18dcce3b89
Model: fpadovani/swe-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
17222 lines
471 KiB
JSON
17222 lines
471 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 7.042253521126761,
|
|
"eval_steps": 500,
|
|
"global_step": 8500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.218191003799438,
|
|
"epoch": 0.004142502071251036,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.8898,
|
|
"mean_token_accuracy": 0.08805170804262161,
|
|
"num_tokens": 9149.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.191673707962036,
|
|
"epoch": 0.008285004142502071,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.7268,
|
|
"mean_token_accuracy": 0.09304728880524635,
|
|
"num_tokens": 16800.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.134524202346801,
|
|
"epoch": 0.012427506213753107,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.8905,
|
|
"mean_token_accuracy": 0.0921505056321621,
|
|
"num_tokens": 26617.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.115550088882446,
|
|
"epoch": 0.016570008285004142,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.8021,
|
|
"mean_token_accuracy": 0.09268744736909866,
|
|
"num_tokens": 35855.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.12396879196167,
|
|
"epoch": 0.020712510356255178,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.0094,
|
|
"mean_token_accuracy": 0.08752714470028877,
|
|
"num_tokens": 45644.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.054446792602539,
|
|
"epoch": 0.024855012427506214,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.8922,
|
|
"mean_token_accuracy": 0.09398627206683159,
|
|
"num_tokens": 54641.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.012255907058716,
|
|
"epoch": 0.02899751449875725,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.7519,
|
|
"mean_token_accuracy": 0.08958807587623596,
|
|
"num_tokens": 63924.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.068031549453735,
|
|
"epoch": 0.033140016570008285,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.89,
|
|
"mean_token_accuracy": 0.09437783882021904,
|
|
"num_tokens": 73452.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.06072301864624,
|
|
"epoch": 0.037282518641259324,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.9455,
|
|
"mean_token_accuracy": 0.0851291574537754,
|
|
"num_tokens": 82731.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.090083169937134,
|
|
"epoch": 0.041425020712510356,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.7734,
|
|
"mean_token_accuracy": 0.09048294499516488,
|
|
"num_tokens": 92626.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.149203157424926,
|
|
"epoch": 0.045567522783761395,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.9497,
|
|
"mean_token_accuracy": 0.08217904865741729,
|
|
"num_tokens": 101380.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.119951963424683,
|
|
"epoch": 0.04971002485501243,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.8604,
|
|
"mean_token_accuracy": 0.09202293381094932,
|
|
"num_tokens": 109895.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.2014998435974125,
|
|
"epoch": 0.053852526926263466,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.8449,
|
|
"mean_token_accuracy": 0.09257669150829315,
|
|
"num_tokens": 119203.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.1111664295196535,
|
|
"epoch": 0.0579950289975145,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.7903,
|
|
"mean_token_accuracy": 0.09657297283411026,
|
|
"num_tokens": 127572.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.04930944442749,
|
|
"epoch": 0.06213753106876554,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.8304,
|
|
"mean_token_accuracy": 0.09063600301742554,
|
|
"num_tokens": 137161.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.044711875915527,
|
|
"epoch": 0.06628003314001657,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.9451,
|
|
"mean_token_accuracy": 0.08874713778495788,
|
|
"num_tokens": 145991.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.130092525482178,
|
|
"epoch": 0.07042253521126761,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.8464,
|
|
"mean_token_accuracy": 0.08955071866512299,
|
|
"num_tokens": 154927.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.090402746200562,
|
|
"epoch": 0.07456503728251865,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.7595,
|
|
"mean_token_accuracy": 0.09832768961787224,
|
|
"num_tokens": 163234.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.08957576751709,
|
|
"epoch": 0.07870753935376967,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.7163,
|
|
"mean_token_accuracy": 0.09379880875349045,
|
|
"num_tokens": 171554.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.044517278671265,
|
|
"epoch": 0.08285004142502071,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.8173,
|
|
"mean_token_accuracy": 0.0955286517739296,
|
|
"num_tokens": 180171.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.050950908660889,
|
|
"epoch": 0.08699254349627175,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.761,
|
|
"mean_token_accuracy": 0.09741089940071106,
|
|
"num_tokens": 189216.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.173562717437744,
|
|
"epoch": 0.09113504556752279,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.9519,
|
|
"mean_token_accuracy": 0.08675800338387489,
|
|
"num_tokens": 200022.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.183744478225708,
|
|
"epoch": 0.09527754763877382,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.8642,
|
|
"mean_token_accuracy": 0.09210581332445145,
|
|
"num_tokens": 208686.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.121539115905762,
|
|
"epoch": 0.09942004971002485,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.8829,
|
|
"mean_token_accuracy": 0.09462191313505172,
|
|
"num_tokens": 218266.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.006736946105957,
|
|
"epoch": 0.1035625517812759,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.7523,
|
|
"mean_token_accuracy": 0.08771666437387467,
|
|
"num_tokens": 227658.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.040027761459351,
|
|
"epoch": 0.10770505385252693,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.7014,
|
|
"mean_token_accuracy": 0.1004502885043621,
|
|
"num_tokens": 236301.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.106470441818237,
|
|
"epoch": 0.11184755592377796,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.8424,
|
|
"mean_token_accuracy": 0.08812138065695763,
|
|
"num_tokens": 246355.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.174679756164551,
|
|
"epoch": 0.115990057995029,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.8174,
|
|
"mean_token_accuracy": 0.0924198254942894,
|
|
"num_tokens": 256023.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.10174069404602,
|
|
"epoch": 0.12013256006628004,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.9036,
|
|
"mean_token_accuracy": 0.08638993799686431,
|
|
"num_tokens": 266893.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.019921684265137,
|
|
"epoch": 0.12427506213753108,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.6606,
|
|
"mean_token_accuracy": 0.08872481286525727,
|
|
"num_tokens": 276075.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.042754888534546,
|
|
"epoch": 0.12841756420878211,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.7532,
|
|
"mean_token_accuracy": 0.09296932220458984,
|
|
"num_tokens": 285529.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.064399909973145,
|
|
"epoch": 0.13256006628003314,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.762,
|
|
"mean_token_accuracy": 0.09406789913773536,
|
|
"num_tokens": 294403.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.152362823486328,
|
|
"epoch": 0.13670256835128416,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.8553,
|
|
"mean_token_accuracy": 0.09358441680669785,
|
|
"num_tokens": 303507.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.071487474441528,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.7534,
|
|
"mean_token_accuracy": 0.09645206108689308,
|
|
"num_tokens": 312393.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 6.956210470199585,
|
|
"epoch": 0.14498757249378624,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.7226,
|
|
"mean_token_accuracy": 0.09953253641724587,
|
|
"num_tokens": 321309.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.2200812816619875,
|
|
"epoch": 0.1491300745650373,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.7879,
|
|
"mean_token_accuracy": 0.09582221284508705,
|
|
"num_tokens": 330217.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 6.964824962615967,
|
|
"epoch": 0.15327257663628832,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.8296,
|
|
"mean_token_accuracy": 0.09584516286849976,
|
|
"num_tokens": 339587.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.104565525054932,
|
|
"epoch": 0.15741507870753935,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.7603,
|
|
"mean_token_accuracy": 0.09563353508710862,
|
|
"num_tokens": 349367.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 6.90956711769104,
|
|
"epoch": 0.1615575807787904,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.6535,
|
|
"mean_token_accuracy": 0.09533214569091797,
|
|
"num_tokens": 358272.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.184430170059204,
|
|
"epoch": 0.16570008285004142,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.8847,
|
|
"mean_token_accuracy": 0.09342379868030548,
|
|
"num_tokens": 367709.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.108258152008057,
|
|
"epoch": 0.16984258492129245,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.7363,
|
|
"mean_token_accuracy": 0.09321728497743606,
|
|
"num_tokens": 376180.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.143853282928466,
|
|
"epoch": 0.1739850869925435,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.8927,
|
|
"mean_token_accuracy": 0.08639989122748375,
|
|
"num_tokens": 385636.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.005528688430786,
|
|
"epoch": 0.17812758906379453,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.7603,
|
|
"mean_token_accuracy": 0.10387283116579056,
|
|
"num_tokens": 394486.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.066776275634766,
|
|
"epoch": 0.18227009113504558,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.7286,
|
|
"mean_token_accuracy": 0.09739693626761436,
|
|
"num_tokens": 403765.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 6.9391695022583,
|
|
"epoch": 0.1864125932062966,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.7198,
|
|
"mean_token_accuracy": 0.09883175939321517,
|
|
"num_tokens": 412988.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.029045248031617,
|
|
"epoch": 0.19055509527754763,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.6219,
|
|
"mean_token_accuracy": 0.09850638806819915,
|
|
"num_tokens": 422096.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.055552911758423,
|
|
"epoch": 0.19469759734879868,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.7907,
|
|
"mean_token_accuracy": 0.08912368193268776,
|
|
"num_tokens": 431333.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.076433277130127,
|
|
"epoch": 0.1988400994200497,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.825,
|
|
"mean_token_accuracy": 0.0998022735118866,
|
|
"num_tokens": 440382.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.098494958877564,
|
|
"epoch": 0.20298260149130073,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.7184,
|
|
"mean_token_accuracy": 0.1035430908203125,
|
|
"num_tokens": 448469.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.065437746047974,
|
|
"epoch": 0.2071251035625518,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.7562,
|
|
"mean_token_accuracy": 0.0973759837448597,
|
|
"num_tokens": 457356.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.103154182434082,
|
|
"epoch": 0.2112676056338028,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.7821,
|
|
"mean_token_accuracy": 0.09721289053559304,
|
|
"num_tokens": 465928.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 6.99681601524353,
|
|
"epoch": 0.21541010770505387,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.6931,
|
|
"mean_token_accuracy": 0.09931288808584213,
|
|
"num_tokens": 475269.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 6.960800266265869,
|
|
"epoch": 0.2195526097763049,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.679,
|
|
"mean_token_accuracy": 0.09512786641716957,
|
|
"num_tokens": 485450.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.0514568328857425,
|
|
"epoch": 0.22369511184755592,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.7347,
|
|
"mean_token_accuracy": 0.10228226557374001,
|
|
"num_tokens": 494029.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 6.9516912460327145,
|
|
"epoch": 0.22783761391880697,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.676,
|
|
"mean_token_accuracy": 0.10202871784567832,
|
|
"num_tokens": 503253.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.018160057067871,
|
|
"epoch": 0.231980115990058,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.6685,
|
|
"mean_token_accuracy": 0.09529435187578202,
|
|
"num_tokens": 512857.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 6.9714985370635985,
|
|
"epoch": 0.23612261806130902,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.7719,
|
|
"mean_token_accuracy": 0.09288227260112762,
|
|
"num_tokens": 522319.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.113205528259277,
|
|
"epoch": 0.24026512013256007,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.7224,
|
|
"mean_token_accuracy": 0.09471780881285667,
|
|
"num_tokens": 531469.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 6.976934480667114,
|
|
"epoch": 0.2444076222038111,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.756,
|
|
"mean_token_accuracy": 0.09480062946677208,
|
|
"num_tokens": 540470.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.011937475204467,
|
|
"epoch": 0.24855012427506215,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.7584,
|
|
"mean_token_accuracy": 0.09700122326612473,
|
|
"num_tokens": 550317.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.0365160465240475,
|
|
"epoch": 0.2526926263463132,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.7067,
|
|
"mean_token_accuracy": 0.09434132501482964,
|
|
"num_tokens": 559409.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 6.928857326507568,
|
|
"epoch": 0.25683512841756423,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.768,
|
|
"mean_token_accuracy": 0.10764710158109665,
|
|
"num_tokens": 568726.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.103991413116455,
|
|
"epoch": 0.2609776304888152,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.7179,
|
|
"mean_token_accuracy": 0.0976541943848133,
|
|
"num_tokens": 577926.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.047203779220581,
|
|
"epoch": 0.2651201325600663,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.7448,
|
|
"mean_token_accuracy": 0.09809661954641342,
|
|
"num_tokens": 586695.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.039609956741333,
|
|
"epoch": 0.26926263463131733,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.771,
|
|
"mean_token_accuracy": 0.0949723981320858,
|
|
"num_tokens": 596250.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.063632202148438,
|
|
"epoch": 0.27340513670256833,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.8221,
|
|
"mean_token_accuracy": 0.09051206335425377,
|
|
"num_tokens": 605761.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.019326877593994,
|
|
"epoch": 0.2775476387738194,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.7365,
|
|
"mean_token_accuracy": 0.09872144013643265,
|
|
"num_tokens": 614313.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.037422132492066,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.7692,
|
|
"mean_token_accuracy": 0.09455474764108658,
|
|
"num_tokens": 624470.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.039082098007202,
|
|
"epoch": 0.28583264291632143,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.6843,
|
|
"mean_token_accuracy": 0.09269843101501465,
|
|
"num_tokens": 633458.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 6.942268037796021,
|
|
"epoch": 0.2899751449875725,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.6939,
|
|
"mean_token_accuracy": 0.09877684414386749,
|
|
"num_tokens": 643251.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 6.99584732055664,
|
|
"epoch": 0.29411764705882354,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7635,
|
|
"mean_token_accuracy": 0.09713540598750114,
|
|
"num_tokens": 652005.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.006586170196533,
|
|
"epoch": 0.2982601491300746,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.6575,
|
|
"mean_token_accuracy": 0.10120807588100433,
|
|
"num_tokens": 661003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.874996137619019,
|
|
"epoch": 0.3024026512013256,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.6294,
|
|
"mean_token_accuracy": 0.09844254851341247,
|
|
"num_tokens": 670221.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.155635738372803,
|
|
"epoch": 0.30654515327257664,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9236,
|
|
"mean_token_accuracy": 0.0882349319756031,
|
|
"num_tokens": 680756.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 6.94540753364563,
|
|
"epoch": 0.3106876553438277,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.6157,
|
|
"mean_token_accuracy": 0.10289054960012436,
|
|
"num_tokens": 689641.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.87992377281189,
|
|
"epoch": 0.3148301574150787,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.5696,
|
|
"mean_token_accuracy": 0.09839320331811904,
|
|
"num_tokens": 698209.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.823869276046753,
|
|
"epoch": 0.31897265948632975,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.5736,
|
|
"mean_token_accuracy": 0.10355777889490128,
|
|
"num_tokens": 706712.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.037634468078613,
|
|
"epoch": 0.3231151615575808,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8017,
|
|
"mean_token_accuracy": 0.09728012606501579,
|
|
"num_tokens": 716087.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.911249494552612,
|
|
"epoch": 0.3272576636288318,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.6465,
|
|
"mean_token_accuracy": 0.09347675815224647,
|
|
"num_tokens": 725913.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.944440317153931,
|
|
"epoch": 0.33140016570008285,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.7201,
|
|
"mean_token_accuracy": 0.10362306237220764,
|
|
"num_tokens": 735020.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.985636854171753,
|
|
"epoch": 0.3355426677713339,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6216,
|
|
"mean_token_accuracy": 0.09469136893749237,
|
|
"num_tokens": 744299.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.883630275726318,
|
|
"epoch": 0.3396851698425849,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.6099,
|
|
"mean_token_accuracy": 0.09694516137242318,
|
|
"num_tokens": 753231.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.825357389450073,
|
|
"epoch": 0.34382767191383595,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.6717,
|
|
"mean_token_accuracy": 0.10230854973196983,
|
|
"num_tokens": 763043.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.016851997375488,
|
|
"epoch": 0.347970173985087,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.6905,
|
|
"mean_token_accuracy": 0.09913048222661018,
|
|
"num_tokens": 771977.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.820244407653808,
|
|
"epoch": 0.352112676056338,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6243,
|
|
"mean_token_accuracy": 0.1011092983186245,
|
|
"num_tokens": 780285.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.956075525283813,
|
|
"epoch": 0.35625517812758906,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6593,
|
|
"mean_token_accuracy": 0.09649045094847679,
|
|
"num_tokens": 789980.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.984846019744873,
|
|
"epoch": 0.3603976801988401,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6275,
|
|
"mean_token_accuracy": 0.10474459528923034,
|
|
"num_tokens": 798700.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.767291498184204,
|
|
"epoch": 0.36454018227009116,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7208,
|
|
"mean_token_accuracy": 0.09600067660212516,
|
|
"num_tokens": 808758.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.046369791030884,
|
|
"epoch": 0.36868268434134216,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.6666,
|
|
"mean_token_accuracy": 0.10458802133798599,
|
|
"num_tokens": 817692.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.95425066947937,
|
|
