,
"step": 980
},
{
"entropy": 7.352199697494507,
"epoch": 0.07521667748463212,
"grad_norm": 1.0546875,
"learning_rate": 0.000492,
"loss": 7.3174,
"mean_token_accuracy": 0.13830707520246505,
"num_tokens": 2457195.0,
"step": 985
},
{
"entropy": 7.230789470672607,
"epoch": 0.0755984880302394,
"grad_norm": 0.9609375,
"learning_rate": 0.0004945,
"loss": 7.2952,
"mean_token_accuracy": 0.13374050483107566,
"num_tokens": 2469844.0,
"step": 990
},
{
"entropy": 7.321718025207519,
"epoch": 0.07598029857584666,
"grad_norm": 1.0390625,
"learning_rate": 0.000497,
"loss": 7.243,
"mean_token_accuracy": 0.12716111689805984,
"num_tokens": 2482499.0,
"step": 995
},
{
"entropy": 7.3011983871459964,
"epoch": 0.07636210912145393,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995,
"loss": 7.3021,
"mean_token_accuracy": 0.12945102751255036,
"num_tokens": 2495190.0,
"step": 1000
},
{
"entropy": 7.2844287872314455,
"epoch": 0.0767439196670612,
"grad_norm": 1.0,
"learning_rate": 0.000499998026082006,
"loss": 7.3269,
"mean_token_accuracy": 0.13009936213493348,
"num_tokens": 2508382.0,
"step": 1005
},
{
"entropy": 7.271711206436157,
"epoch": 0.07712573021266847,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999900070995136,
"loss": 7.3099,
"mean_token_accuracy": 0.1319010838866234,
"num_tokens": 2521116.0,
"step": 1010
},
{
"entropy": 7.27166862487793,
"epoch": 0.07750754075827575,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999758199023239,
"loss": 7.3058,
"mean_token_accuracy": 0.1321793980896473,
"num_tokens": 2532921.0,
"step": 1015
},
{
"entropy": 7.230968236923218,
"epoch": 0.07788935130388301,
"grad_norm": 1.078125,
"learning_rate": 0.0004999554648793858,
"loss": 7.3285,
"mean_token_accuracy": 0.13184018954634666,
"num_tokens": 2545066.0,
"step": 1020
},
{
"entropy": 7.381369876861572,
"epoch": 0.07827116184949029,
"grad_norm": 0.96875,
"learning_rate": 0.0004999289425887425,
"loss": 7.3144,
"mean_token_accuracy": 0.13187807202339172,
"num_tokens": 2558406.0,
"step": 1025
},
{
"entropy": 7.266106033325196,
"epoch": 0.07865297239509755,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998962537575161,
"loss": 7.2969,
"mean_token_accuracy": 0.13618498742580415,
"num_tokens": 2570977.0,
"step": 1030
},
{
"entropy": 7.252181482315064,
"epoch": 0.07903478294070482,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998573992818874,
"loss": 7.2746,
"mean_token_accuracy": 0.1367061749100685,
"num_tokens": 2582459.0,
"step": 1035
},
{
"entropy": 7.311231231689453,
"epoch": 0.0794165934863121,
"grad_norm": 1.046875,
"learning_rate": 0.0004998123802270715,
"loss": 7.2741,
"mean_token_accuracy": 0.13370620161294938,
"num_tokens": 2597437.0,
"step": 1040
},
{
"entropy": 7.280791568756103,
"epoch": 0.07979840403191936,
"grad_norm": 1.0,
"learning_rate": 0.0004997611978272886,
"loss": 7.2458,
"mean_token_accuracy": 0.13519275933504105,
"num_tokens": 2609550.0,
"step": 1045
},
{
"entropy": 7.248421764373779,