"epoch": 0.3728251864125932,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7146,
|
|
"mean_token_accuracy": 0.10474570766091347,
|
|
"num_tokens": 827052.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.895345544815063,
|
|
"epoch": 0.37696768848384427,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7139,
|
|
"mean_token_accuracy": 0.1091882586479187,
|
|
"num_tokens": 836442.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.922871923446655,
|
|
"epoch": 0.38111019055509526,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.5775,
|
|
"mean_token_accuracy": 0.10118989571928978,
|
|
"num_tokens": 846441.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.8815388679504395,
|
|
"epoch": 0.3852526926263463,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6215,
|
|
"mean_token_accuracy": 0.10475475937128068,
|
|
"num_tokens": 856108.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.848587274551392,
|
|
"epoch": 0.38939519469759737,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5833,
|
|
"mean_token_accuracy": 0.10137399956583977,
|
|
"num_tokens": 865788.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.886804008483887,
|
|
"epoch": 0.39353769676884837,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6302,
|
|
"mean_token_accuracy": 0.09514767676591873,
|
|
"num_tokens": 874833.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.751422643661499,
|
|
"epoch": 0.3976801988400994,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.5997,
|
|
"mean_token_accuracy": 0.10276172533631325,
|
|
"num_tokens": 883685.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.881311464309692,
|
|
"epoch": 0.40182270091135047,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6001,
|
|
"mean_token_accuracy": 0.09796027019619942,
|
|
"num_tokens": 893908.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.934821891784668,
|
|
"epoch": 0.40596520298260147,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7142,
|
|
"mean_token_accuracy": 0.09827135354280472,
|
|
"num_tokens": 903081.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.7868632793426515,
|
|
"epoch": 0.4101077050538525,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5522,
|
|
"mean_token_accuracy": 0.10336750745773315,
|
|
"num_tokens": 912515.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.886235475540161,
|
|
"epoch": 0.4142502071251036,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.5797,
|
|
"mean_token_accuracy": 0.1070313923060894,
|
|
"num_tokens": 921056.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.4142502071251036,
|
|
"eval_entropy": 6.836038604728194,
|
|
"eval_loss": 6.684774875640869,
|
|
"eval_mean_token_accuracy": 0.10272799217021034,
|
|
"eval_num_tokens": 921056.0,
|
|
"eval_runtime": 2.7891,
|
|
"eval_samples_per_second": 1343.421,
|
|
"eval_steps_per_second": 168.152,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.7657403469085695,
|
|
"epoch": 0.4183927091963546,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.5317,
|
|
"mean_token_accuracy": 0.10417452603578567,
|
|
"num_tokens": 931112.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.821895503997803,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6605,
|
|
"mean_token_accuracy": 0.0978618748486042,
|
|
"num_tokens": 941120.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.886211729049682,
|
|
"epoch": 0.4266777133388567,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6044,
|
|
"mean_token_accuracy": 0.09737976416945457,
|
|
"num_tokens": 950041.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.879851579666138,
|
|
"epoch": 0.43082021541010773,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5907,
|
|
"mean_token_accuracy": 0.10606569275259972,
|
|
"num_tokens": 958608.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.886484861373901,
|
|
"epoch": 0.43496271748135873,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.7158,
|
|
"mean_token_accuracy": 0.09990910813212395,
|
|
"num_tokens": 966901.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.767313003540039,
|
|
"epoch": 0.4391052195526098,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.5504,
|
|
"mean_token_accuracy": 0.10768762230873108,
|
|
"num_tokens": 976198.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.892781782150268,
|
|
"epoch": 0.44324772162386084,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.624,
|
|
"mean_token_accuracy": 0.09861580505967141,
|
|
"num_tokens": 985248.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.772344255447388,
|
|
"epoch": 0.44739022369511183,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5014,
|
|
"mean_token_accuracy": 0.1071978710591793,
|
|
"num_tokens": 993289.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.820696449279785,
|
|
"epoch": 0.4515327257663629,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5909,
|
|
"mean_token_accuracy": 0.10533165782690049,
|
|
"num_tokens": 1001784.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.935409688949585,
|
|
"epoch": 0.45567522783761394,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5627,
|
|
"mean_token_accuracy": 0.10804210156202317,
|
|
"num_tokens": 1010445.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.745790100097656,
|
|
"epoch": 0.45981772990886494,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6184,
|
|
"mean_token_accuracy": 0.10709234476089477,
|
|
"num_tokens": 1019084.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.7490733623504635,
|
|
"epoch": 0.463960231980116,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5249,
|
|
"mean_token_accuracy": 0.10461713373661041,
|
|
"num_tokens": 1027263.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.876884937286377,
|
|
"epoch": 0.46810273405136704,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6233,
|
|
"mean_token_accuracy": 0.09752989709377288,
|
|
"num_tokens": 1036275.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.765736055374146,
|
|
"epoch": 0.47224523612261804,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7015,
|
|
"mean_token_accuracy": 0.09858408868312836,
|
|
"num_tokens": 1045208.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.9024957656860355,
|
|
"epoch": 0.4763877381938691,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6287,
|
|
"mean_token_accuracy": 0.1026283286511898,
|
|
"num_tokens": 1054703.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.647345924377442,
|
|
"epoch": 0.48053024026512015,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4841,
|
|
"mean_token_accuracy": 0.10212503597140313,
|
|
"num_tokens": 1063652.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.717900657653809,
|
|
"epoch": 0.48467274233637114,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5873,
|
|
"mean_token_accuracy": 0.10145951434969902,
|
|
"num_tokens": 1072070.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.997959756851197,
|
|
"epoch": 0.4888152444076222,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6143,
|
|
"mean_token_accuracy": 0.10064447894692422,
|
|
"num_tokens": 1079992.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.657160854339599,
|
|
"epoch": 0.49295774647887325,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.628,
|
|
"mean_token_accuracy": 0.10435748845338821,
|
|
"num_tokens": 1088862.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.773271942138672,
|
|
"epoch": 0.4971002485501243,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4317,
|
|
"mean_token_accuracy": 0.10795770138502121,
|
|
"num_tokens": 1097614.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.7102379322052,
|
|
"epoch": 0.5012427506213754,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6288,
|
|
"mean_token_accuracy": 0.10706980749964715,
|
|
"num_tokens": 1107347.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.8672473430633545,
|
|
"epoch": 0.5053852526926264,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6445,
|
|
"mean_token_accuracy": 0.10073774978518486,
|
|
"num_tokens": 1116707.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.764383840560913,
|
|
"epoch": 0.5095277547638773,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6549,
|
|
"mean_token_accuracy": 0.10234303548932075,
|
|
"num_tokens": 1126356.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.787193298339844,
|
|
"epoch": 0.5136702568351285,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.5847,
|
|
"mean_token_accuracy": 0.10079671740531922,
|
|
"num_tokens": 1136281.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.713733196258545,
|
|
"epoch": 0.5178127589063795,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5355,
|
|
"mean_token_accuracy": 0.10429632142186165,
|
|
"num_tokens": 1145466.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.7503923892974855,
|
|
"epoch": 0.5219552609776305,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.5409,
|
|
"mean_token_accuracy": 0.10918259546160698,
|
|
"num_tokens": 1154225.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.761834907531738,
|
|
"epoch": 0.5260977630488816,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5885,
|
|
"mean_token_accuracy": 0.09673135951161385,
|
|
"num_tokens": 1163872.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.69431324005127,
|
|
"epoch": 0.5302402651201326,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5244,
|
|
"mean_token_accuracy": 0.10705070868134499,
|
|
"num_tokens": 1172695.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.651590061187744,
|
|
"epoch": 0.5343827671913836,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.3962,
|
|
"mean_token_accuracy": 0.11006554961204529,
|
|
"num_tokens": 1181291.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.782713270187378,
|
|
"epoch": 0.5385252692626347,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5715,
|
|
"mean_token_accuracy": 0.10394523739814758,
|
|
"num_tokens": 1189927.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.705166435241699,
|
|
"epoch": 0.5426677713338857,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5043,
|
|
"mean_token_accuracy": 0.10940467119216919,
|
|
"num_tokens": 1197913.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.670699262619019,
|
|
"epoch": 0.5468102734051367,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4309,
|
|
"mean_token_accuracy": 0.11147961542010307,
|
|
"num_tokens": 1207032.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.640204334259034,
|
|
"epoch": 0.5509527754763878,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5397,
|
|
"mean_token_accuracy": 0.10571447387337685,
|
|
"num_tokens": 1215710.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.694819068908691,
|
|
"epoch": 0.5550952775476388,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.52,
|
|
"mean_token_accuracy": 0.10049431845545768,
|
|
"num_tokens": 1225619.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.649594163894653,
|
|
"epoch": 0.5592377796188898,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4407,
|
|
"mean_token_accuracy": 0.11157044768333435,
|
|
"num_tokens": 1235000.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.612611961364746,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4462,
|
|
"mean_token_accuracy": 0.108580531924963,
|
|
"num_tokens": 1243594.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.843221950531006,
|
|
"epoch": 0.5675227837613919,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6445,
|
|
"mean_token_accuracy": 0.10460855960845947,
|
|
"num_tokens": 1252279.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.760235929489136,
|
|
"epoch": 0.5716652858326429,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.559,
|
|
"mean_token_accuracy": 0.10665302500128745,
|
|
"num_tokens": 1261314.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.55326566696167,
|
|
"epoch": 0.575807787903894,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4337,
|
|
"mean_token_accuracy": 0.11188110187649727,
|
|
"num_tokens": 1270854.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.618375587463379,
|
|
"epoch": 0.579950289975145,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3929,
|
|
"mean_token_accuracy": 0.11392810270190239,
|
|
"num_tokens": 1279183.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.626208257675171,
|
|
"epoch": 0.584092792046396,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5015,
|
|
"mean_token_accuracy": 0.10761247500777245,
|
|
"num_tokens": 1288347.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.6855669021606445,
|
|
"epoch": 0.5882352941176471,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5036,
|
|
"mean_token_accuracy": 0.10823310390114785,
|
|
"num_tokens": 1297449.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.710147094726563,
|
|
"epoch": 0.5923777961888981,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4688,
|
|
"mean_token_accuracy": 0.10904998034238815,
|
|
"num_tokens": 1306232.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.6553794860839846,
|
|
"epoch": 0.5965202982601492,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4857,
|
|
"mean_token_accuracy": 0.11109372302889824,
|
|
"num_tokens": 1315073.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.674065732955933,
|
|
"epoch": 0.6006628003314002,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5301,
|
|
"mean_token_accuracy": 0.10648825243115426,
|
|
"num_tokens": 1324629.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.700368881225586,
|
|
"epoch": 0.6048053024026512,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6444,
|
|
"mean_token_accuracy": 0.10443340390920638,
|
|
"num_tokens": 1334152.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.779498624801636,
|
|
"epoch": 0.6089478044739023,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5946,
|
|
"mean_token_accuracy": 0.10130336955189705,
|
|
"num_tokens": 1343063.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.684298419952393,
|
|
"epoch": 0.6130903065451533,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5241,
|
|
"mean_token_accuracy": 0.10516620799899101,
|
|
"num_tokens": 1352632.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.582866573333741,
|
|
"epoch": 0.6172328086164043,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4712,
|
|
"mean_token_accuracy": 0.10765027329325676,
|
|
"num_tokens": 1361103.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.643687963485718,
|
|
"epoch": 0.6213753106876554,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4642,
|
|
"mean_token_accuracy": 0.10832677409052849,
|
|
"num_tokens": 1370658.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.64506516456604,
|
|
"epoch": 0.6255178127589064,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4878,
|
|
"mean_token_accuracy": 0.10808288529515267,
|
|
"num_tokens": 1379817.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.640752458572388,
|
|
"epoch": 0.6296603148301574,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4418,
|
|
"mean_token_accuracy": 0.10874675288796425,
|
|
"num_tokens": 1389329.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.4627049446105955,
|
|
"epoch": 0.6338028169014085,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4902,
|
|
"mean_token_accuracy": 0.10606684386730195,
|
|
"num_tokens": 1399279.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.621781063079834,
|
|
"epoch": 0.6379453189726595,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4146,
|
|
"mean_token_accuracy": 0.11267002299427986,
|
|
"num_tokens": 1408765.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.736465692520142,
|
|
"epoch": 0.6420878210439105,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5632,
|
|
"mean_token_accuracy": 0.10607504546642303,
|
|
"num_tokens": 1418615.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.616378879547119,
|
|
"epoch": 0.6462303231151616,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4664,
|
|
"mean_token_accuracy": 0.10207068845629692,
|
|
"num_tokens": 1427359.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.70065336227417,
|
|
"epoch": 0.6503728251864126,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.622,
|
|
"mean_token_accuracy": 0.1116428442299366,
|
|
"num_tokens": 1435516.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.709064102172851,
|
|
"epoch": 0.6545153272576636,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5704,
|
|
"mean_token_accuracy": 0.10583743155002594,
|
|
"num_tokens": 1444438.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.671351766586303,
|
|
"epoch": 0.6586578293289147,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4996,
|
|
"mean_token_accuracy": 0.11326526403427124,
|
|
"num_tokens": 1453740.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.614573907852173,
|
|
"epoch": 0.6628003314001657,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4874,
|
|
"mean_token_accuracy": 0.11111863479018211,
|
|
"num_tokens": 1462083.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.484545516967773,
|
|
"epoch": 0.6669428334714167,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4127,
|
|
"mean_token_accuracy": 0.1100592203438282,
|
|
"num_tokens": 1471459.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.621589279174804,
|
|
"epoch": 0.6710853355426678,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3972,
|
|
"mean_token_accuracy": 0.11681128293275833,
|
|
"num_tokens": 1480516.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.421737241744995,
|
|
"epoch": 0.6752278376139188,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3,
|
|
"mean_token_accuracy": 0.11318568140268326,
|
|
"num_tokens": 1489268.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.700246906280517,
|
|
"epoch": 0.6793703396851698,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5174,
|
|
"mean_token_accuracy": 0.11110908538103104,
|
|
"num_tokens": 1498819.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.622490644454956,
|
|
"epoch": 0.6835128417564209,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.501,
|
|
"mean_token_accuracy": 0.11033953949809075,
|
|
"num_tokens": 1507852.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.59730544090271,
|
|
"epoch": 0.6876553438276719,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5302,
|
|
"mean_token_accuracy": 0.11122860461473465,
|
|
"num_tokens": 1517538.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.458763217926025,
|
|
"epoch": 0.6917978458989229,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4061,
|
|
"mean_token_accuracy": 0.11324542313814163,
|
|
"num_tokens": 1525389.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.57511887550354,
|
|
"epoch": 0.695940347970174,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3741,
|
|
"mean_token_accuracy": 0.11116622686386109,
|
|
"num_tokens": 1534117.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.522304344177246,
|
|
"epoch": 0.700082850041425,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4029,
|
|
"mean_token_accuracy": 0.11315450444817543,
|
|
"num_tokens": 1542610.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.5175107479095455,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4206,
|
|
"mean_token_accuracy": 0.1143906556069851,
|
|
"num_tokens": 1551771.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.620167636871338,
|
|
"epoch": 0.7083678541839271,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5779,
|
|
"mean_token_accuracy": 0.10512799099087715,
|
|
"num_tokens": 1560231.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.589151382446289,
|
|
"epoch": 0.7125103562551781,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5346,
|
|
"mean_token_accuracy": 0.10532297939062119,
|
|
"num_tokens": 1569578.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.563750839233398,
|
|
"epoch": 0.7166528583264291,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4187,
|
|
"mean_token_accuracy": 0.11381370946764946,
|
|
"num_tokens": 1579308.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.597287034988403,
|
|
"epoch": 0.7207953603976802,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.6081,
|
|
"mean_token_accuracy": 0.10267190784215927,
|
|
"num_tokens": 1589506.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.482245397567749,
|
|
"epoch": 0.7249378624689312,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3671,
|
|
"mean_token_accuracy": 0.11368186473846435,
|
|
"num_tokens": 1598127.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.548954200744629,
|
|
"epoch": 0.7290803645401823,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4125,
|
|
"mean_token_accuracy": 0.11406067088246345,
|
|
"num_tokens": 1608073.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.551317119598389,
|
|
"epoch": 0.7332228666114333,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4304,
|
|
"mean_token_accuracy": 0.11178040653467178,
|
|
"num_tokens": 1616272.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.496135663986206,
|
|
"epoch": 0.7373653686826843,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.447,
|
|
"mean_token_accuracy": 0.10845504105091094,
|
|
"num_tokens": 1625577.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.685812950134277,
|
|
"epoch": 0.7415078707539354,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.533,
|
|
"mean_token_accuracy": 0.10158233270049095,
|
|
"num_tokens": 1634902.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.520436191558838,
|
|
"epoch": 0.7456503728251864,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4674,
|
|
"mean_token_accuracy": 0.10865833014249801,
|
|
"num_tokens": 1644131.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.461379718780518,
|
|
"epoch": 0.7497928748964374,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.3694,
|
|
"mean_token_accuracy": 0.12158194407820702,
|
|
"num_tokens": 1652272.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.533070278167725,
|
|
"epoch": 0.7539353769676885,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.4632,
|
|
"mean_token_accuracy": 0.11447574123740197,
|
|
"num_tokens": 1660962.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.475569725036621,
|
|
"epoch": 0.7580778790389395,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4006,
|
|
"mean_token_accuracy": 0.11205699220299721,
|
|
"num_tokens": 1670570.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.526782608032226,
|
|
"epoch": 0.7622203811101905,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4497,
|
|
"mean_token_accuracy": 0.1096197433769703,
|
|
"num_tokens": 1679674.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.530770015716553,
|
|
"epoch": 0.7663628831814416,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4324,
|
|
"mean_token_accuracy": 0.11466658413410187,
|
|
"num_tokens": 1688849.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.4163538932800295,
|
|
"epoch": 0.7705053852526926,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3128,
|
|
"mean_token_accuracy": 0.1098814107477665,
|
|
"num_tokens": 1698565.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.5036052703857425,
|
|
"epoch": 0.7746478873239436,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3986,
|
|
"mean_token_accuracy": 0.11683695837855339,
|
|
"num_tokens": 1708009.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.518198871612549,
|
|
"epoch": 0.7787903893951947,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3878,
|
|
"mean_token_accuracy": 0.1126109592616558,
|
|
"num_tokens": 1716497.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.415347719192505,
|
|
"epoch": 0.7829328914664457,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4082,
|
|
"mean_token_accuracy": 0.12070046439766884,
|
|
"num_tokens": 1725609.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.477306175231933,
|
|
"epoch": 0.7870753935376967,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3456,
|
|
"mean_token_accuracy": 0.11454906165599824,
|
|
"num_tokens": 1734165.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.454520893096924,
|
|
"epoch": 0.7912178956089478,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4546,
|
|
"mean_token_accuracy": 0.10918105840682983,
|
|
"num_tokens": 1744856.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.4711943626403805,
|
|
"epoch": 0.7953603976801988,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4509,
|
|
"mean_token_accuracy": 0.10578904524445534,
|
|
"num_tokens": 1753386.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.4536961078643795,
|
|
"epoch": 0.7995028997514498,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3053,
|
|
"mean_token_accuracy": 0.11872726455330848,
|
|
"num_tokens": 1762521.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.34936490058899,
|
|
"epoch": 0.8036454018227009,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3765,
|
|
"mean_token_accuracy": 0.11444728896021843,
|
|
"num_tokens": 1772403.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.375087881088257,
|
|
"epoch": 0.8077879038939519,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.2884,
|
|
"mean_token_accuracy": 0.11179185584187508,
|
|
"num_tokens": 1780924.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.547906732559204,
|
|
"epoch": 0.8119304059652029,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4367,
|
|
"mean_token_accuracy": 0.11225800514221192,
|
|
"num_tokens": 1789992.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.421040678024292,
|
|
"epoch": 0.816072908036454,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3919,
|
|
"mean_token_accuracy": 0.11761699616909027,
|
|
"num_tokens": 1799513.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.332476711273193,
|
|
"epoch": 0.820215410107705,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.2287,
|
|
"mean_token_accuracy": 0.11279567033052444,
|
|
"num_tokens": 1807393.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.438618326187134,
|
|
"epoch": 0.824357912178956,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.5079,
|
|
"mean_token_accuracy": 0.10579989776015282,
|
|
"num_tokens": 1817284.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.449818515777588,
|
|
"epoch": 0.8285004142502072,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2996,
|
|
"mean_token_accuracy": 0.11799877136945724,
|
|
"num_tokens": 1825591.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8285004142502072,
|
|
"eval_entropy": 6.222835092402216,
|
|
"eval_loss": 6.475013256072998,
|
|
"eval_mean_token_accuracy": 0.11269166863866961,
|
|
"eval_num_tokens": 1825591.0,
|
|
"eval_runtime": 2.7611,
|
|
"eval_samples_per_second": 1357.068,
|
|
"eval_steps_per_second": 169.86,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.457831478118896,
|
|
"epoch": 0.8326429163214581,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004999998547681232,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.11798396334052086,
|
|
"num_tokens": 1834655.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.26599349975586,
|
|
"epoch": 0.8367854183927091,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999992647639447,
|
|
"loss": 6.2578,
|
|
"mean_token_accuracy": 0.11858200430870056,
|
|
"num_tokens": 1844187.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.415271520614624,
|
|
"epoch": 0.8409279204639603,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999982209116618,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.11406200677156449,
|
|
"num_tokens": 1852768.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.355557584762574,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999967232133796,
|
|
"loss": 6.3913,
|
|
"mean_token_accuracy": 0.12348653748631477,
|
|
"num_tokens": 1862095.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.427043008804321,
|
|
"epoch": 0.8492129246064622,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999947716721196,
|
|
"loss": 6.4206,
|
|
"mean_token_accuracy": 0.10961534604430198,
|
|
"num_tokens": 1871239.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.509608554840088,
|
|
"epoch": 0.8533554266777134,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999923662918179,
|
|
"loss": 6.4323,
|
|
"mean_token_accuracy": 0.11334251016378402,
|
|
"num_tokens": 1880065.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.417975521087646,
|
|
"epoch": 0.8574979287489644,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999895070773266,
|
|
"loss": 6.4608,
|
|
"mean_token_accuracy": 0.10985229611396789,
|
|
"num_tokens": 1889065.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.429206514358521,
|
|
"epoch": 0.8616404308202155,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999861940344129,
|
|
"loss": 6.4721,
|
|
"mean_token_accuracy": 0.11190373301506043,
|
|
"num_tokens": 1898238.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.462369585037232,
|
|
"epoch": 0.8657829328914665,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004999824271697597,
|
|
"loss": 6.3929,
|
|
"mean_token_accuracy": 0.12136606723070145,
|
|
"num_tokens": 1906780.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.506088638305664,
|
|
"epoch": 0.8699254349627175,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499978206490965,
|
|
"loss": 6.4745,
|
|
"mean_token_accuracy": 0.11197563111782075,
|
|
"num_tokens": 1916183.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.427097368240356,
|
|
"epoch": 0.8740679370339686,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999735320065427,
|
|
"loss": 6.3371,
|
|
"mean_token_accuracy": 0.11927392706274986,
|
|
"num_tokens": 1925382.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.292429256439209,
|
|
"epoch": 0.8782104391052196,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999684037259216,
|
|
"loss": 6.1827,
|
|
"mean_token_accuracy": 0.1270924136042595,
|
|
"num_tokens": 1933639.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.3600146770477295,
|
|
"epoch": 0.8823529411764706,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999628216594456,
|
|
"loss": 6.464,
|
|
"mean_token_accuracy": 0.11409285366535186,
|
|
"num_tokens": 1943044.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.460871028900146,
|
|
"epoch": 0.8864954432477217,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000499956785818375,
|
|
"loss": 6.303,
|
|
"mean_token_accuracy": 0.11871922463178634,
|
|
"num_tokens": 1951777.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.376981210708618,
|
|
"epoch": 0.8906379453189727,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999502962148843,
|
|
"loss": 6.3666,
|
|
"mean_token_accuracy": 0.1170460693538189,
|
|
"num_tokens": 1961356.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.439280414581299,
|
|
"epoch": 0.8947804473902237,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499943352862064,
|
|
"loss": 6.4248,
|
|
"mean_token_accuracy": 0.11459407210350037,
|
|
"num_tokens": 1970341.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.394289827346801,
|
|
"epoch": 0.8989229494614748,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999359557739194,
|
|
"loss": 6.2796,
|
|
"mean_token_accuracy": 0.12198276743292809,
|
|
"num_tokens": 1978576.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.319045782089233,
|
|
"epoch": 0.9030654515327258,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999281049653713,
|
|
"loss": 6.3227,
|
|
"mean_token_accuracy": 0.11601013913750649,
|
|
"num_tokens": 1987229.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.376017141342163,
|
|
"epoch": 0.9072079536039768,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999198004522557,
|
|
"loss": 6.261,
|
|
"mean_token_accuracy": 0.12269015237689018,
|
|
"num_tokens": 1996473.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.2405853271484375,
|
|
"epoch": 0.9113504556752279,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999110422513237,
|
|
"loss": 6.286,
|
|
"mean_token_accuracy": 0.12056776732206345,
|
|
"num_tokens": 2005551.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.398191165924072,
|
|
"epoch": 0.9154929577464789,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999018303802414,
|
|
"loss": 6.3869,
|
|
"mean_token_accuracy": 0.11947718560695648,
|
|
"num_tokens": 2015012.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.374700450897217,
|
|
"epoch": 0.9196354598177299,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998921648575904,
|
|
"loss": 6.3999,
|
|
"mean_token_accuracy": 0.1203969657421112,
|
|
"num_tokens": 2024435.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.380674362182617,
|
|
"epoch": 0.923777961888981,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998820457028669,
|
|
"loss": 6.3005,
|
|
"mean_token_accuracy": 0.12089794874191284,
|
|
"num_tokens": 2033991.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.409522771835327,
|
|
"epoch": 0.927920463960232,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998714729364824,
|
|
"loss": 6.3017,
|
|
"mean_token_accuracy": 0.1228084571659565,
|
|
"num_tokens": 2043117.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.326260423660278,
|
|
"epoch": 0.932062966031483,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004998604465797634,
|
|
"loss": 6.3305,
|
|
"mean_token_accuracy": 0.11992078274488449,
|
|
"num_tokens": 2052304.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.251798295974732,
|
|
"epoch": 0.9362054681027341,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998489666549512,
|
|
"loss": 6.2779,
|
|
"mean_token_accuracy": 0.12114684730768203,
|
|
"num_tokens": 2061217.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.319644594192505,
|
|
"epoch": 0.9403479701739851,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998370331852021,
|
|
"loss": 6.2517,
|
|
"mean_token_accuracy": 0.11879112273454666,
|
|
"num_tokens": 2070796.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.465651321411133,
|
|
"epoch": 0.9444904722452361,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998246461945872,
|
|
"loss": 6.4037,
|
|
"mean_token_accuracy": 0.12191760018467904,
|
|
"num_tokens": 2080202.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.320812797546386,
|
|
"epoch": 0.9486329743164872,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998118057080923,
|
|
"loss": 6.3045,
|
|
"mean_token_accuracy": 0.11891168132424354,
|
|
"num_tokens": 2089788.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.301202344894409,
|
|
"epoch": 0.9527754763877382,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997985117516183,
|
|
"loss": 6.3462,
|
|
"mean_token_accuracy": 0.11142431199550629,
|
|
"num_tokens": 2098727.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.344118309020996,
|
|
"epoch": 0.9569179784589892,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997847643519803,
|
|
"loss": 6.226,
|
|
"mean_token_accuracy": 0.11924557387828827,
|
|
"num_tokens": 2107645.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.1790977954864506,
|
|
"epoch": 0.9610604805302403,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997705635369085,
|
|
"loss": 6.1991,
|
|
"mean_token_accuracy": 0.11988648250699044,
|
|
"num_tokens": 2116613.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.284418439865112,
|
|
"epoch": 0.9652029826014913,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004997559093350474,
|
|
"loss": 6.1881,
|
|
"mean_token_accuracy": 0.12116428092122078,
|
|
"num_tokens": 2125585.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.396596622467041,
|
|
"epoch": 0.9693454846727423,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997408017759562,
|
|
"loss": 6.3359,
|
|
"mean_token_accuracy": 0.11936862170696258,
|
|
"num_tokens": 2134795.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.273751878738404,
|
|
"epoch": 0.9734879867439934,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997252408901084,
|
|
"loss": 6.3271,
|
|
"mean_token_accuracy": 0.11899949684739113,
|
|
"num_tokens": 2143699.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.386678028106689,
|
|
"epoch": 0.9776304888152444,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997092267088922,
|
|
"loss": 6.2707,
|
|
"mean_token_accuracy": 0.11861606612801552,
|
|
"num_tokens": 2152187.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.275117111206055,
|
|
"epoch": 0.9817729908864954,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996927592646097,
|
|
"loss": 6.3412,
|
|
"mean_token_accuracy": 0.11767763197422028,
|
|
"num_tokens": 2160941.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.339710283279419,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004996758385904776,
|
|
"loss": 6.3119,
|
|
"mean_token_accuracy": 0.11789651140570641,
|
|
"num_tokens": 2168946.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.181374216079712,
|
|
"epoch": 0.9900579950289975,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004996584647206269,
|
|
"loss": 6.2237,
|
|
"mean_token_accuracy": 0.12055367529392243,
|
|
"num_tokens": 2177662.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.316773986816406,
|
|
"epoch": 0.9942004971002486,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996406376901026,
|
|
"loss": 6.3268,
|
|
"mean_token_accuracy": 0.12757316455245019,
|
|
"num_tokens": 2187461.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.283385133743286,
|
|
"epoch": 0.9983429991714996,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996223575348635,
|
|
"loss": 6.2207,
|
|
"mean_token_accuracy": 0.12578138560056687,
|
|
"num_tokens": 2196878.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.311019659042358,
|
|
"epoch": 1.0024855012427507,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996036242917828,
|
|
"loss": 6.2021,
|
|
"mean_token_accuracy": 0.1151522897183895,
|
|
"num_tokens": 2206178.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.25421667098999,
|
|
"epoch": 1.0066280033140016,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995844379986476,
|
|
"loss": 6.0526,
|
|
"mean_token_accuracy": 0.12476176023483276,
|
|
"num_tokens": 2215330.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.309921216964722,
|
|
"epoch": 1.0107705053852527,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995647986941587,
|
|
"loss": 6.0354,
|
|
"mean_token_accuracy": 0.13005029112100602,
|
|
"num_tokens": 2224146.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.241363286972046,
|
|
"epoch": 1.0149130074565038,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995447064179308,
|
|
"loss": 6.0171,
|
|
"mean_token_accuracy": 0.13421598002314566,
|
|
"num_tokens": 2233646.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.058859920501709,
|
|
"epoch": 1.0190555095277547,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499524161210492,
|
|
"loss": 5.9379,
|
|
"mean_token_accuracy": 0.12560308948159218,
|
|
"num_tokens": 2243663.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.270004224777222,
|
|
"epoch": 1.0231980115990058,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995031631132844,
|
|
"loss": 6.1035,
|
|
"mean_token_accuracy": 0.13170127123594283,
|
|
"num_tokens": 2252564.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.2710247993469235,
|
|
"epoch": 1.027340513670257,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004994817121686633,
|
|
"loss": 6.099,
|
|
"mean_token_accuracy": 0.12465815395116805,
|
|
"num_tokens": 2262504.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.193396997451782,
|
|
"epoch": 1.0314830157415078,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004994598084198979,
|
|
"loss": 5.9843,
|
|
"mean_token_accuracy": 0.1374910719692707,
|
|
"num_tokens": 2270553.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.136552810668945,
|
|
"epoch": 1.035625517812759,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004994374519111701,
|
|
"loss": 5.977,
|
|
"mean_token_accuracy": 0.1272726461291313,
|
|
"num_tokens": 2279319.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.217378282546997,
|
|
"epoch": 1.03976801988401,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004994146426875757,
|
|
"loss": 6.123,
|
|
"mean_token_accuracy": 0.126935476064682,
|
|
"num_tokens": 2288749.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.218695068359375,
|
|
"epoch": 1.043910521955261,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993913807951232,
|
|
"loss": 6.0156,
|
|
"mean_token_accuracy": 0.12562426924705505,
|
|
"num_tokens": 2298272.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.221994352340698,
|
|
"epoch": 1.048053024026512,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004993676662807345,
|
|
"loss": 6.047,
|
|
"mean_token_accuracy": 0.12884460240602494,
|
|
"num_tokens": 2307230.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.183428144454956,
|
|
"epoch": 1.0521955260977631,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004993434991922443,
|
|
"loss": 6.1317,
|
|
"mean_token_accuracy": 0.12460145726799965,
|
|
"num_tokens": 2316306.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.22090425491333,
|
|
"epoch": 1.056338028169014,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004993188795784002,
|
|
"loss": 6.0531,
|
|
"mean_token_accuracy": 0.1241947129368782,
|
|
"num_tokens": 2325672.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.1453857421875,
|
|
"epoch": 1.0604805302402651,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004992938074888628,
|
|
"loss": 6.066,
|
|
"mean_token_accuracy": 0.1274007849395275,
|
|
"num_tokens": 2335251.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.180039167404175,
|
|
"epoch": 1.0646230323115162,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004992682829742052,
|
|
"loss": 6.0216,
|
|
"mean_token_accuracy": 0.1308726042509079,
|
|
"num_tokens": 2344054.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.086296463012696,
|
|
"epoch": 1.0687655343827671,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004992423060859132,
|
|
"loss": 6.0236,
|
|
"mean_token_accuracy": 0.127706591039896,
|
|
"num_tokens": 2353493.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.223158740997315,
|
|
"epoch": 1.0729080364540182,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004992158768763849,
|
|
"loss": 6.1321,
|
|
"mean_token_accuracy": 0.12850543335080147,
|
|
"num_tokens": 2362687.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.2506636619567875,
|
|
"epoch": 1.0770505385252693,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991889953989308,
|
|
"loss": 6.0965,
|
|
"mean_token_accuracy": 0.12563316524028778,
|
|