"epoch": 0.08018021457752664,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997038534857298,
"loss": 7.3345,
"mean_token_accuracy": 0.12652327120304108,
"num_tokens": 2622182.0,
"step": 1050
},
{
"entropy": 7.309385108947754,
"epoch": 0.0805620251231339,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996403487745194,
"loss": 7.3162,
"mean_token_accuracy": 0.13302057534456252,
"num_tokens": 2634501.0,
"step": 1055
},
{
"entropy": 7.3574216842651365,
"epoch": 0.08094383566874117,
"grad_norm": 0.89453125,
"learning_rate": 0.000499570685434671,
"loss": 7.2931,
"mean_token_accuracy": 0.12851919308304788,
"num_tokens": 2648059.0,
"step": 1060
},
{
"entropy": 7.23109016418457,
"epoch": 0.08132564621434844,
"grad_norm": 1.4609375,
"learning_rate": 0.0004994948653760405,
"loss": 7.2801,
"mean_token_accuracy": 0.12933025807142257,
"num_tokens": 2661120.0,
"step": 1065
},
{
"entropy": 7.3374021530151365,
"epoch": 0.08170745675995571,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994128906772729,
"loss": 7.2938,
"mean_token_accuracy": 0.12528049424290658,
"num_tokens": 2673047.0,
"step": 1070
},
{
"entropy": 7.225849246978759,
"epoch": 0.08208926730556299,
"grad_norm": 0.99609375,
"learning_rate": 0.000499324763585746,
"loss": 7.2236,
"mean_token_accuracy": 0.13942873403429984,
"num_tokens": 2685854.0,
"step": 1075
},
{
"entropy": 7.18076605796814,
"epoch": 0.08247107785117025,
"grad_norm": 0.95703125,
"learning_rate": 0.0004992304865175085,
"loss": 7.2302,
"mean_token_accuracy": 0.13722492158412933,
"num_tokens": 2697221.0,
"step": 1080
},
{
"entropy": 7.301717758178711,
"epoch": 0.08285288839677751,
"grad_norm": 1.125,
"learning_rate": 0.0004991300620572138,
"loss": 7.2891,
"mean_token_accuracy": 0.1294270284473896,
"num_tokens": 2709300.0,
"step": 1085
},
{
"entropy": 7.199479055404663,
"epoch": 0.08323469894238479,
"grad_norm": 0.96484375,
"learning_rate": 0.0004990234929580494,
"loss": 7.2945,
"mean_token_accuracy": 0.12281102836132049,
"num_tokens": 2721754.0,
"step": 1090
},
{
"entropy": 7.232211399078369,
"epoch": 0.08361650948799206,
"grad_norm": 1.2734375,
"learning_rate": 0.0004989107821416609,
"loss": 7.2394,
"mean_token_accuracy": 0.14128851890563965,
"num_tokens": 2734629.0,
"step": 1095
},
{
"entropy": 7.24570665359497,
"epoch": 0.08399832003359933,
"grad_norm": 1.046875,
"learning_rate": 0.0004987919326980723,
"loss": 7.1761,
"mean_token_accuracy": 0.13926326259970664,
"num_tokens": 2745917.0,
"step": 1100
},
{
"entropy": 7.159485530853272,
"epoch": 0.0843801305792066,
"grad_norm": 1.03125,
"learning_rate": 0.0004986669478856011,
"loss": 7.1583,
"mean_token_accuracy": 0.13639277145266532,
"num_tokens": 2758823.0,
"step": 1105
},
{
"entropy": 7.099758672714233,
"epoch": 0.08476194112481386,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985358311307688,
"loss": 7.2162,
"mean_token_accuracy": 0.1342034213244915,
"num_tokens": 2771074.0,
"step": 1110
},
{
"entropy": 7.159941816329956,
"epoch": 0.08514375167042114,
"grad_norm": 0.953125,
"learning_rate": 0.0004983985860282081,
"loss": 7.1831,
"mean_token_accuracy": 0.14291902631521225,
"num_tokens": 2782452.0,
"step": 1115
},
{
"entropy": 7.25389347076416,
"epoch": 0.0855255622160284,
"grad_norm": 1.0234375,
"learning_rate": 0.0004982552163405623,
"loss": 7.2808,
"mean_token_accuracy": 0.13367804288864135,
"num_tokens": 2794409.0,
"step": 1120
},
{
"entropy": 7.203687047958374,
"epoch": 0.08590737276163568,
"grad_norm": 0.92578125,
"learning_rate": 0.0004981057259983839,
"loss": 7.199,
"mean_token_accuracy": 0.13216371312737465,
"num_tokens": 2807153.0,
"step": 1125
},
{
"entropy": 7.218121194839478,
"epoch": 0.08628918330724294,
"grad_norm": 0.95703125,
"learning_rate": 0.0004979501191000262,
"loss": 7.2498,
"mean_token_accuracy": 0.12889372780919076,
"num_tokens": 2818876.0,
"step": 1130
},
{
"entropy": 7.169486379623413,
"epoch": 0.08667099385285021,
"grad_norm": 1.0546875,
"learning_rate": 0.0004977883999115311,
"loss": 7.17,
"mean_token_accuracy": 0.14080205485224723,
"num_tokens": 2830558.0,
"step": 1135
},
{
"entropy": 7.224482917785645,
"epoch": 0.08705280439845749,
"grad_norm": 1.1328125,
"learning_rate": 0.0004976205728665113,
"loss": 7.2366,
"mean_token_accuracy": 0.13714844435453416,
"num_tokens": 2843574.0,
"step": 1140
},
{
"entropy": 7.168094062805176,
"epoch": 0.08743461494406475,
"grad_norm": 1.03125,
"learning_rate": 0.0004974466425660307,
"loss": 7.2152,
"mean_token_accuracy": 0.13099872693419456,
"num_tokens": 2855601.0,
"step": 1145
},
{
"entropy": 7.221179533004761,
"epoch": 0.08781642548967203,
"grad_norm": 0.95703125,
"learning_rate": 0.0004972666137784759,
"loss": 7.1849,
"mean_token_accuracy": 0.13660483956336975,
"num_tokens": 2867881.0,
"step": 1150
},
{
"entropy": 7.176160526275635,
"epoch": 0.08819823603527929,
"grad_norm": 0.93359375,
"learning_rate": 0.0004970804914394271,
"loss": 7.1732,
"mean_token_accuracy": 0.13679502308368682,
"num_tokens": 2879924.0,
"step": 1155
},
{
"entropy": 7.203902864456177,
"epoch": 0.08858004658088657,
"grad_norm": 1.0234375,
"learning_rate": 0.0004968882806515225,
"loss": 7.1931,
"mean_token_accuracy": 0.13587966412305832,
"num_tokens": 2892208.0,
"step": 1160
},
{
"entropy": 7.066990375518799,
"epoch": 0.08896185712649383,
"grad_norm": 1.0390625,
"learning_rate": 0.0004966899866843177,
"loss": 7.1121,
"mean_token_accuracy": 0.15332832857966422,
"num_tokens": 2904837.0,
"step": 1165
},
{
"entropy": 7.1723103523254395,
"epoch": 0.0893436676721011,
"grad_norm": 0.9453125,
"learning_rate": 0.000496485614974142,
"loss": 7.1617,
"mean_token_accuracy": 0.13764103949069978,
"num_tokens": 2917151.0,
"step": 1170
},
{
"entropy": 7.128491449356079,
"epoch": 0.08972547821770838,
"grad_norm": 0.96484375,
"learning_rate": 0.0004962751711239492,
"loss": 7.0893,
"mean_token_accuracy": 0.14346047043800353,
"num_tokens": 2929214.0,
"step": 1175
},
{
"entropy": 7.145873880386352,
"epoch": 0.09010728876331564,
"grad_norm": 1.0078125,