"num_tokens": 2372325.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.173602199554443,
|
|
"epoch": 1.0811930405965202,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991616617077741,
|
|
"loss": 6.12,
|
|
"mean_token_accuracy": 0.1287991665303707,
|
|
"num_tokens": 2381987.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.307486963272095,
|
|
"epoch": 1.0853355426677713,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004991338758580496,
|
|
"loss": 6.0234,
|
|
"mean_token_accuracy": 0.13178783878684044,
|
|
"num_tokens": 2390901.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.034789133071899,
|
|
"epoch": 1.0894780447390224,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004991056379058045,
|
|
"loss": 6.0813,
|
|
"mean_token_accuracy": 0.12450421676039695,
|
|
"num_tokens": 2400756.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.188017988204956,
|
|
"epoch": 1.0936205468102733,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004990769479079976,
|
|
"loss": 5.9232,
|
|
"mean_token_accuracy": 0.12984196096658707,
|
|
"num_tokens": 2409526.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.086554956436157,
|
|
"epoch": 1.0977630488815244,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004990478059225003,
|
|
"loss": 6.0931,
|
|
"mean_token_accuracy": 0.12972966134548186,
|
|
"num_tokens": 2419583.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.2457873821258545,
|
|
"epoch": 1.1019055509527755,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004990182120080945,
|
|
"loss": 6.0338,
|
|
"mean_token_accuracy": 0.12990806251764297,
|
|
"num_tokens": 2429439.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.078778219223023,
|
|
"epoch": 1.1060480530240264,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004989881662244748,
|
|
"loss": 6.0416,
|
|
"mean_token_accuracy": 0.12380676046013832,
|
|
"num_tokens": 2438467.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.138932418823242,
|
|
"epoch": 1.1101905550952775,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989576686322468,
|
|
"loss": 5.9919,
|
|
"mean_token_accuracy": 0.1332716926932335,
|
|
"num_tokens": 2446619.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.1281578540802,
|
|
"epoch": 1.1143330571665286,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004989267192929272,
|
|
"loss": 6.0379,
|
|
"mean_token_accuracy": 0.12955363765358924,
|
|
"num_tokens": 2454998.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.09554443359375,
|
|
"epoch": 1.1184755592377795,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004988953182689445,
|
|
"loss": 6.0072,
|
|
"mean_token_accuracy": 0.13532012328505516,
|
|
"num_tokens": 2463497.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.134703922271728,
|
|
"epoch": 1.1226180613090306,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988634656236378,
|
|
"loss": 5.9216,
|
|
"mean_token_accuracy": 0.13609320148825646,
|
|
"num_tokens": 2471763.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.145527791976929,
|
|
"epoch": 1.1267605633802817,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004988311614212576,
|
|
"loss": 6.0016,
|
|
"mean_token_accuracy": 0.1402701161801815,
|
|
"num_tokens": 2479949.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.116470623016357,
|
|
"epoch": 1.1309030654515326,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004987984057269648,
|
|
"loss": 6.0539,
|
|
"mean_token_accuracy": 0.13444141745567323,
|
|
"num_tokens": 2489097.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.972180604934692,
|
|
"epoch": 1.1350455675227837,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987651986068311,
|
|
"loss": 5.8781,
|
|
"mean_token_accuracy": 0.13601028621196748,
|
|
"num_tokens": 2497625.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.143597984313965,
|
|
"epoch": 1.1391880695940348,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004987315401278394,
|
|
"loss": 6.1252,
|
|
"mean_token_accuracy": 0.1272761717438698,
|
|
"num_tokens": 2507107.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.286170959472656,
|
|
"epoch": 1.143330571665286,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004986974303578819,
|
|
"loss": 6.1408,
|
|
"mean_token_accuracy": 0.12348597794771195,
|
|
"num_tokens": 2516629.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.077923250198364,
|
|
"epoch": 1.1474730737365368,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004986628693657621,
|
|
"loss": 6.021,
|
|
"mean_token_accuracy": 0.1280151717364788,
|
|
"num_tokens": 2525924.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.229060649871826,
|
|
"epoch": 1.151615575807788,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004986278572211933,
|
|
"loss": 6.1723,
|
|
"mean_token_accuracy": 0.1280426375567913,
|
|
"num_tokens": 2535586.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.189875984191895,
|
|
"epoch": 1.1557580778790388,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004985923939947987,
|
|
"loss": 6.2595,
|
|
"mean_token_accuracy": 0.1247698113322258,
|
|
"num_tokens": 2546685.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.148470640182495,
|
|
"epoch": 1.15990057995029,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004985564797581115,
|
|
"loss": 5.9681,
|
|
"mean_token_accuracy": 0.12971177995204924,
|
|
"num_tokens": 2556873.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.143459701538086,
|
|
"epoch": 1.164043082021541,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004985201145835747,
|
|
"loss": 5.9836,
|
|
"mean_token_accuracy": 0.13127154856920242,
|
|
"num_tokens": 2565222.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.031957960128784,
|
|
"epoch": 1.1681855840927922,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000498483298544541,
|
|
"loss": 5.9875,
|
|
"mean_token_accuracy": 0.13211632743477822,
|
|
"num_tokens": 2575467.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.136318731307983,
|
|
"epoch": 1.172328086164043,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000498446031715272,
|
|
"loss": 6.0125,
|
|
"mean_token_accuracy": 0.13508019223809242,
|
|
"num_tokens": 2583963.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.133963680267334,
|
|
"epoch": 1.1764705882352942,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004984083141709391,
|
|
"loss": 6.042,
|
|
"mean_token_accuracy": 0.13524071872234344,
|
|
"num_tokens": 2593350.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.975436115264893,
|
|
"epoch": 1.180613090306545,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000498370145987623,
|
|
"loss": 5.9652,
|
|
"mean_token_accuracy": 0.13682079091668128,
|
|
"num_tokens": 2602191.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.170718050003051,
|
|
"epoch": 1.1847555923777962,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004983315272423127,
|
|
"loss": 6.0999,
|
|
"mean_token_accuracy": 0.1291058838367462,
|
|
"num_tokens": 2613590.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.049332237243652,
|
|
"epoch": 1.1888980944490473,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004982924580129066,
|
|
"loss": 5.8895,
|
|
"mean_token_accuracy": 0.1340464949607849,
|
|
"num_tokens": 2622072.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.093293142318726,
|
|
"epoch": 1.1930405965202984,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004982529383782116,
|
|
"loss": 6.1012,
|
|
"mean_token_accuracy": 0.13022035285830497,
|
|
"num_tokens": 2630904.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.189996242523193,
|
|
"epoch": 1.1971830985915493,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004982129684179432,
|
|
"loss": 6.0594,
|
|
"mean_token_accuracy": 0.13088259026408194,
|
|
"num_tokens": 2639925.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.1187131881713865,
|
|
"epoch": 1.2013256006628004,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004981725482127251,
|
|
"loss": 5.977,
|
|
"mean_token_accuracy": 0.13439981341362,
|
|
"num_tokens": 2649397.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.16315655708313,
|
|
"epoch": 1.2054681027340513,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004981316778440892,
|
|
"loss": 6.081,
|
|
"mean_token_accuracy": 0.13515593931078912,
|
|
"num_tokens": 2658877.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.063838720321655,
|
|
"epoch": 1.2096106048053024,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004980903573944756,
|
|
"loss": 6.0171,
|
|
"mean_token_accuracy": 0.13121431767940522,
|
|
"num_tokens": 2668518.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.059180212020874,
|
|
"epoch": 1.2137531068765535,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004980485869472322,
|
|
"loss": 6.1148,
|
|
"mean_token_accuracy": 0.12996160089969636,
|
|
"num_tokens": 2677459.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.223527908325195,
|
|
"epoch": 1.2178956089478046,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004980063665866143,
|
|
"loss": 6.0758,
|
|
"mean_token_accuracy": 0.12482839822769165,
|
|
"num_tokens": 2686668.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.935272264480591,
|
|
"epoch": 1.2220381110190555,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004979636963977853,
|
|
"loss": 5.8825,
|
|
"mean_token_accuracy": 0.1352998949587345,
|
|
"num_tokens": 2695545.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.066101598739624,
|
|
"epoch": 1.2261806130903066,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004979205764668156,
|
|
"loss": 6.0569,
|
|
"mean_token_accuracy": 0.13228241056203843,
|
|
"num_tokens": 2705514.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.138315343856812,
|
|
"epoch": 1.2303231151615575,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004978770068806825,
|
|
"loss": 5.9599,
|
|
"mean_token_accuracy": 0.13136665150523186,
|
|
"num_tokens": 2714373.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.061770391464234,
|
|
"epoch": 1.2344656172328086,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497832987727271,
|
|
"loss": 6.0715,
|
|
"mean_token_accuracy": 0.1264137916266918,
|
|
"num_tokens": 2723887.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.061756038665772,
|
|
"epoch": 1.2386081193040597,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004977885190953724,
|
|
"loss": 5.956,
|
|
"mean_token_accuracy": 0.13353569880127908,
|
|
"num_tokens": 2733591.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.104292392730713,
|
|
"epoch": 1.2427506213753108,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004977436010746846,
|
|
"loss": 6.066,
|
|
"mean_token_accuracy": 0.13720986992120743,
|
|
"num_tokens": 2741638.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2427506213753108,
|
|
"eval_entropy": 6.029743814773397,
|
|
"eval_loss": 6.2611613273620605,
|
|
"eval_mean_token_accuracy": 0.12890694399219332,
|
|
"eval_num_tokens": 2741638.0,
|
|
"eval_runtime": 2.9518,
|
|
"eval_samples_per_second": 1269.399,
|
|
"eval_steps_per_second": 158.887,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.078821802139283,
|
|
"epoch": 1.2468931234465617,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004976982337558124,
|
|
"loss": 5.9412,
|
|
"mean_token_accuracy": 0.13084649220108985,
|
|
"num_tokens": 2749711.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.094208097457885,
|
|
"epoch": 1.2510356255178128,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004976524172302665,
|
|
"loss": 6.121,
|
|
"mean_token_accuracy": 0.13192041367292404,
|
|
"num_tokens": 2758121.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.159745693206787,
|
|
"epoch": 1.2551781275890637,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004976061515904641,
|
|
"loss": 6.1181,
|
|
"mean_token_accuracy": 0.12667254582047463,
|
|
"num_tokens": 2766764.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.116680526733399,
|
|
"epoch": 1.2593206296603148,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004975594369297277,
|
|
"loss": 6.0557,
|
|
"mean_token_accuracy": 0.1348518520593643,
|
|
"num_tokens": 2775561.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.08063235282898,
|
|
"epoch": 1.2634631317315659,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004975122733422861,
|
|
"loss": 6.0201,
|
|
"mean_token_accuracy": 0.13085112273693084,
|
|
"num_tokens": 2784691.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.129113435745239,
|
|
"epoch": 1.267605633802817,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004974646609232732,
|
|
"loss": 5.9816,
|
|
"mean_token_accuracy": 0.13641836866736412,
|
|
"num_tokens": 2793893.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.108260202407837,
|
|
"epoch": 1.2717481358740679,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974165997687288,
|
|
"loss": 6.0751,
|
|
"mean_token_accuracy": 0.13353196606040002,
|
|
"num_tokens": 2804479.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.16022138595581,
|
|
"epoch": 1.275890637945319,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004973680899755972,
|
|
"loss": 6.107,
|
|
"mean_token_accuracy": 0.1326654002070427,
|
|
"num_tokens": 2813966.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.007886743545532,
|
|
"epoch": 1.2800331400165699,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000497319131641728,
|
|
"loss": 5.9326,
|
|
"mean_token_accuracy": 0.13846119418740271,
|
|
"num_tokens": 2822328.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.003260755538941,
|
|
"epoch": 1.284175642087821,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004972697248658755,
|
|
"loss": 5.8899,
|
|
"mean_token_accuracy": 0.14066707715392113,
|
|
"num_tokens": 2829845.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.948832178115845,
|
|
"epoch": 1.288318144159072,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004972198697476987,
|
|
"loss": 5.8783,
|
|
"mean_token_accuracy": 0.13914238885045052,
|
|
"num_tokens": 2839040.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.961882209777832,
|
|
"epoch": 1.2924606462303232,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004971695663877607,
|
|
"loss": 5.9665,
|
|
"mean_token_accuracy": 0.13561231270432472,
|
|
"num_tokens": 2848574.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.077086877822876,
|
|
"epoch": 1.296603148301574,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004971188148875289,
|
|
"loss": 5.9421,
|
|
"mean_token_accuracy": 0.14444561377167703,
|
|
"num_tokens": 2856812.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.051652336120606,
|
|
"epoch": 1.3007456503728252,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004970676153493748,
|
|
"loss": 6.0283,
|
|
"mean_token_accuracy": 0.13189572617411613,
|
|
"num_tokens": 2866001.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.1068840503692625,
|
|
"epoch": 1.3048881524440763,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004970159678765731,
|
|
"loss": 6.0164,
|
|
"mean_token_accuracy": 0.13239546120166779,
|
|
"num_tokens": 2875181.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.049808597564697,
|
|
"epoch": 1.3090306545153272,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004969638725733028,
|
|
"loss": 5.9587,
|
|
"mean_token_accuracy": 0.13397156223654746,
|
|
"num_tokens": 2885103.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.92131872177124,
|
|
"epoch": 1.3131731565865783,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004969113295446455,
|
|
"loss": 5.9116,
|
|
"mean_token_accuracy": 0.1383065924048424,
|
|
"num_tokens": 2894055.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.116864538192749,
|
|
"epoch": 1.3173156586578294,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004968583388965863,
|
|
"loss": 5.9846,
|
|
"mean_token_accuracy": 0.133075200766325,
|
|
"num_tokens": 2903391.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.006745624542236,
|
|
"epoch": 1.3214581607290803,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004968049007360132,
|
|
"loss": 5.9854,
|
|
"mean_token_accuracy": 0.13297903537750244,
|
|
"num_tokens": 2911760.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.926331329345703,
|
|
"epoch": 1.3256006628003314,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004967510151707169,
|
|
"loss": 5.9369,
|
|
"mean_token_accuracy": 0.13469143435359002,
|
|
"num_tokens": 2920589.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.15295524597168,
|
|
"epoch": 1.3297431648715825,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004966966823093903,
|
|
"loss": 6.0746,
|
|
"mean_token_accuracy": 0.1302391268312931,
|
|
"num_tokens": 2929635.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.986531114578247,
|
|
"epoch": 1.3338856669428334,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004966419022616287,
|
|
"loss": 5.9349,
|
|
"mean_token_accuracy": 0.13593596145510672,
|
|
"num_tokens": 2938778.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.039755535125733,
|
|
"epoch": 1.3380281690140845,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004965866751379296,
|
|
"loss": 6.0503,
|
|
"mean_token_accuracy": 0.13109695687890052,
|
|
"num_tokens": 2947836.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.133180904388428,
|
|
"epoch": 1.3421706710853356,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004965310010496922,
|
|
"loss": 6.0032,
|
|
"mean_token_accuracy": 0.1280514419078827,
|
|
"num_tokens": 2957028.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.026159048080444,
|
|
"epoch": 1.3463131731565865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004964748801092168,
|
|
"loss": 5.9571,
|
|
"mean_token_accuracy": 0.1338861681520939,
|
|
"num_tokens": 2966088.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.078959226608276,
|
|
"epoch": 1.3504556752278376,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004964183124297059,
|
|
"loss": 6.0438,
|
|
"mean_token_accuracy": 0.13178688064217567,
|
|
"num_tokens": 2975271.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.992210865020752,
|
|
"epoch": 1.3545981772990887,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004963612981252625,
|
|
"loss": 6.0524,
|
|
"mean_token_accuracy": 0.1321226365864277,
|
|
"num_tokens": 2983799.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.056794929504394,
|
|
"epoch": 1.3587406793703396,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004963038373108908,
|
|
"loss": 6.0179,
|
|
"mean_token_accuracy": 0.13609340488910676,
|
|
"num_tokens": 2992675.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.155933427810669,
|
|
"epoch": 1.3628831814415907,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004962459301024954,
|
|
"loss": 6.0838,
|
|
"mean_token_accuracy": 0.13211635649204254,
|
|
"num_tokens": 3002483.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.027872896194458,
|
|
"epoch": 1.3670256835128418,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004961875766168816,
|
|
"loss": 5.985,
|
|
"mean_token_accuracy": 0.13883458971977233,
|
|
"num_tokens": 3010611.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.057289028167725,
|
|
"epoch": 1.371168185584093,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004961287769717547,
|
|
"loss": 5.9397,
|
|
"mean_token_accuracy": 0.13197456523776055,
|
|
"num_tokens": 3019928.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.9325989246368405,
|
|
"epoch": 1.3753106876553438,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004960695312857199,
|
|
"loss": 5.9279,
|
|
"mean_token_accuracy": 0.13408055454492568,
|
|
"num_tokens": 3029003.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.0389604568481445,
|
|
"epoch": 1.379453189726595,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004960098396782824,
|
|
"loss": 5.9464,
|
|
"mean_token_accuracy": 0.1375861033797264,
|
|
"num_tokens": 3038049.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.969662141799927,
|
|
"epoch": 1.3835956917978458,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004959497022698466,
|
|
"loss": 5.9515,
|
|
"mean_token_accuracy": 0.14408120438456534,
|
|
"num_tokens": 3046674.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.0561059474945065,
|
|
"epoch": 1.387738193869097,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004958891191817163,
|
|
"loss": 5.9927,
|
|
"mean_token_accuracy": 0.12801324650645257,
|
|
"num_tokens": 3056459.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.032327270507812,
|
|
"epoch": 1.391880695940348,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000495828090536094,
|
|
"loss": 5.9845,
|
|
"mean_token_accuracy": 0.13036760538816453,
|
|
"num_tokens": 3066294.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.0512574195861815,
|
|
"epoch": 1.3960231980115991,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004957666164560815,
|
|
"loss": 5.9412,
|
|
"mean_token_accuracy": 0.13139644563198088,
|
|
"num_tokens": 3075420.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.063203239440918,
|
|
"epoch": 1.40016570008285,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004957046970656784,
|
|
"loss": 6.0385,
|
|
"mean_token_accuracy": 0.13581736162304878,
|
|