"learning_rate": 0.0004960586609031636,
"loss": 7.1499,
"mean_token_accuracy": 0.13344688415527345,
"num_tokens": 2940374.0,
"step": 1180
},
{
"entropy": 7.090103673934936,
"epoch": 0.09048909930892292,
"grad_norm": 0.94921875,
"learning_rate": 0.0004958360902475224,
"loss": 7.2597,
"mean_token_accuracy": 0.12683168947696685,
"num_tokens": 2953690.0,
"step": 1185
},
{
"entropy": 7.255402946472168,
"epoch": 0.09087090985453018,
"grad_norm": 1.0859375,
"learning_rate": 0.0004956074652589125,
"loss": 7.1885,
"mean_token_accuracy": 0.14467285722494125,
"num_tokens": 2967441.0,
"step": 1190
},
{
"entropy": 7.058232259750366,
"epoch": 0.09125272040013745,
"grad_norm": 0.921875,
"learning_rate": 0.0004953727922052035,
"loss": 7.2005,
"mean_token_accuracy": 0.13181116729974746,
"num_tokens": 2980203.0,
"step": 1195
},
{
"entropy": 7.229273128509521,
"epoch": 0.09163453094574472,
"grad_norm": 0.9921875,
"learning_rate": 0.0004951320775200756,
"loss": 7.1873,
"mean_token_accuracy": 0.12479368448257447,
"num_tokens": 2994621.0,
"step": 1200
},
{
"entropy": 7.185740375518799,
"epoch": 0.09201634149135199,
"grad_norm": 0.984375,
"learning_rate": 0.0004948853278028436,
"loss": 7.2108,
"mean_token_accuracy": 0.13122743889689445,
"num_tokens": 3007363.0,
"step": 1205
},
{
"entropy": 7.096544647216797,
"epoch": 0.09239815203695927,
"grad_norm": 1.6328125,
"learning_rate": 0.0004946325498182755,
"loss": 7.1552,
"mean_token_accuracy": 0.12754447683691977,
"num_tokens": 3020741.0,
"step": 1210
},
{
"entropy": 7.197258853912354,
"epoch": 0.09277996258256653,
"grad_norm": 0.9140625,
"learning_rate": 0.0004943737504964076,
"loss": 7.2048,
"mean_token_accuracy": 0.13166032060980798,
"num_tokens": 3033784.0,
"step": 1215
},
{
"entropy": 7.043015718460083,
"epoch": 0.0931617731281738,
"grad_norm": 0.8671875,
"learning_rate": 0.000494108936932354,
"loss": 7.184,
"mean_token_accuracy": 0.13357840329408646,
"num_tokens": 3047770.0,
"step": 1220
},
{
"entropy": 7.2477373600006105,
"epoch": 0.09354358367378107,
"grad_norm": 1.046875,
"learning_rate": 0.0004938381163861124,
"loss": 7.191,
"mean_token_accuracy": 0.13460689783096313,
"num_tokens": 3060402.0,
"step": 1225
},
{
"entropy": 7.158150339126587,
"epoch": 0.09392539421938834,
"grad_norm": 0.94921875,
"learning_rate": 0.0004935612962823645,
"loss": 7.1943,
"mean_token_accuracy": 0.12998826950788497,
"num_tokens": 3072710.0,
"step": 1230
},
{
"entropy": 7.093478393554688,
"epoch": 0.09430720476499561,
"grad_norm": 0.98828125,
"learning_rate": 0.0004932784842102739,
"loss": 7.1102,
"mean_token_accuracy": 0.14277689903974533,
"num_tokens": 3085751.0,
"step": 1235
},
{
"entropy": 7.256224966049194,
"epoch": 0.09468901531060288,
"grad_norm": 0.8828125,
"learning_rate": 0.0004929896879232758,
"loss": 7.1816,
"mean_token_accuracy": 0.13075839653611182,
"num_tokens": 3099066.0,
"step": 1240
},
{
"entropy": 7.079723644256592,
"epoch": 0.09507082585621014,
"grad_norm": 0.91796875,