"num_tokens": 3084373.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.00097541809082,
|
|
"epoch": 1.4043082021541011,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004956423324897831,
|
|
"loss": 6.0257,
|
|
"mean_token_accuracy": 0.13937310948967935,
|
|
"num_tokens": 3094212.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.043115425109863,
|
|
"epoch": 1.408450704225352,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004955795228541917,
|
|
"loss": 5.9589,
|
|
"mean_token_accuracy": 0.14452596232295037,
|
|
"num_tokens": 3102769.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.040110731124878,
|
|
"epoch": 1.4125932062966031,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000495516268285598,
|
|
"loss": 5.9805,
|
|
"mean_token_accuracy": 0.13440805301070213,
|
|
"num_tokens": 3111465.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.978239822387695,
|
|
"epoch": 1.4167357083678542,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004954525689115935,
|
|
"loss": 5.9177,
|
|
"mean_token_accuracy": 0.13326097056269645,
|
|
"num_tokens": 3121106.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.007881164550781,
|
|
"epoch": 1.4208782104391053,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004953884248606669,
|
|
"loss": 5.945,
|
|
"mean_token_accuracy": 0.14831372052431108,
|
|
"num_tokens": 3130662.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.97642674446106,
|
|
"epoch": 1.4250207125103562,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004953238362622034,
|
|
"loss": 5.9812,
|
|
"mean_token_accuracy": 0.13328263983130456,
|
|
"num_tokens": 3139538.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.0256367206573485,
|
|
"epoch": 1.4291632145816073,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004952588032464855,
|
|
"loss": 6.0298,
|
|
"mean_token_accuracy": 0.12707214802503586,
|
|
"num_tokens": 3148552.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.996851205825806,
|
|
"epoch": 1.4333057166528582,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004951933259446919,
|
|
"loss": 5.9365,
|
|
"mean_token_accuracy": 0.14197203740477563,
|
|
"num_tokens": 3156549.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.114988136291504,
|
|
"epoch": 1.4374482187241093,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004951274044888975,
|
|
"loss": 6.1006,
|
|
"mean_token_accuracy": 0.12147112190723419,
|
|
"num_tokens": 3166821.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.960144329071045,
|
|
"epoch": 1.4415907207953604,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004950610390120731,
|
|
"loss": 5.9848,
|
|
"mean_token_accuracy": 0.135729306191206,
|
|
"num_tokens": 3174962.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.087535095214844,
|
|
"epoch": 1.4457332228666115,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000494994229648085,
|
|
"loss": 5.8903,
|
|
"mean_token_accuracy": 0.13537728488445283,
|
|
"num_tokens": 3184027.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.957475805282593,
|
|
"epoch": 1.4498757249378624,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004949269765316948,
|
|
"loss": 6.0321,
|
|
"mean_token_accuracy": 0.13071285635232927,
|
|
"num_tokens": 3193186.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.171153354644775,
|
|
"epoch": 1.4540182270091135,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 0.0004948592797985599,
|
|
"loss": 6.0525,
|
|
"mean_token_accuracy": 0.13555259853601456,
|
|
"num_tokens": 3202676.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.123908805847168,
|
|
"epoch": 1.4581607290803644,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004947911395852317,
|
|
"loss": 6.1524,
|
|
"mean_token_accuracy": 0.1352410912513733,
|
|
"num_tokens": 3211950.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.120599699020386,
|
|
"epoch": 1.4623032311516155,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004947225560291565,
|
|
"loss": 6.0758,
|
|
"mean_token_accuracy": 0.12961760386824608,
|
|
"num_tokens": 3221135.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.962483358383179,
|
|
"epoch": 1.4664457332228666,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004946535292686746,
|
|
"loss": 5.8975,
|
|
"mean_token_accuracy": 0.13825839310884475,
|
|
"num_tokens": 3230174.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.979943370819091,
|
|
"epoch": 1.4705882352941178,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004945840594430209,
|
|
"loss": 5.8871,
|
|
"mean_token_accuracy": 0.1381668671965599,
|
|
"num_tokens": 3238898.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.973515844345092,
|
|
"epoch": 1.4747307373653686,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004945141466923235,
|
|
"loss": 5.9292,
|
|
"mean_token_accuracy": 0.14479121565818787,
|
|
"num_tokens": 3247550.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.113377714157105,
|
|
"epoch": 1.4788732394366197,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004944437911576041,
|
|
"loss": 6.0779,
|
|
"mean_token_accuracy": 0.12900342419743538,
|
|
"num_tokens": 3256810.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.9020613670349125,
|
|
"epoch": 1.4830157415078706,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004943729929807772,
|
|
"loss": 5.877,
|
|
"mean_token_accuracy": 0.139550269395113,
|
|
"num_tokens": 3265886.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.06021089553833,
|
|
"epoch": 1.4871582435791217,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004943017523046507,
|
|
"loss": 5.9273,
|
|
"mean_token_accuracy": 0.137468421459198,
|
|
"num_tokens": 3274522.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.027210426330567,
|
|
"epoch": 1.4913007456503728,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000494230069272925,
|
|
"loss": 6.0486,
|
|
"mean_token_accuracy": 0.13558472618460654,
|
|
"num_tokens": 3283388.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.989331531524658,
|
|
"epoch": 1.495443247721624,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004941579440301923,
|
|
"loss": 5.9224,
|
|
"mean_token_accuracy": 0.1305856943130493,
|
|
"num_tokens": 3291887.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.96136965751648,
|
|
"epoch": 1.4995857497928748,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004940853767219372,
|
|
"loss": 5.9685,
|
|
"mean_token_accuracy": 0.13564062863588333,
|
|
"num_tokens": 3300674.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.9364429950714115,
|
|
"epoch": 1.503728251864126,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004940123674945359,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.13808463513851166,
|
|
"num_tokens": 3309745.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.076629638671875,
|
|
"epoch": 1.5078707539353768,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004939389164952561,
|
|
"loss": 5.9722,
|
|
"mean_token_accuracy": 0.1378110609948635,
|
|
"num_tokens": 3318888.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.8970863819122314,
|
|
"epoch": 1.512013256006628,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004938650238722564,
|
|
"loss": 5.9296,
|
|
"mean_token_accuracy": 0.13823849260807036,
|
|
"num_tokens": 3328274.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.003948831558228,
|
|
"epoch": 1.516155758077879,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004937906897745861,
|
|
"loss": 5.9387,
|
|
"mean_token_accuracy": 0.1468382939696312,
|
|
"num_tokens": 3337579.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.045965003967285,
|
|
"epoch": 1.5202982601491302,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004937159143521856,
|
|
"loss": 5.9081,
|
|
"mean_token_accuracy": 0.14555162489414214,
|
|
"num_tokens": 3346262.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.972390079498291,
|
|
"epoch": 1.5244407622203813,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004936406977558847,
|
|
"loss": 6.0156,
|
|
"mean_token_accuracy": 0.13030771315097808,
|
|
"num_tokens": 3354833.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.051387977600098,
|
|
"epoch": 1.5285832642916322,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004935650401374039,
|
|
"loss": 6.0906,
|
|
"mean_token_accuracy": 0.14119647368788718,
|
|
"num_tokens": 3364314.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.046140813827515,
|
|
"epoch": 1.532725766362883,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004934889416493524,
|
|
"loss": 5.9157,
|
|
"mean_token_accuracy": 0.13524662256240844,
|
|
"num_tokens": 3373648.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.863892507553101,
|
|
"epoch": 1.5368682684341342,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004934124024452295,
|
|
"loss": 5.7869,
|
|
"mean_token_accuracy": 0.1471114367246628,
|
|
"num_tokens": 3382919.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.888700532913208,
|
|
"epoch": 1.5410107705053853,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004933354226794232,
|
|
"loss": 5.9383,
|
|
"mean_token_accuracy": 0.1432707317173481,
|
|
"num_tokens": 3391814.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.080925798416137,
|
|
"epoch": 1.5451532725766364,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004932580025072098,
|
|
"loss": 5.9919,
|
|
"mean_token_accuracy": 0.13580586463212968,
|
|
"num_tokens": 3400392.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.996384525299073,
|
|
"epoch": 1.5492957746478875,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004931801420847545,
|
|
"loss": 6.036,
|
|
"mean_token_accuracy": 0.13089759424328803,
|
|
"num_tokens": 3410430.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.932257795333863,
|
|
"epoch": 1.5534382767191384,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004931018415691102,
|
|
"loss": 5.877,
|
|
"mean_token_accuracy": 0.13694921880960464,
|
|
"num_tokens": 3420654.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.951343011856079,
|
|
"epoch": 1.5575807787903893,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004930231011182176,
|
|
"loss": 5.9986,
|
|
"mean_token_accuracy": 0.13937177062034606,
|
|
"num_tokens": 3430514.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.071512508392334,
|
|
"epoch": 1.5617232808616404,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004929439208909048,
|
|
"loss": 5.9331,
|
|
"mean_token_accuracy": 0.1375361390411854,
|
|
"num_tokens": 3439551.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.902947330474854,
|
|
"epoch": 1.5658657829328915,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004928643010468869,
|
|
"loss": 5.8911,
|
|
"mean_token_accuracy": 0.13622887432575226,
|
|
"num_tokens": 3448160.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.993059539794922,
|
|
"epoch": 1.5700082850041426,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004927842417467659,
|
|
"loss": 5.9517,
|
|
"mean_token_accuracy": 0.14548188596963882,
|
|
"num_tokens": 3457296.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.027637386322022,
|
|
"epoch": 1.5741507870753937,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004927037431520302,
|
|
"loss": 5.9073,
|
|
"mean_token_accuracy": 0.14102519080042838,
|
|
"num_tokens": 3466108.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.975209665298462,
|
|
"epoch": 1.5782932891466446,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004926228054250541,
|
|
"loss": 5.9941,
|
|
"mean_token_accuracy": 0.14009249955415726,
|
|
"num_tokens": 3475336.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.073916673660278,
|
|
"epoch": 1.5824357912178955,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004925414287290979,
|
|
"loss": 5.9668,
|
|
"mean_token_accuracy": 0.13723743334412575,
|
|
"num_tokens": 3483963.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.848618030548096,
|
|
"epoch": 1.5865782932891466,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004924596132283075,
|
|
"loss": 5.8765,
|
|
"mean_token_accuracy": 0.14485997408628465,
|
|
"num_tokens": 3492947.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.966853284835816,
|
|
"epoch": 1.5907207953603977,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004923773590877135,
|
|
"loss": 5.9651,
|
|
"mean_token_accuracy": 0.13334692791104316,
|
|
"num_tokens": 3501877.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.042904996871949,
|
|
"epoch": 1.5948632974316488,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004922946664732314,
|
|
"loss": 6.0304,
|
|
"mean_token_accuracy": 0.13204361274838447,
|
|
"num_tokens": 3510610.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.0583656311035154,
|
|
"epoch": 1.5990057995029,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004922115355516615,
|
|
"loss": 5.9841,
|
|
"mean_token_accuracy": 0.13304307833313941,
|
|
"num_tokens": 3519829.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.015839862823486,
|
|
"epoch": 1.6031483015741508,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004921279664906877,
|
|
"loss": 5.9661,
|
|
"mean_token_accuracy": 0.13071200698614122,
|
|
"num_tokens": 3529408.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.924501991271972,
|
|
"epoch": 1.6072908036454017,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004920439594588781,
|
|
"loss": 5.8955,
|
|
"mean_token_accuracy": 0.14675913602113724,
|
|
"num_tokens": 3538068.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.051935148239136,
|
|
"epoch": 1.6114333057166528,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000491959514625684,
|
|
"loss": 6.0177,
|
|
"mean_token_accuracy": 0.13845598921179772,
|
|
"num_tokens": 3547302.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.956210565567017,
|
|
"epoch": 1.6155758077879039,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00049187463216144,
|
|
"loss": 5.9024,
|
|
"mean_token_accuracy": 0.145584287494421,
|
|
"num_tokens": 3556087.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.046857452392578,
|
|
"epoch": 1.619718309859155,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000491789312237363,
|
|
"loss": 5.9857,
|
|
"mean_token_accuracy": 0.13579782098531723,
|
|
"num_tokens": 3566235.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.983897924423218,
|
|
"epoch": 1.623860811930406,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004917035550255528,
|
|
"loss": 5.899,
|
|
"mean_token_accuracy": 0.14092317298054696,
|
|
"num_tokens": 3574762.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.904513597488403,
|
|
"epoch": 1.628003314001657,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000491617360698991,
|
|
"loss": 5.8837,
|
|
"mean_token_accuracy": 0.139695380628109,
|
|
"num_tokens": 3583359.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.920687866210938,
|
|
"epoch": 1.6321458160729079,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004915307294315412,
|
|
"loss": 5.8505,
|
|
"mean_token_accuracy": 0.14011457338929176,
|
|
"num_tokens": 3591436.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.926318550109864,
|
|
"epoch": 1.636288318144159,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000491443661397948,
|
|
"loss": 5.9355,
|
|
"mean_token_accuracy": 0.13468224480748175,
|
|
"num_tokens": 3601054.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.974443817138672,
|
|
"epoch": 1.64043082021541,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004913561567738368,
|
|
"loss": 5.9014,
|
|
"mean_token_accuracy": 0.1383673779666424,
|
|
"num_tokens": 3610641.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.912512636184692,
|
|
"epoch": 1.6445733222866612,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004912682157357146,
|
|
"loss": 5.8447,
|
|
"mean_token_accuracy": 0.13975749984383584,
|
|
"num_tokens": 3619351.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.853101539611816,
|
|
"epoch": 1.6487158243579123,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004911798384609675,
|
|
"loss": 5.8306,
|
|
"mean_token_accuracy": 0.1472959376871586,
|
|
"num_tokens": 3627341.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.942045879364014,
|
|
"epoch": 1.6528583264291632,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004910910251278626,
|
|
"loss": 5.8918,
|
|
"mean_token_accuracy": 0.14801610112190247,
|
|
"num_tokens": 3637132.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.919465351104736,
|
|
"epoch": 1.6570008285004143,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004910017759155458,
|
|
"loss": 5.9479,
|
|
"mean_token_accuracy": 0.1400071859359741,
|
|
"num_tokens": 3646196.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6570008285004143,
|
|
"eval_entropy": 5.874394122979789,
|
|
"eval_loss": 6.1148576736450195,
|
|
"eval_mean_token_accuracy": 0.1363023624046525,
|
|
"eval_num_tokens": 3646196.0,
|
|
"eval_runtime": 2.7574,
|
|
"eval_samples_per_second": 1358.871,
|
|
"eval_steps_per_second": 170.085,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.967284250259399,
|
|
"epoch": 1.6611433305716652,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004909120910040427,
|
|
"loss": 5.8839,
|
|
"mean_token_accuracy": 0.1461731344461441,
|
|
"num_tokens": 3654658.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.045904684066772,
|
|
"epoch": 1.6652858326429163,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004908219705742574,
|
|
"loss": 6.1605,
|
|
"mean_token_accuracy": 0.13489469960331918,
|
|
"num_tokens": 3664509.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.002130031585693,
|
|
"epoch": 1.6694283347141674,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004907314148079727,
|
|
"loss": 5.8567,
|
|
"mean_token_accuracy": 0.1425239197909832,
|
|
"num_tokens": 3672983.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.708412790298462,
|
|
"epoch": 1.6735708367854185,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004906404238878496,
|
|
"loss": 5.7221,
|
|
"mean_token_accuracy": 0.14639501571655272,
|
|
"num_tokens": 3681196.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.97982873916626,
|
|
"epoch": 1.6777133388566694,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004905489979974265,
|
|
"loss": 5.9098,
|
|
"mean_token_accuracy": 0.14081615954637527,
|
|
"num_tokens": 3690077.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.942774772644043,
|
|
"epoch": 1.6818558409279205,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004904571373211197,
|
|
"loss": 5.9906,
|
|
"mean_token_accuracy": 0.13563016131520272,
|
|
"num_tokens": 3700121.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.897730588912964,
|
|
"epoch": 1.6859983429991714,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000490364842044222,
|
|
"loss": 5.8074,
|
|
"mean_token_accuracy": 0.14530285224318504,
|
|
"num_tokens": 3709618.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.954602289199829,
|
|
"epoch": 1.6901408450704225,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004902721123529031,
|
|
"loss": 5.9869,
|
|
"mean_token_accuracy": 0.12953415215015412,
|
|
"num_tokens": 3719570.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.074023342132568,
|
|
"epoch": 1.6942833471416736,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000490178948434209,
|
|
"loss": 5.9267,
|
|
"mean_token_accuracy": 0.14249598532915114,
|
|
"num_tokens": 3729354.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.833365964889526,
|
|
"epoch": 1.6984258492129247,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004900853504760614,
|
|
"loss": 5.8409,
|
|
"mean_token_accuracy": 0.14450331926345825,
|
|
"num_tokens": 3738453.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.888468265533447,
|
|
"epoch": 1.7025683512841756,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004899913186672577,
|
|
"loss": 5.8165,
|
|
"mean_token_accuracy": 0.14293194860219954,
|
|
"num_tokens": 3747815.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.914711236953735,
|
|
"epoch": 1.7067108533554267,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004898968531974702,
|
|
"loss": 5.9249,
|
|
"mean_token_accuracy": 0.14642315953969956,
|
|
"num_tokens": 3757298.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.978992033004761,
|
|
"epoch": 1.7108533554266776,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004898019542572462,
|
|
"loss": 5.9124,
|
|
"mean_token_accuracy": 0.14719650894403458,
|
|
"num_tokens": 3765939.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.005469894409179,
|
|
"epoch": 1.7149958574979287,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004897066220380073,
|
|
"loss": 5.933,
|
|
"mean_token_accuracy": 0.13779435828328132,
|
|
"num_tokens": 3775160.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.921374797821045,
|
|
"epoch": 1.7191383595691798,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004896108567320489,
|
|
"loss": 5.831,
|
|
"mean_token_accuracy": 0.14186858981847764,
|
|
"num_tokens": 3785181.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.972924375534058,
|
|
"epoch": 1.723280861640431,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048951465853254,
|
|
"loss": 5.9444,
|
|