"learning_rate": 0.0004926949153388668,
"loss": 7.1662,
"mean_token_accuracy": 0.1339078389108181,
"num_tokens": 3112145.0,
"step": 1245
},
{
"entropy": 7.142389106750488,
"epoch": 0.09545263640181742,
"grad_norm": 1.0234375,
"learning_rate": 0.0004923941745383859,
"loss": 7.1683,
"mean_token_accuracy": 0.131107147783041,
"num_tokens": 3124581.0,
"step": 1250
},
{
"entropy": 7.142371654510498,
"epoch": 0.09583444694742468,
"grad_norm": 0.96484375,
"learning_rate": 0.000492087473766794,
"loss": 7.1939,
"mean_token_accuracy": 0.13225287646055223,
"num_tokens": 3136368.0,
"step": 1255
},
{
"entropy": 7.149948930740356,
"epoch": 0.09621625749303196,
"grad_norm": 1.4375,
"learning_rate": 0.000491774821432448,
"loss": 7.1206,
"mean_token_accuracy": 0.13162810131907463,
"num_tokens": 3149528.0,
"step": 1260
},
{
"entropy": 7.1567319393157955,
"epoch": 0.09659806803863923,
"grad_norm": 0.87890625,
"learning_rate": 0.0004914562261068693,
"loss": 7.1336,
"mean_token_accuracy": 0.13130421340465545,
"num_tokens": 3162443.0,
"step": 1265
},
{
"entropy": 7.112925052642822,
"epoch": 0.09697987858424649,
"grad_norm": 1.0,
"learning_rate": 0.0004911316965245098,
"loss": 7.072,
"mean_token_accuracy": 0.14194170609116555,
"num_tokens": 3174561.0,
"step": 1270
},
{
"entropy": 6.972229814529419,
"epoch": 0.09736168912985377,
"grad_norm": 0.953125,
"learning_rate": 0.000490801241582512,
"loss": 7.0694,
"mean_token_accuracy": 0.13742200583219527,
"num_tokens": 3187781.0,
"step": 1275
},
{
"entropy": 7.20753846168518,
"epoch": 0.09774349967546103,
"grad_norm": 0.87109375,
"learning_rate": 0.000490464870340465,
"loss": 7.1693,
"mean_token_accuracy": 0.131757465749979,
"num_tokens": 3200898.0,
"step": 1280
},
{
"entropy": 7.110137605667115,
"epoch": 0.09812531022106831,
"grad_norm": 0.91796875,
"learning_rate": 0.0004901225920201563,
"loss": 7.1792,
"mean_token_accuracy": 0.12826146185398102,
"num_tokens": 3213919.0,
"step": 1285
},
{
"entropy": 7.155130624771118,
"epoch": 0.09850712076667557,
"grad_norm": 1.1171875,
"learning_rate": 0.000489774416005319,
"loss": 7.1013,
"mean_token_accuracy": 0.13860852047801017,
"num_tokens": 3226414.0,
"step": 1290
},
{
"entropy": 7.098326826095581,
"epoch": 0.09888893131228285,
"grad_norm": 1.125,
"learning_rate": 0.0004894203518413742,
"loss": 7.1132,
"mean_token_accuracy": 0.13262501135468482,
"num_tokens": 3238786.0,
"step": 1295
},
{
"entropy": 7.04604754447937,
"epoch": 0.09927074185789012,
"grad_norm": 1.0546875,
"learning_rate": 0.0004890604092351701,
"loss": 7.1213,
"mean_token_accuracy": 0.13990525901317596,
"num_tokens": 3251584.0,
"step": 1300
},
{
"entropy": 7.154106521606446,
"epoch": 0.09965255240349738,
"grad_norm": 0.953125,
"learning_rate": 0.000488694598054715,
"loss": 7.138,
"mean_token_accuracy": 0.13530943915247917,
"num_tokens": 3264805.0,
"step": 1305
},
{
"entropy": 7.134760904312134,
"epoch": 0.10003436294910466,
"grad_norm": 1.0234375,
"learning_rate": 0.0004883229283289071,
"loss": 7.1899,