"mean_token_accuracy": 0.14323145076632499,
|
|
"num_tokens": 3794531.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.953138160705566,
|
|
"epoch": 1.7274233637116818,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004894180276335231,
|
|
"loss": 5.934,
|
|
"mean_token_accuracy": 0.1396125227212906,
|
|
"num_tokens": 3804602.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.9721667766571045,
|
|
"epoch": 1.731565865782933,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004893209642299132,
|
|
"loss": 5.97,
|
|
"mean_token_accuracy": 0.1395590640604496,
|
|
"num_tokens": 3813394.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.97849326133728,
|
|
"epoch": 1.7357083678541838,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004892234685174976,
|
|
"loss": 5.962,
|
|
"mean_token_accuracy": 0.13808128908276557,
|
|
"num_tokens": 3821604.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.967495775222778,
|
|
"epoch": 1.739850869925435,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004891255406929359,
|
|
"loss": 5.9564,
|
|
"mean_token_accuracy": 0.13652343899011612,
|
|
"num_tokens": 3831977.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.98737063407898,
|
|
"epoch": 1.743993371996686,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004890271809537595,
|
|
"loss": 5.9017,
|
|
"mean_token_accuracy": 0.1393343411386013,
|
|
"num_tokens": 3840684.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.870862340927124,
|
|
"epoch": 1.7481358740679371,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004889283894983704,
|
|
"loss": 5.8147,
|
|
"mean_token_accuracy": 0.1389534741640091,
|
|
"num_tokens": 3851147.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.002704238891601,
|
|
"epoch": 1.752278376139188,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000488829166526042,
|
|
"loss": 5.9838,
|
|
"mean_token_accuracy": 0.13991605415940284,
|
|
"num_tokens": 3859881.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.948231506347656,
|
|
"epoch": 1.7564208782104391,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004887295122369176,
|
|
"loss": 6.0107,
|
|
"mean_token_accuracy": 0.13371825441718102,
|
|
"num_tokens": 3870151.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.99525203704834,
|
|
"epoch": 1.76056338028169,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000488629426832011,
|
|
"loss": 5.9313,
|
|
"mean_token_accuracy": 0.13850879445672035,
|
|
"num_tokens": 3878398.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.924004030227661,
|
|
"epoch": 1.7647058823529411,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004885289105132054,
|
|
"loss": 5.8893,
|
|
"mean_token_accuracy": 0.14114601090550422,
|
|
"num_tokens": 3888007.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.930246257781983,
|
|
"epoch": 1.7688483844241922,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004884279634832532,
|
|
"loss": 5.8728,
|
|
"mean_token_accuracy": 0.14278061985969542,
|
|
"num_tokens": 3896945.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.940092372894287,
|
|
"epoch": 1.7729908864954433,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004883265859457754,
|
|
"loss": 5.8905,
|
|
"mean_token_accuracy": 0.1457919418811798,
|
|
"num_tokens": 3906049.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.866062164306641,
|
|
"epoch": 1.7771333885666944,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048822477810526186,
|
|
"loss": 5.8422,
|
|
"mean_token_accuracy": 0.1388966679573059,
|
|
"num_tokens": 3915050.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.966696119308471,
|
|
"epoch": 1.7812758906379453,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00048812254016706994,
|
|
"loss": 5.9205,
|
|
"mean_token_accuracy": 0.14728582575917243,
|
|
"num_tokens": 3924031.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.915832757949829,
|
|
"epoch": 1.7854183927091962,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004880198723374247,
|
|
"loss": 5.9776,
|
|
"mean_token_accuracy": 0.14149753525853156,
|
|
"num_tokens": 3933756.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.959774684906006,
|
|
"epoch": 1.7895608947804473,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048791677482341845,
|
|
"loss": 5.9592,
|
|
"mean_token_accuracy": 0.13750381171703338,
|
|
"num_tokens": 3943129.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.881538438796997,
|
|
"epoch": 1.7937033968516984,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048781324783301016,
|
|
"loss": 5.7912,
|
|
"mean_token_accuracy": 0.14508036673069,
|
|
"num_tokens": 3951494.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.946735525131226,
|
|
"epoch": 1.7978458989229495,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048770929157502505,
|
|
"loss": 5.9565,
|
|
"mean_token_accuracy": 0.143115097284317,
|
|
"num_tokens": 3960588.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.9805809497833256,
|
|
"epoch": 1.8019884009942007,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004876049062591543,
|
|
"loss": 5.9226,
|
|
"mean_token_accuracy": 0.13506680428981782,
|
|
"num_tokens": 3970135.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.950060653686523,
|
|
"epoch": 1.8061309030654515,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048750009209595443,
|
|
"loss": 5.9042,
|
|
"mean_token_accuracy": 0.13944360390305519,
|
|
"num_tokens": 3979569.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.88271656036377,
|
|
"epoch": 1.8102734051367024,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048739484929684714,
|
|
"loss": 5.8446,
|
|
"mean_token_accuracy": 0.144245932251215,
|
|
"num_tokens": 3987885.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.933677864074707,
|
|
"epoch": 1.8144159072079535,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048728917807411856,
|
|
"loss": 6.0216,
|
|
"mean_token_accuracy": 0.13125594183802605,
|
|
"num_tokens": 3997726.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.008653068542481,
|
|
"epoch": 1.8185584092792046,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004871830786409191,
|
|
"loss": 5.8436,
|
|
"mean_token_accuracy": 0.14898036047816277,
|
|
"num_tokens": 4006437.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.856683444976807,
|
|
"epoch": 1.8227009113504558,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00048707655121126296,
|
|
"loss": 5.8124,
|
|
"mean_token_accuracy": 0.14927435666322708,
|
|
"num_tokens": 4015326.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.924995994567871,
|
|
"epoch": 1.8268434134217069,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048696959600002767,
|
|
"loss": 5.8722,
|
|
"mean_token_accuracy": 0.14032577201724053,
|
|
"num_tokens": 4024294.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.881852293014527,
|
|
"epoch": 1.8309859154929577,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048686221322295347,
|
|
"loss": 5.9028,
|
|
"mean_token_accuracy": 0.14340249821543694,
|
|
"num_tokens": 4033341.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.951443433761597,
|
|
"epoch": 1.8351284175642086,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048675440309664314,
|
|
"loss": 5.8446,
|
|
"mean_token_accuracy": 0.1474481076002121,
|
|
"num_tokens": 4042446.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.832607173919678,
|
|
"epoch": 1.8392709196354597,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048664616583856166,
|
|
"loss": 5.8924,
|
|
"mean_token_accuracy": 0.13868724405765534,
|
|
"num_tokens": 4051849.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.026660823822022,
|
|
"epoch": 1.8434134217067109,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048653750166703516,
|
|
"loss": 5.9855,
|
|
"mean_token_accuracy": 0.1381051518023014,
|
|
"num_tokens": 4061075.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.976817560195923,
|
|
"epoch": 1.847555923777962,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004864284108012513,
|
|
"loss": 5.9524,
|
|
"mean_token_accuracy": 0.13498588278889656,
|
|
"num_tokens": 4071022.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.884273719787598,
|
|
"epoch": 1.851698425849213,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004863188934612583,
|
|
"loss": 5.854,
|
|
"mean_token_accuracy": 0.1435965158045292,
|
|
"num_tokens": 4080728.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.8848789691925045,
|
|
"epoch": 1.855840927920464,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004862089498679645,
|
|
"loss": 5.8121,
|
|
"mean_token_accuracy": 0.15146355628967284,
|
|
"num_tokens": 4088712.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.904728364944458,
|
|
"epoch": 1.8599834299917148,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004860985802431383,
|
|
"loss": 5.8445,
|
|
"mean_token_accuracy": 0.14107747822999955,
|
|
"num_tokens": 4097488.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.815548610687256,
|
|
"epoch": 1.864125932062966,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004859877848094071,
|
|
"loss": 5.8328,
|
|
"mean_token_accuracy": 0.14647450745105745,
|
|
"num_tokens": 4106485.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.969652986526489,
|
|
"epoch": 1.868268434134217,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004858765637902575,
|
|
"loss": 5.8016,
|
|
"mean_token_accuracy": 0.14588006138801574,
|
|
"num_tokens": 4115806.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.982199335098267,
|
|
"epoch": 1.8724109362054682,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004857649174100344,
|
|
"loss": 5.9353,
|
|
"mean_token_accuracy": 0.14478921219706536,
|
|
"num_tokens": 4124636.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.869907474517822,
|
|
"epoch": 1.8765534382767193,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004856528458939409,
|
|
"loss": 5.8948,
|
|
"mean_token_accuracy": 0.13403457626700402,
|
|
"num_tokens": 4133876.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.836872100830078,
|
|
"epoch": 1.8806959403479702,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004855403494680373,
|
|
"loss": 5.8159,
|
|
"mean_token_accuracy": 0.14096100330352784,
|
|
"num_tokens": 4143555.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.871636962890625,
|
|
"epoch": 1.884838442419221,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048542742835924133,
|
|
"loss": 5.7773,
|
|
"mean_token_accuracy": 0.14253296032547952,
|
|
"num_tokens": 4153007.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.766701364517212,
|
|
"epoch": 1.8889809444904722,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048531408279532705,
|
|
"loss": 5.8171,
|
|
"mean_token_accuracy": 0.14552054405212403,
|
|
"num_tokens": 4162107.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.884533357620239,
|
|
"epoch": 1.8931234465617233,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048520031300492507,
|
|
"loss": 5.8208,
|
|
"mean_token_accuracy": 0.14203491136431695,
|
|
"num_tokens": 4170920.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.9367118835449215,
|
|
"epoch": 1.8972659486329744,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048508611921752124,
|
|
"loss": 5.8366,
|
|
"mean_token_accuracy": 0.14547978863120078,
|
|
"num_tokens": 4180226.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.804201984405518,
|
|
"epoch": 1.9014084507042255,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048497150166345703,
|
|
"loss": 5.7799,
|
|
"mean_token_accuracy": 0.14307984560728074,
|
|
"num_tokens": 4189283.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.833661937713623,
|
|
"epoch": 1.9055509527754764,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004848564605739287,
|
|
"loss": 5.8181,
|
|
"mean_token_accuracy": 0.14286661818623542,
|
|
"num_tokens": 4199017.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.985245418548584,
|
|
"epoch": 1.9096934548467275,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048474099618098644,
|
|
"loss": 5.9557,
|
|
"mean_token_accuracy": 0.13776218444108962,
|
|
"num_tokens": 4207756.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.926583385467529,
|
|
"epoch": 1.9138359569179784,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004846251087175348,
|
|
"loss": 5.8467,
|
|
"mean_token_accuracy": 0.14270823299884797,
|
|
"num_tokens": 4216721.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.816466999053955,
|
|
"epoch": 1.9179784589892295,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004845087984173312,
|
|
"loss": 5.8214,
|
|
"mean_token_accuracy": 0.14952016696333886,
|
|
"num_tokens": 4225596.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.975145244598389,
|
|
"epoch": 1.9221209610604806,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048439206551498644,
|
|
"loss": 5.8881,
|
|
"mean_token_accuracy": 0.14604572355747222,
|
|
"num_tokens": 4234500.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.83410792350769,
|
|
"epoch": 1.9262634631317317,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004842749102459635,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.14477182626724244,
|
|
"num_tokens": 4243240.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.799895906448365,
|
|
"epoch": 1.9304059652029826,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048415733284657727,
|
|
"loss": 5.8249,
|
|
"mean_token_accuracy": 0.14724751114845275,
|
|
"num_tokens": 4251892.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.914888668060303,
|
|
"epoch": 1.9345484672742337,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004840393335539943,
|
|
"loss": 5.784,
|
|
"mean_token_accuracy": 0.14537777602672577,
|
|
"num_tokens": 4261447.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.93565788269043,
|
|
"epoch": 1.9386909693454846,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048392091260623195,
|
|
"loss": 5.9202,
|
|
"mean_token_accuracy": 0.1471864774823189,
|
|
"num_tokens": 4270120.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.905978488922119,
|
|
"epoch": 1.9428334714167357,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048380207024215836,
|
|
"loss": 5.7991,
|
|
"mean_token_accuracy": 0.1482786476612091,
|
|
"num_tokens": 4278300.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.905145263671875,
|
|
"epoch": 1.9469759734879868,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00048368280670149145,
|
|
"loss": 5.8487,
|
|
"mean_token_accuracy": 0.1461447760462761,
|
|
"num_tokens": 4286454.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.805000495910645,
|
|
"epoch": 1.951118475559238,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004835631222247989,
|
|
"loss": 5.7243,
|
|
"mean_token_accuracy": 0.15050614029169082,
|
|
"num_tokens": 4294752.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.907592630386352,
|
|
"epoch": 1.9552609776304888,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048344301705349735,
|
|
"loss": 5.7974,
|
|
"mean_token_accuracy": 0.14830030649900436,
|
|
"num_tokens": 4303195.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.841587924957276,
|
|
"epoch": 1.95940347970174,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048332249142985195,
|
|
"loss": 5.9211,
|
|
"mean_token_accuracy": 0.1373597078025341,
|
|
"num_tokens": 4311961.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.900353670120239,
|
|
"epoch": 1.9635459817729908,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004832015455969763,
|
|
"loss": 5.7748,
|
|
"mean_token_accuracy": 0.13879901468753814,
|
|
"num_tokens": 4322267.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.774641275405884,
|
|
"epoch": 1.9676884838442419,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048308017979883114,
|
|
"loss": 5.8838,
|
|
"mean_token_accuracy": 0.1426581546664238,
|
|
"num_tokens": 4330584.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.840176010131836,
|
|
"epoch": 1.971830985915493,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048295839428022454,
|
|
"loss": 5.7215,
|
|
"mean_token_accuracy": 0.14949245899915695,
|
|
"num_tokens": 4339215.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.869688034057617,
|
|
"epoch": 1.975973487986744,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048283618928681127,
|
|
"loss": 5.7418,
|
|
"mean_token_accuracy": 0.1501791574060917,
|
|
"num_tokens": 4347711.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.764533996582031,
|
|
"epoch": 1.980115990057995,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004827135650650923,
|
|
"loss": 5.7445,
|
|
"mean_token_accuracy": 0.14758301600813867,
|
|
"num_tokens": 4356982.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.770398235321045,
|
|
"epoch": 1.984258492129246,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004825905218624139,
|
|
"loss": 5.7696,
|
|
"mean_token_accuracy": 0.14773683473467827,
|
|
"num_tokens": 4365378.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.9908288478851315,
|
|
"epoch": 1.988400994200497,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048246705992696756,
|
|
"loss": 5.8174,
|
|
"mean_token_accuracy": 0.14142358154058457,
|
|
"num_tokens": 4374007.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.9741987705230715,
|
|
"epoch": 1.992543496271748,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048234317950778975,
|
|
"loss": 5.9346,
|
|
"mean_token_accuracy": 0.1409923940896988,
|
|
"num_tokens": 4383909.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.967488718032837,
|
|
"epoch": 1.9966859983429992,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004822188808547607,
|
|
"loss": 5.9967,
|
|
"mean_token_accuracy": 0.13560240790247918,
|
|
"num_tokens": 4393145.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.751282596588135,
|
|
"epoch": 2.0008285004142503,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004820941642186043,
|
|
"loss": 5.7146,
|
|
"mean_token_accuracy": 0.14043126851320267,
|
|
"num_tokens": 4402024.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.91456356048584,
|
|
"epoch": 2.0049710024855014,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004819690298508879,
|
|
"loss": 5.5395,
|
|
"mean_token_accuracy": 0.1603303700685501,
|
|
"num_tokens": 4411377.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.758482313156128,
|
|
"epoch": 2.009113504556752,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048184347800402083,
|
|
"loss": 5.4299,
|
|
"mean_token_accuracy": 0.16564757078886033,
|
|
"num_tokens": 4419781.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.727555131912231,
|
|
"epoch": 2.013256006628003,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004817175089312551,
|
|
"loss": 5.4154,
|
|
"mean_token_accuracy": 0.16400155574083328,
|
|
"num_tokens": 4428427.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.749885892868042,
|
|
"epoch": 2.0173985086992543,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048159112288668403,
|
|
"loss": 5.4586,
|
|
"mean_token_accuracy": 0.15933601409196854,
|
|
"num_tokens": 4438156.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.876013088226318,
|
|
"epoch": 2.0215410107705054,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048146432012524215,
|
|
"loss": 5.4819,
|
|
"mean_token_accuracy": 0.16306492537260056,
|
|
"num_tokens": 4446342.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.738983201980591,
|
|
"epoch": 2.0256835128417565,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00048133710090270446,
|
|
"loss": 5.5229,
|
|
"mean_token_accuracy": 0.1542331963777542,
|
|
"num_tokens": 4455582.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.815389633178711,
|
|
"epoch": 2.0298260149130076,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048120946547568596,
|
|
"loss": 5.5375,
|
|
"mean_token_accuracy": 0.15604802072048188,
|
|
"num_tokens": 4464518.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.927202749252319,
|
|
"epoch": 2.0339685169842583,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00048108141410164133,
|
|
"loss": 5.6588,
|
|
"mean_token_accuracy": 0.14534952044486998,
|
|
"num_tokens": 4475031.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.8722837448120115,
|
|
"epoch": 2.0381110190555094,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004809529470388642,
|
|
"loss": 5.5396,
|
|
"mean_token_accuracy": 0.1577153742313385,
|
|
"num_tokens": 4484609.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.787671709060669,
|
|
"epoch": 2.0422535211267605,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004808240645464866,
|
|
"loss": 5.5698,
|
|
"mean_token_accuracy": 0.15477177649736404,
|
|
"num_tokens": 4494238.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.684235668182373,
|
|
"epoch": 2.0463960231980116,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004806947668844787,
|
|
"loss": 5.4573,
|
|
"mean_token_accuracy": 0.16172746568918228,
|
|
"num_tokens": 4503258.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.786364412307739,
|
|
"epoch": 2.0505385252692627,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048056505431364794,
|
|
"loss": 5.5091,
|
|
"mean_token_accuracy": 0.15975141078233718,
|
|
"num_tokens": 4512071.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.736668491363526,
|
|
"epoch": 2.054681027340514,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048043492709563877,
|
|
"loss": 5.3461,
|
|
"mean_token_accuracy": 0.17059434503316878,
|
|
"num_tokens": 4520219.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.672813034057617,
|
|
"epoch": 2.0588235294117645,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048030438549293193,
|
|
"loss": 5.5568,
|
|
"mean_token_accuracy": 0.15284566581249237,
|
|
"num_tokens": 4529386.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.830966377258301,
|
|
"epoch": 2.0629660314830156,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048017342976884424,
|
|
"loss": 5.5638,
|
|
"mean_token_accuracy": 0.15199215039610864,
|
|
"num_tokens": 4538860.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.761076402664185,
|
|
"epoch": 2.0671085335542667,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048004206018752773,
|
|
"loss": 5.5299,
|
|
"mean_token_accuracy": 0.15335657596588134,
|
|
"num_tokens": 4547842.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.733208799362183,
|
|
"epoch": 2.071251035625518,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047991027701396895,
|
|
"loss": 5.5026,
|
|
"mean_token_accuracy": 0.15411428660154342,
|
|
"num_tokens": 4556238.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.071251035625518,
|
|
"eval_entropy": 5.642364594473768,
|
|
"eval_loss": 6.032641410827637,
|
|
"eval_mean_token_accuracy": 0.1416530203399882,
|
|
"eval_num_tokens": 4556238.0,
|
|
"eval_runtime": 2.7817,
|
|
"eval_samples_per_second": 1347.034,
|
|
"eval_steps_per_second": 168.604,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.732073068618774,
|
|
"epoch": 2.075393537696769,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004797780805139892,
|
|
"loss": 5.5069,
|
|
"mean_token_accuracy": 0.1587075412273407,
|
|
"num_tokens": 4565684.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.722462224960327,
|
|
"epoch": 2.07953603976802,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004796454709542432,
|
|
"loss": 5.467,
|
|
"mean_token_accuracy": 0.1590041533112526,
|
|
"num_tokens": 4575001.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.797696113586426,
|
|
"epoch": 2.083678541839271,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004795124486022189,
|
|
"loss": 5.5732,
|
|
"mean_token_accuracy": 0.15362634360790253,
|
|
"num_tokens": 4584128.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.711380529403686,
|
|
"epoch": 2.087821043910522,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047937901372623694,
|
|
"loss": 5.4191,
|
|
"mean_token_accuracy": 0.16894411891698838,
|
|
"num_tokens": 4593011.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.688450145721435,
|
|
"epoch": 2.091963545981773,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047924516659545005,
|
|
"loss": 5.4668,
|
|
"mean_token_accuracy": 0.16164053976535797,
|
|
"num_tokens": 4602676.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.79151463508606,
|
|
"epoch": 2.096106048053024,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004791109074798425,
|
|
"loss": 5.5522,
|
|
"mean_token_accuracy": 0.15433578938245773,
|
|
"num_tokens": 4611915.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.748647403717041,
|
|
"epoch": 2.100248550124275,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004789762366502296,
|
|
"loss": 5.5184,
|
|
"mean_token_accuracy": 0.15959300100803375,
|
|
"num_tokens": 4621396.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.7374598503112795,
|
|
"epoch": 2.1043910521955262,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004788411543782571,
|
|
"loss": 5.5201,
|
|
"mean_token_accuracy": 0.1591552659869194,
|
|
"num_tokens": 4630533.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.678498649597168,
|
|
"epoch": 2.108533554266777,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004787056609364008,
|
|
"loss": 5.4599,
|
|
"mean_token_accuracy": 0.1542729303240776,
|
|
"num_tokens": 4639023.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.711474943161011,
|
|
"epoch": 2.112676056338028,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004785697565979657,
|
|
"loss": 5.4732,
|
|
"mean_token_accuracy": 0.15900285094976424,
|
|
"num_tokens": 4647975.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.72883243560791,
|
|
"epoch": 2.116818558409279,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004784334416370857,
|
|
"loss": 5.498,
|
|
"mean_token_accuracy": 0.15594277828931807,
|
|
"num_tokens": 4657555.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.745093822479248,
|
|
"epoch": 2.1209610604805302,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004782967163287231,
|
|
"loss": 5.4193,
|
|
"mean_token_accuracy": 0.16100153774023057,
|
|
"num_tokens": 4667228.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.716836738586426,
|
|
"epoch": 2.1251035625517813,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047815958094866756,
|
|
"loss": 5.4957,
|
|
"mean_token_accuracy": 0.1513967014849186,
|
|
"num_tokens": 4676723.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.7276163578033445,
|
|
"epoch": 2.1292460646230325,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004780220357735363,
|
|
"loss": 5.4896,
|
|
"mean_token_accuracy": 0.15479894131422042,
|
|
"num_tokens": 4686241.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.764811038970947,
|
|
"epoch": 2.1333885666942836,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047788408108077294,
|
|
"loss": 5.5404,
|
|
"mean_token_accuracy": 0.1561874568462372,
|
|
"num_tokens": 4695195.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.739376497268677,
|
|
"epoch": 2.1375310687655342,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004777457171486471,
|
|
"loss": 5.5871,
|
|
"mean_token_accuracy": 0.15551834553480148,
|
|
"num_tokens": 4704716.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.769967555999756,
|
|
"epoch": 2.1416735708367853,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004776069442562541,
|
|
"loss": 5.5562,
|
|
"mean_token_accuracy": 0.15339450240135194,
|
|
"num_tokens": 4713720.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.733342504501342,
|
|
"epoch": 2.1458160729080364,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004774677626835138,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.15301723331212996,
|
|
"num_tokens": 4722428.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.787648677825928,
|
|
"epoch": 2.1499585749792876,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047732817271117084,
|
|
"loss": 5.5365,
|
|
"mean_token_accuracy": 0.15334974452853203,
|
|
"num_tokens": 4732395.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.773520803451538,
|
|
"epoch": 2.1541010770505387,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004771881746207934,
|
|
"loss": 5.5206,
|
|
"mean_token_accuracy": 0.15674114525318145,
|
|
"num_tokens": 4741071.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.607013797760009,
|
|
"epoch": 2.1582435791217898,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047704776869477285,
|
|
"loss": 5.5153,
|
|
"mean_token_accuracy": 0.15076877027750016,
|
|
"num_tokens": 4749876.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.784313154220581,
|
|
"epoch": 2.1623860811930404,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047690695521632344,
|
|
"loss": 5.4669,
|
|
"mean_token_accuracy": 0.16002854704856873,
|
|
"num_tokens": 4758721.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.6919313907623295,
|
|
"epoch": 2.1665285832642915,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047676573446948114,
|
|
"loss": 5.4832,
|
|
"mean_token_accuracy": 0.15925751626491547,
|
|
"num_tokens": 4767162.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.71550760269165,
|
|
"epoch": 2.1706710853355426,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047662410673910376,
|
|
"loss": 5.5151,
|
|
"mean_token_accuracy": 0.1629578560590744,
|
|
"num_tokens": 4776397.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.665811157226562,
|
|
"epoch": 2.1748135874067938,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047648207231087,
|
|
"loss": 5.4579,
|
|
"mean_token_accuracy": 0.15894875824451446,
|
|
"num_tokens": 4784822.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.71003270149231,
|
|
"epoch": 2.178956089478045,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004763396314712786,
|
|
"loss": 5.5297,
|
|
"mean_token_accuracy": 0.15831661075353623,
|
|
"num_tokens": 4794272.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.64026665687561,
|
|
"epoch": 2.183098591549296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004761967845076484,
|
|
"loss": 5.453,
|
|
"mean_token_accuracy": 0.15510014593601226,
|
|
"num_tokens": 4804314.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.6847330093383786,
|
|
"epoch": 2.1872410936205466,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004760535317081174,
|
|
"loss": 5.4694,
|
|
"mean_token_accuracy": 0.15845878273248673,
|
|
"num_tokens": 4812685.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.715235567092895,
|
|
"epoch": 2.1913835956917977,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047590987336164207,
|
|
"loss": 5.5532,
|
|
"mean_token_accuracy": 0.1562129870057106,
|
|
"num_tokens": 4821761.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.741413831710815,
|
|
"epoch": 2.195526097763049,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047576580975799705,
|
|
"loss": 5.5307,
|
|
"mean_token_accuracy": 0.1623195379972458,
|
|
"num_tokens": 4830515.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.782008934020996,
|
|
"epoch": 2.1996685998343,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004756213411877743,
|
|
"loss": 5.5059,
|
|
"mean_token_accuracy": 0.14879967421293258,
|
|
"num_tokens": 4840741.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.612497425079345,
|
|
"epoch": 2.203811101905551,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047547646794238277,
|
|
"loss": 5.453,
|
|
"mean_token_accuracy": 0.16198884546756745,
|
|
"num_tokens": 4849571.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.689143991470337,
|
|
"epoch": 2.207953603976802,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047533119031404757,
|
|
"loss": 5.4138,
|
|
"mean_token_accuracy": 0.1696420982480049,
|
|
"num_tokens": 4857194.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.632621717453003,
|
|
"epoch": 2.212096106048053,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047518550859580966,
|
|
"loss": 5.5074,
|
|
"mean_token_accuracy": 0.1592964008450508,
|
|
"num_tokens": 4865894.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.74289755821228,
|
|
"epoch": 2.216238608119304,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004750394230815249,
|
|
"loss": 5.5268,
|
|
"mean_token_accuracy": 0.15829544067382811,
|
|
"num_tokens": 4875131.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.670539522171021,
|
|
"epoch": 2.220381110190555,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047489293406586375,
|
|
"loss": 5.3777,
|
|
"mean_token_accuracy": 0.1620385766029358,
|
|
"num_tokens": 4884098.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.657269620895386,
|
|
"epoch": 2.224523612261806,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004747460418443106,
|
|
"loss": 5.4373,
|
|
"mean_token_accuracy": 0.15941016003489494,
|
|
"num_tokens": 4892963.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.690485191345215,
|
|
"epoch": 2.2286661143330573,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047459874671316305,
|
|
"loss": 5.4903,
|
|
"mean_token_accuracy": 0.15749651044607163,
|
|
"num_tokens": 4902097.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.608357000350952,
|
|
"epoch": 2.2328086164043084,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047445104896953153,
|
|
"loss": 5.4015,
|
|
"mean_token_accuracy": 0.16532779037952422,
|
|
"num_tokens": 4910469.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.82715744972229,
|
|
"epoch": 2.236951118475559,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047430294891133846,
|
|
"loss": 5.6892,
|
|
"mean_token_accuracy": 0.1501491665840149,
|
|
"num_tokens": 4921024.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.749764776229858,
|
|
"epoch": 2.24109362054681,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004741544468373179,
|
|
"loss": 5.5329,
|
|
"mean_token_accuracy": 0.15474629402160645,
|
|
"num_tokens": 4930297.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.687776947021485,
|
|
"epoch": 2.2452361226180613,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047400554304701474,
|
|
"loss": 5.4692,
|
|
"mean_token_accuracy": 0.16049361675977708,
|
|
"num_tokens": 4938971.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.571053695678711,
|
|
"epoch": 2.2493786246893124,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047385623784078427,
|
|
"loss": 5.4899,
|
|
"mean_token_accuracy": 0.15803302526474,
|
|
"num_tokens": 4948531.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.787879133224488,
|
|
"epoch": 2.2535211267605635,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004737065315197913,
|
|
"loss": 5.5664,
|
|
"mean_token_accuracy": 0.1587745025753975,
|
|
"num_tokens": 4957430.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.720958280563354,
|
|
"epoch": 2.2576636288318146,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047355642438600987,
|
|
"loss": 5.5478,
|
|
"mean_token_accuracy": 0.16044806241989135,
|
|
"num_tokens": 4966681.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.638753175735474,
|
|
"epoch": 2.2618061309030653,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004734059167422225,
|
|
"loss": 5.4123,
|
|
"mean_token_accuracy": 0.1639780133962631,
|
|
"num_tokens": 4975327.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.634887552261352,
|
|
"epoch": 2.2659486329743164,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047325500889201956,
|
|
"loss": 5.4725,
|
|
"mean_token_accuracy": 0.16207436099648476,
|
|
"num_tokens": 4983741.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.673438644409179,
|
|
"epoch": 2.2700911350455675,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004731037011397986,
|
|
"loss": 5.5662,
|
|
"mean_token_accuracy": 0.15884149968624114,
|
|
"num_tokens": 4993171.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.752307653427124,
|
|
"epoch": 2.2742336371168186,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047295199379076403,
|
|
"loss": 5.6,
|
|
"mean_token_accuracy": 0.15322241187095642,
|
|
"num_tokens": 5002810.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.7073346138000485,
|
|
"epoch": 2.2783761391880697,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000472799887150926,
|
|
"loss": 5.4893,
|
|
"mean_token_accuracy": 0.15741414427757264,
|
|
"num_tokens": 5011875.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.679959297180176,
|
|
"epoch": 2.282518641259321,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004726473815271003,
|
|
"loss": 5.5349,
|
|
"mean_token_accuracy": 0.15855301022529603,
|
|
"num_tokens": 5020780.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.669374132156372,
|
|
"epoch": 2.286661143330572,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047249447722690747,
|
|
"loss": 5.4331,
|
|
"mean_token_accuracy": 0.16046710908412934,
|
|
"num_tokens": 5029896.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.652454280853272,
|
|
"epoch": 2.2908036454018226,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047234117455877215,
|
|
"loss": 5.4932,
|
|
"mean_token_accuracy": 0.1545606277883053,
|
|
"num_tokens": 5038956.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.639697694778443,
|
|
"epoch": 2.2949461474730737,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004721874738319225,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.17258906960487366,
|
|
"num_tokens": 5047005.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.655842065811157,
|
|
"epoch": 2.299088649544325,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004720333753563898,
|
|
"loss": 5.5008,
|
|
"mean_token_accuracy": 0.15116662830114364,
|
|
"num_tokens": 5056475.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.704644584655762,
|
|
"epoch": 2.303231151615576,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047187887944300747,
|
|
"loss": 5.4825,
|
|
"mean_token_accuracy": 0.16439782381057738,
|
|
"num_tokens": 5064978.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.726413679122925,
|
|
"epoch": 2.307373653686827,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004717239864034107,
|
|
"loss": 5.578,
|
|
"mean_token_accuracy": 0.15506226867437362,
|
|
"num_tokens": 5074235.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.662968683242798,
|
|
"epoch": 2.3115161557580777,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047156869655003554,
|
|
"loss": 5.4481,
|
|
"mean_token_accuracy": 0.16141191571950914,
|
|
"num_tokens": 5082386.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.631324100494385,
|
|
"epoch": 2.315658657829329,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047141301019611875,
|
|
"loss": 5.4164,
|
|
"mean_token_accuracy": 0.16636510640382768,
|
|
"num_tokens": 5090634.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.818879795074463,
|
|
"epoch": 2.31980115990058,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004712569276556967,
|
|
"loss": 5.5906,
|
|
"mean_token_accuracy": 0.14937217235565187,
|
|
"num_tokens": 5099003.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.617757177352905,
|
|
"epoch": 2.323943661971831,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00047110044924360484,
|
|
"loss": 5.6681,
|
|
"mean_token_accuracy": 0.1565680392086506,
|
|
"num_tokens": 5108631.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.732515859603882,
|
|
"epoch": 2.328086164043082,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047094357527547725,
|
|
"loss": 5.4545,
|
|
"mean_token_accuracy": 0.15828853249549865,
|
|
"num_tokens": 5117309.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.686485624313354,
|
|
"epoch": 2.332228666114333,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004707863060677459,
|
|
"loss": 5.5411,
|
|
"mean_token_accuracy": 0.1627561032772064,
|
|
"num_tokens": 5126305.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.821760654449463,
|
|
"epoch": 2.3363711681855843,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047062864193764,
|
|
"loss": 5.6022,
|
|
"mean_token_accuracy": 0.15580221712589265,
|
|
"num_tokens": 5135338.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.676449155807495,
|
|
"epoch": 2.340513670256835,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047047058320318525,
|
|
"loss": 5.5298,
|
|
"mean_token_accuracy": 0.16162888556718827,
|
|
"num_tokens": 5144518.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.702890157699585,
|
|
"epoch": 2.344656172328086,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004703121301832034,
|
|
"loss": 5.6262,
|
|
"mean_token_accuracy": 0.15101130157709122,
|
|
"num_tokens": 5154102.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.7891247272491455,
|
|
"epoch": 2.348798674399337,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047015328319731153,
|
|
"loss": 5.5642,
|
|
"mean_token_accuracy": 0.15612353384494781,
|
|
"num_tokens": 5163719.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.677855682373047,
|
|
"epoch": 2.3529411764705883,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004699940425659215,
|
|
"loss": 5.5247,
|
|
"mean_token_accuracy": 0.15519936084747316,
|
|
"num_tokens": 5173192.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.622324132919312,
|
|
"epoch": 2.3570836785418394,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046983440861023877,
|
|
"loss": 5.4626,
|
|
"mean_token_accuracy": 0.1663823500275612,
|
|
"num_tokens": 5182438.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.699106454849243,
|
|
"epoch": 2.36122618061309,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046967438165226275,
|
|
"loss": 5.6015,
|
|
"mean_token_accuracy": 0.15352728664875032,
|
|
"num_tokens": 5191595.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.659265756607056,
|
|
"epoch": 2.365368682684341,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004695139620147851,
|
|
"loss": 5.4682,
|
|
"mean_token_accuracy": 0.1643056571483612,
|
|
"num_tokens": 5200174.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.713242053985596,
|
|
"epoch": 2.3695111847555923,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046935315002138994,
|
|
"loss": 5.5804,
|
|
"mean_token_accuracy": 0.15417435318231582,
|
|
"num_tokens": 5210454.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.727554702758789,
|
|
"epoch": 2.3736536868268434,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004691919459964525,
|
|
"loss": 5.5019,
|
|
"mean_token_accuracy": 0.16056040227413176,
|