"mean_token_accuracy": 0.12837898582220078,
"num_tokens": 3278681.0,
"step": 1310
},
{
"entropy": 7.063161373138428,
"epoch": 0.10041617349471192,
"grad_norm": 1.046875,
"learning_rate": 0.00048794541024725993,
"loss": 7.0764,
"mean_token_accuracy": 0.13971355333924293,
"num_tokens": 3290457.0,
"step": 1315
},
{
"entropy": 7.017997169494629,
"epoch": 0.1007979840403192,
"grad_norm": 1.1953125,
"learning_rate": 0.0004875620541596221,
"loss": 7.0086,
"mean_token_accuracy": 0.13909627720713616,
"num_tokens": 3303353.0,
"step": 1320
},
{
"entropy": 7.113060474395752,
"epoch": 0.10117979458592646,
"grad_norm": 0.9921875,
"learning_rate": 0.00048717287057589454,
"loss": 7.0367,
"mean_token_accuracy": 0.13749843761324881,
"num_tokens": 3316040.0,
"step": 1325
},
{
"entropy": 7.0832137107849125,
"epoch": 0.10156160513153373,
"grad_norm": 0.88671875,
"learning_rate": 0.0004867778701657417,
"loss": 7.1423,
"mean_token_accuracy": 0.13220136389136314,
"num_tokens": 3329803.0,
"step": 1330
},
{
"entropy": 7.106856870651245,
"epoch": 0.101943415677141,
"grad_norm": 0.84765625,
"learning_rate": 0.00048637706375829955,
"loss": 7.0984,
"mean_token_accuracy": 0.12772538587450982,
"num_tokens": 3343245.0,
"step": 1335
},
{
"entropy": 7.078064727783203,
"epoch": 0.10232522622274827,
"grad_norm": 0.9296875,
"learning_rate": 0.000485970462341878,
"loss": 7.0651,
"mean_token_accuracy": 0.13379976749420167,
"num_tokens": 3355569.0,
"step": 1340
},
{
"entropy": 7.107391738891602,
"epoch": 0.10270703676835555,
"grad_norm": 0.90234375,
"learning_rate": 0.00048555807706366044,
"loss": 7.1131,
"mean_token_accuracy": 0.13194383233785628,
"num_tokens": 3367359.0,
"step": 1345
},
{
"entropy": 7.016172742843628,
"epoch": 0.10308884731396281,
"grad_norm": 0.98046875,
"learning_rate": 0.00048513991922939756,
"loss": 7.0139,
"mean_token_accuracy": 0.1402280993759632,
"num_tokens": 3379586.0,
"step": 1350
},
{
"entropy": 7.109827327728271,
"epoch": 0.10347065785957008,
"grad_norm": 0.90625,
"learning_rate": 0.00048471600030309744,
"loss": 7.0253,
"mean_token_accuracy": 0.1351835697889328,
"num_tokens": 3391561.0,
"step": 1355
},
{
"entropy": 7.081918478012085,
"epoch": 0.10385246840517735,
"grad_norm": 0.94140625,
"learning_rate": 0.00048428633190671186,
"loss": 7.0985,
"mean_token_accuracy": 0.13951607942581176,
"num_tokens": 3403818.0,
"step": 1360
},
{
"entropy": 7.053944683074951,
"epoch": 0.10423427895078462,
"grad_norm": 0.87890625,
"learning_rate": 0.0004838509258198167,
"loss": 7.0883,
"mean_token_accuracy": 0.1354493446648121,
"num_tokens": 3416071.0,
"step": 1365
},
{
"entropy": 7.128853988647461,
"epoch": 0.1046160894963919,
"grad_norm": 0.89453125,
"learning_rate": 0.00048340979397929,
"loss": 7.1362,
"mean_token_accuracy": 0.13550538271665574,
"num_tokens": 3428667.0,
"step": 1370
},
{
"entropy": 6.962645530700684,
"epoch": 0.10499790004199916,
"grad_norm": 0.953125,
"learning_rate": 0.00048296294847898386,
"loss": 7.0338,
"mean_token_accuracy": 0.14013547897338868,
"num_tokens": 3440841.0,
"step": 1375
},
{
"entropy": 7.072249460220337,
"epoch": 0.10537971058760642,
"grad_norm": 0.921875,
"learning_rate": 0.0004825104015693934,
"loss": 6.9938,
"mean_token_accuracy": 0.1363446034491062,
"num_tokens": 3452335.0,
"step": 1380
},
{
"entropy": 7.003687334060669,
"epoch": 0.1057615211332137,
"grad_norm": 0.92578125,
"learning_rate": 0.0004820521656573208,
"loss": 6.9918,
"mean_token_accuracy": 0.1383456476032734,
"num_tokens": 3464239.0,
"step": 1385
},
{
"entropy": 6.971402740478515,
"epoch": 0.10614333167882097,
"grad_norm": 0.8359375,
"learning_rate": 0.00048158825330553505,
"loss": 7.035,
"mean_token_accuracy": 0.1379629008471966,
"num_tokens": 3478423.0,
"step": 1390
},
{
"entropy": 7.111477661132812,
"epoch": 0.10652514222442824,
"grad_norm": 1.0078125,
"learning_rate": 0.00048111867723242763,
"loss": 7.1073,
"mean_token_accuracy": 0.1306366391479969,
"num_tokens": 3491579.0,
"step": 1395
},
{
"entropy": 7.006403064727783,
"epoch": 0.10690695277003551,
"grad_norm": 0.984375,
"learning_rate": 0.0004806434503116637,
"loss": 7.0787,
"mean_token_accuracy": 0.13708655759692193,
"num_tokens": 3503313.0,
"step": 1400
},
{
"entropy": 7.118660974502563,
"epoch": 0.10728876331564277,
"grad_norm": 0.93359375,
"learning_rate": 0.0004801625855718296,
"loss": 7.0623,
"mean_token_accuracy": 0.13492181822657584,
"num_tokens": 3514377.0,
"step": 1405
},
{
"entropy": 7.0437156677246096,
"epoch": 0.10767057386125005,
"grad_norm": 0.9453125,
"learning_rate": 0.00047967609619607477,
"loss": 7.0477,
"mean_token_accuracy": 0.1366609439253807,
"num_tokens": 3527041.0,
"step": 1410
},
{
"entropy": 7.0386721134185795,
"epoch": 0.10805238440685731,
"grad_norm": 0.9453125,
"learning_rate": 0.0004791839955217513,
"loss": 7.0804,
"mean_token_accuracy": 0.1295987658202648,
"num_tokens": 3540300.0,
"step": 1415
},
{
"entropy": 7.046014022827149,
"epoch": 0.10843419495246459,
"grad_norm": 0.9375,
"learning_rate": 0.00047868629704004786,
"loss": 7.0416,
"mean_token_accuracy": 0.14144471287727356,
"num_tokens": 3553111.0,
"step": 1420
},
{
"entropy": 7.037079954147339,
"epoch": 0.10881600549807185,
"grad_norm": 0.8515625,
"learning_rate": 0.00047818301439561965,
"loss": 6.994,
"mean_token_accuracy": 0.14127341657876968,
"num_tokens": 3564892.0,
"step": 1425
},
{
"entropy": 7.0140142917633055,
"epoch": 0.10919781604367913,
"grad_norm": 0.9296875,
"learning_rate": 0.00047767416138621454,
"loss": 7.0735,
"mean_token_accuracy": 0.135311059653759,
"num_tokens": 3576244.0,
"step": 1430
},
{
"entropy": 7.0694409847259525,
"epoch": 0.1095796265892864,
"grad_norm": 0.890625,
"learning_rate": 0.000477159751962295,
"loss": 7.0074,
"mean_token_accuracy": 0.14243306890130042,
"num_tokens": 3588700.0,
"step": 1435
},
{
"entropy": 7.005630970001221,
"epoch": 0.10996143713489366,
"grad_norm": 0.98828125,
"learning_rate": 0.00047663980022665507,
"loss": 6.9865,
"mean_token_accuracy": 0.13463270515203477,
"num_tokens": 3601075.0,
"step": 1440
},
{
"entropy": 7.035998058319092,
"epoch": 0.11034324768050094,
"grad_norm": 0.88671875,
"learning_rate": 0.00047611432043403437,
"loss": 7.0024,
"mean_token_accuracy": 0.1404828667640686,
"num_tokens": 3613479.0,
"step": 1445
},
{
"entropy": 6.8482097625732425,
"epoch": 0.1107250582261082,
"grad_norm": 0.98046875,
"learning_rate": 0.0004755833269907267,
"loss": 6.9792,
"mean_token_accuracy": 0.1464216187596321,
"num_tokens": 3626329.0,
"step": 1450
},
{
"entropy": 7.078008937835693,
"epoch": 0.11110686877171548,
"grad_norm": 0.94921875,
"learning_rate": 0.0004750468344541857,
"loss": 7.0241,
"mean_token_accuracy": 0.14053741693496705,
"num_tokens": 3639067.0,
"step": 1455
},
{
"entropy": 6.932734251022339,
"epoch": 0.11148867931732274,
"grad_norm": 1.109375,
"learning_rate": 0.00047450485753262525,
"loss": 6.9073,
"mean_token_accuracy": 0.153934246301651,
"num_tokens": 3651231.0,
"step": 1460
},
{
"entropy": 6.968956708908081,
"epoch": 0.11187048986293001,
"grad_norm": 0.98046875,
"learning_rate": 0.00047395741108461633,
"loss": 6.9901,
"mean_token_accuracy": 0.1365182176232338,
"num_tokens": 3662075.0,
"step": 1465
},
{
"entropy": 6.9668042182922365,
"epoch": 0.11225230040853729,
"grad_norm": 1.0546875,
"learning_rate": 0.00047340451011867985,
"loss": 7.0298,
"mean_token_accuracy": 0.1391049027442932,
"num_tokens": 3674590.0,
"step": 1470
},
{
"entropy": 7.0620503425598145,
"epoch": 0.11263411095414455,
"grad_norm": 0.86328125,
"learning_rate": 0.00047284616979287515,
"loss": 6.9845,
"mean_token_accuracy": 0.14239439144730567,
"num_tokens": 3687366.0,
"step": 1475
},
{
"entropy": 6.983460140228272,
"epoch": 0.11301592149975183,
"grad_norm": 1.234375,
"learning_rate": 0.00047228240541438433,
"loss": 6.9413,
"mean_token_accuracy": 0.1429337441921234,
"num_tokens": 3699476.0,
"step": 1480
},
{
"entropy": 6.999633312225342,
"epoch": 0.11339773204535909,
"grad_norm": 1.984375,
"learning_rate": 0.00047171323243909257,
"loss": 7.057,
"mean_token_accuracy": 0.13819182813167571,
"num_tokens": 3712974.0,
"step": 1485
},
{
"entropy": 7.101782512664795,
"epoch": 0.11377954259096636,
"grad_norm": 0.92578125,
"learning_rate": 0.00047113866647116457,
"loss": 7.1066,
"mean_token_accuracy": 0.13542261794209481,
"num_tokens": 3725548.0,
"step": 1490
},
{
"entropy": 7.0012349605560305,
"epoch": 0.11416135313657363,
"grad_norm": 0.95703125,
"learning_rate": 0.0004705587232626164,
"loss": 7.0509,
"mean_token_accuracy": 0.12673078402876853,
"num_tokens": 3737158.0,
"step": 1495
},
{
"entropy": 7.032653379440307,
"epoch": 0.1145431636821809,
"grad_norm": 0.96875,
"learning_rate": 0.00046997341871288424,
"loss": 6.9872,
"mean_token_accuracy": 0.13459232226014137,
"num_tokens": 3749254.0,
"step": 1500
}
],
"logging_steps": 5,
"max_steps": 4000,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 5563152322560000.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue Copy Permalink