|
"num_tokens": 5219243.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.5862212657928465,
|
|
"epoch": 2.3777961888980945,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046903035026513894,
|
|
"loss": 5.4319,
|
|
"mean_token_accuracy": 0.1613735228776932,
|
|
"num_tokens": 5228656.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.587289905548095,
|
|
"epoch": 2.3819386909693456,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004688683631534056,
|
|
"loss": 5.4859,
|
|
"mean_token_accuracy": 0.156366066634655,
|
|
"num_tokens": 5237807.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.775411653518677,
|
|
"epoch": 2.3860811930405967,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004687059849879981,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.1609719067811966,
|
|
"num_tokens": 5247432.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.597807741165161,
|
|
"epoch": 2.3902236951118474,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046854321609645093,
|
|
"loss": 5.4145,
|
|
"mean_token_accuracy": 0.16111126840114592,
|
|
"num_tokens": 5256680.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.6380726337432865,
|
|
"epoch": 2.3943661971830985,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 0.00046838005680708677,
|
|
"loss": 5.5296,
|
|
"mean_token_accuracy": 0.14821895360946655,
|
|
"num_tokens": 5266793.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.710390281677246,
|
|
"epoch": 2.3985086992543496,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046821650744901573,
|
|
"loss": 5.4452,
|
|
"mean_token_accuracy": 0.15857197493314742,
|
|
"num_tokens": 5275028.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.598506593704224,
|
|
"epoch": 2.4026512013256007,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004680525683521347,
|
|
"loss": 5.5226,
|
|
"mean_token_accuracy": 0.1566852644085884,
|
|
"num_tokens": 5284162.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.757242631912232,
|
|
"epoch": 2.406793703396852,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004678882398471269,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.15474308133125306,
|
|
"num_tokens": 5293263.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.654598569869995,
|
|
"epoch": 2.4109362054681025,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046772352226546066,
|
|
"loss": 5.477,
|
|
"mean_token_accuracy": 0.1666618600487709,
|
|
"num_tokens": 5301373.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.643638467788696,
|
|
"epoch": 2.4150787075393536,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004675584159393895,
|
|
"loss": 5.5151,
|
|
"mean_token_accuracy": 0.15203802064061164,
|
|
"num_tokens": 5311009.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.699149656295776,
|
|
"epoch": 2.4192212096106047,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046739292120195083,
|
|
"loss": 5.5245,
|
|
"mean_token_accuracy": 0.16085631400346756,
|
|
"num_tokens": 5320341.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.642888498306275,
|
|
"epoch": 2.423363711681856,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004672270383869657,
|
|
"loss": 5.4896,
|
|
"mean_token_accuracy": 0.1610630214214325,
|
|
"num_tokens": 5329250.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.62935357093811,
|
|
"epoch": 2.427506213753107,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004670607678290379,
|
|
"loss": 5.4471,
|
|
"mean_token_accuracy": 0.16571807861328125,
|
|
"num_tokens": 5336515.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.655415058135986,
|
|
"epoch": 2.431648715824358,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004668941098635532,
|
|
"loss": 5.5345,
|
|
"mean_token_accuracy": 0.1595415398478508,
|
|
"num_tokens": 5346043.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.642036104202271,
|
|
"epoch": 2.435791217895609,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046672706482667897,
|
|
"loss": 5.4925,
|
|
"mean_token_accuracy": 0.15832271426916122,
|
|
"num_tokens": 5355124.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.647988367080688,
|
|
"epoch": 2.43993371996686,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004665596330553633,
|
|
"loss": 5.5051,
|
|
"mean_token_accuracy": 0.1597411409020424,
|
|
"num_tokens": 5364556.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.67459397315979,
|
|
"epoch": 2.444076222038111,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046639181488733445,
|
|
"loss": 5.5316,
|
|
"mean_token_accuracy": 0.15941161364316941,
|
|
"num_tokens": 5373778.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.725692987442017,
|
|
"epoch": 2.448218724109362,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004662236106610999,
|
|
"loss": 5.5123,
|
|
"mean_token_accuracy": 0.15816445648670197,
|
|
"num_tokens": 5382633.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.5606146335601805,
|
|
"epoch": 2.452361226180613,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046605502071594604,
|
|
"loss": 5.4163,
|
|
"mean_token_accuracy": 0.16387673169374467,
|
|
"num_tokens": 5391442.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.569208765029908,
|
|
"epoch": 2.4565037282518642,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004658860453919371,
|
|
"loss": 5.4388,
|
|
"mean_token_accuracy": 0.1621423229575157,
|
|
"num_tokens": 5400504.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.587312078475952,
|
|
"epoch": 2.460646230323115,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004657166850299149,
|
|
"loss": 5.4448,
|
|
"mean_token_accuracy": 0.16361159086227417,
|
|
"num_tokens": 5408939.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.629118537902832,
|
|
"epoch": 2.464788732394366,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004655469399714978,
|
|
"loss": 5.4722,
|
|
"mean_token_accuracy": 0.1575179159641266,
|
|
"num_tokens": 5418934.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.572103595733642,
|
|
"epoch": 2.468931234465617,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004653768105590799,
|
|
"loss": 5.4955,
|
|
"mean_token_accuracy": 0.16008511185646057,
|
|
"num_tokens": 5427894.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.683330011367798,
|
|
"epoch": 2.4730737365368682,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046520629713583115,
|
|
"loss": 5.5003,
|
|
"mean_token_accuracy": 0.1613202139735222,
|
|
"num_tokens": 5436830.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.597742080688477,
|
|
"epoch": 2.4772162386081193,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046503540004569555,
|
|
"loss": 5.4814,
|
|
"mean_token_accuracy": 0.16423845291137695,
|
|
"num_tokens": 5445504.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.679887342453003,
|
|
"epoch": 2.4813587406793705,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046486411963339125,
|
|
"loss": 5.5405,
|
|
"mean_token_accuracy": 0.15501321628689765,
|
|
"num_tokens": 5455378.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.712473011016845,
|
|
"epoch": 2.4855012427506216,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046469245624440955,
|
|
"loss": 5.6107,
|
|
"mean_token_accuracy": 0.1589423656463623,
|
|
"num_tokens": 5465334.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.4855012427506216,
|
|
"eval_entropy": 5.590163738234465,
|
|
"eval_loss": 5.965892314910889,
|
|
"eval_mean_token_accuracy": 0.14734836894947329,
|
|
"eval_num_tokens": 5465334.0,
|
|
"eval_runtime": 2.7526,
|
|
"eval_samples_per_second": 1361.234,
|
|
"eval_steps_per_second": 170.381,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.751126146316528,
|
|
"epoch": 2.4896437448218722,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046452041022501423,
|
|
"loss": 5.5858,
|
|
"mean_token_accuracy": 0.15555662512779236,
|
|
"num_tokens": 5475701.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.699931526184082,
|
|
"epoch": 2.4937862468931233,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004643479819222411,
|
|
"loss": 5.4828,
|
|
"mean_token_accuracy": 0.16229774355888366,
|
|
"num_tokens": 5485244.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.596950387954712,
|
|
"epoch": 2.4979287489643744,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046417517168389664,
|
|
"loss": 5.4434,
|
|
"mean_token_accuracy": 0.1682340309023857,
|
|
"num_tokens": 5493587.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.6203169345855715,
|
|
"epoch": 2.5020712510356256,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004640019798585581,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.15738681554794312,
|
|
"num_tokens": 5503117.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.762682390213013,
|
|
"epoch": 2.5062137531068767,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046382840679557233,
|
|
"loss": 5.5352,
|
|
"mean_token_accuracy": 0.16024575680494307,
|
|
"num_tokens": 5512742.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.647499704360962,
|
|
"epoch": 2.5103562551781273,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046365445284505516,
|
|
"loss": 5.4968,
|
|
"mean_token_accuracy": 0.16166605204343795,
|
|
"num_tokens": 5521862.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.597900772094727,
|
|
"epoch": 2.5144987572493784,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004634801183578907,
|
|
"loss": 5.488,
|
|
"mean_token_accuracy": 0.1608501434326172,
|
|
"num_tokens": 5531244.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.655751419067383,
|
|
"epoch": 2.5186412593206295,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046330540368573063,
|
|
"loss": 5.5329,
|
|
"mean_token_accuracy": 0.15176820158958435,
|
|
"num_tokens": 5540786.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.7291028022766115,
|
|
"epoch": 2.5227837613918807,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046313030918099365,
|
|
"loss": 5.5224,
|
|
"mean_token_accuracy": 0.1602902241051197,
|
|
"num_tokens": 5549961.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.620108366012573,
|
|
"epoch": 2.5269262634631318,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046295483519686437,
|
|
"loss": 5.4591,
|
|
"mean_token_accuracy": 0.16417455673217773,
|
|
"num_tokens": 5558941.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.644694232940674,
|
|
"epoch": 2.531068765534383,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046277898208729306,
|
|
"loss": 5.4673,
|
|
"mean_token_accuracy": 0.16716189235448836,
|
|
"num_tokens": 5567630.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.586954879760742,
|
|
"epoch": 2.535211267605634,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004626027502069947,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.16521373838186265,
|
|
"num_tokens": 5576430.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.650178289413452,
|
|
"epoch": 2.539353769676885,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046242613991144814,
|
|
"loss": 5.5444,
|
|
"mean_token_accuracy": 0.1609214127063751,
|
|
"num_tokens": 5584785.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.585511493682861,
|
|
"epoch": 2.5434962717481358,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046224915155689586,
|
|
"loss": 5.478,
|
|
"mean_token_accuracy": 0.16044741719961167,
|
|
"num_tokens": 5594583.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.638197040557861,
|
|
"epoch": 2.547638773819387,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046207178550034244,
|
|
"loss": 5.4306,
|
|
"mean_token_accuracy": 0.1608849883079529,
|
|
"num_tokens": 5603498.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.7161966323852536,
|
|
"epoch": 2.551781275890638,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004618940420995547,
|
|
"loss": 5.5685,
|
|
"mean_token_accuracy": 0.15440558269619942,
|
|
"num_tokens": 5613101.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.665849494934082,
|
|
"epoch": 2.555923777961889,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004617159217130606,
|
|
"loss": 5.5409,
|
|
"mean_token_accuracy": 0.1569191187620163,
|
|
"num_tokens": 5621821.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.711861181259155,
|
|
"epoch": 2.5600662800331397,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004615374247001483,
|
|
"loss": 5.5738,
|
|
"mean_token_accuracy": 0.15701586306095122,
|
|
"num_tokens": 5631006.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.674423551559448,
|
|
"epoch": 2.564208782104391,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004613585514208659,
|
|
"loss": 5.5295,
|
|
"mean_token_accuracy": 0.15232180580496787,
|
|
"num_tokens": 5640159.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.592060613632202,
|
|
"epoch": 2.568351284175642,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046117930223602016,
|
|
"loss": 5.4003,
|
|
"mean_token_accuracy": 0.1694741129875183,
|
|
"num_tokens": 5648369.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.5982965469360355,
|
|
"epoch": 2.572493786246893,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004609996775071765,
|
|
"loss": 5.459,
|
|
"mean_token_accuracy": 0.1591326355934143,
|
|
"num_tokens": 5657669.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.56780686378479,
|
|
"epoch": 2.576636288318144,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004608196775966575,
|
|
"loss": 5.5319,
|
|
"mean_token_accuracy": 0.1610974356532097,
|
|
"num_tokens": 5666601.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.639773225784301,
|
|
"epoch": 2.5807787903893953,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046063930286754267,
|
|
"loss": 5.4465,
|
|
"mean_token_accuracy": 0.16618527621030807,
|
|
"num_tokens": 5675243.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.56855001449585,
|
|
"epoch": 2.5849212924606464,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046045855368366756,
|
|
"loss": 5.466,
|
|
"mean_token_accuracy": 0.1630399063229561,
|
|
"num_tokens": 5683646.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.682586193084717,
|
|
"epoch": 2.5890637945318975,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000460277430409623,
|
|
"loss": 5.5129,
|
|
"mean_token_accuracy": 0.17162407487630843,
|
|
"num_tokens": 5693069.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.587219524383545,
|
|
"epoch": 2.593206296603148,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046009593341075444,
|
|
"loss": 5.3754,
|
|
"mean_token_accuracy": 0.16499391198158264,
|
|
"num_tokens": 5701527.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.561956834793091,
|
|
"epoch": 2.5973487986743993,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004599140630531612,
|
|
"loss": 5.5403,
|
|
"mean_token_accuracy": 0.15682477951049806,
|
|
"num_tokens": 5710379.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.69692587852478,
|
|
"epoch": 2.6014913007456504,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004597318197036956,
|
|
"loss": 5.4524,
|
|
"mean_token_accuracy": 0.1643553927540779,
|
|
"num_tokens": 5719217.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.629226970672607,
|
|
"epoch": 2.6056338028169015,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004595492037299624,
|
|
"loss": 5.5535,
|
|
"mean_token_accuracy": 0.16214104294776915,
|
|
"num_tokens": 5728331.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.608061027526856,
|
|
"epoch": 2.6097763048881526,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045936621550031806,
|
|
"loss": 5.5107,
|
|
"mean_token_accuracy": 0.16073764860630035,
|
|
"num_tokens": 5737696.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.731885099411011,
|
|
"epoch": 2.6139188069594033,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045918285538386973,
|
|
"loss": 5.578,
|
|
"mean_token_accuracy": 0.16164542436599733,
|
|
"num_tokens": 5746191.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.703127813339234,
|
|
"epoch": 2.6180613090306544,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045899912375047485,
|
|
"loss": 5.5619,
|
|
"mean_token_accuracy": 0.1629435494542122,
|
|
"num_tokens": 5754824.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.638229656219482,
|
|
"epoch": 2.6222038111019055,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045881502097074023,
|
|
"loss": 5.537,
|
|
"mean_token_accuracy": 0.1518124133348465,
|
|
"num_tokens": 5764863.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.614318180084228,
|
|
"epoch": 2.6263463131731566,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004586305474160212,
|
|
"loss": 5.5459,
|
|
"mean_token_accuracy": 0.15642993301153182,
|
|
"num_tokens": 5774273.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.666326427459717,
|
|
"epoch": 2.6304888152444077,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045844570345842116,
|
|
"loss": 5.5044,
|
|
"mean_token_accuracy": 0.16042826995253562,
|
|
"num_tokens": 5784085.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.659371709823608,
|
|
"epoch": 2.634631317315659,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004582604894707906,
|
|
"loss": 5.5336,
|
|
"mean_token_accuracy": 0.16884370446205138,
|
|
"num_tokens": 5793921.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.664378356933594,
|
|
"epoch": 2.63877381938691,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045807490582672634,
|
|
"loss": 5.4972,
|
|
"mean_token_accuracy": 0.15294444859027861,
|
|
"num_tokens": 5803902.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.542481708526611,
|
|
"epoch": 2.6429163214581606,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045788895290057083,
|
|
"loss": 5.4129,
|
|
"mean_token_accuracy": 0.16788192689418793,
|
|
"num_tokens": 5812089.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.6802209377288815,
|
|
"epoch": 2.6470588235294117,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045770263106741137,
|
|
"loss": 5.5784,
|
|
"mean_token_accuracy": 0.1578851893544197,
|
|
"num_tokens": 5821072.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.628516340255738,
|
|
"epoch": 2.651201325600663,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004575159407030797,
|
|
"loss": 5.4029,
|
|
"mean_token_accuracy": 0.16388181746006011,
|
|
"num_tokens": 5829725.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.500157403945923,
|
|
"epoch": 2.655343827671914,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004573288821841505,
|
|
"loss": 5.4825,
|
|
"mean_token_accuracy": 0.16496249735355378,
|
|
"num_tokens": 5838153.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.616616106033325,
|
|
"epoch": 2.659486329743165,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004571414558879413,
|
|
"loss": 5.4409,
|
|
"mean_token_accuracy": 0.15448423773050307,
|
|
"num_tokens": 5847270.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.6490891456604,
|
|
"epoch": 2.6636288318144157,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004569536621925114,
|
|
"loss": 5.5472,
|
|
"mean_token_accuracy": 0.15474478006362916,
|
|
"num_tokens": 5855954.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.634416961669922,
|
|
"epoch": 2.667771333885667,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004567655014766612,
|
|
"loss": 5.4529,
|
|
"mean_token_accuracy": 0.16163208782672883,
|
|
"num_tokens": 5864435.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.624909925460815,
|
|
"epoch": 2.671913835956918,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004565769741199314,
|
|
"loss": 5.5135,
|
|
"mean_token_accuracy": 0.1572384186089039,
|
|
"num_tokens": 5873964.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.727962255477905,
|
|
"epoch": 2.676056338028169,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045638808050260237,
|
|
"loss": 5.4648,
|
|
"mean_token_accuracy": 0.16628578454256057,
|
|
"num_tokens": 5882738.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.587652635574341,
|
|
"epoch": 2.68019884009942,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045619882100569317,
|
|
"loss": 5.5925,
|
|
"mean_token_accuracy": 0.15854242593050002,
|
|
"num_tokens": 5891749.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.627515649795532,
|
|
"epoch": 2.684341342170671,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004560091960109608,
|
|
"loss": 5.5372,
|
|
"mean_token_accuracy": 0.15811603218317033,
|
|
"num_tokens": 5900551.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.661735343933105,
|
|
"epoch": 2.6884838442419223,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004558192059008996,
|
|
"loss": 5.5561,
|
|
"mean_token_accuracy": 0.1517203629016876,
|
|
"num_tokens": 5909254.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.599635648727417,
|
|
"epoch": 2.692626346313173,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004556288510587406,
|
|
"loss": 5.4987,
|
|
"mean_token_accuracy": 0.15817683488130568,
|
|
"num_tokens": 5918357.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.662895679473877,
|
|
"epoch": 2.696768848384424,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045543813186845004,
|
|
"loss": 5.5021,
|
|
"mean_token_accuracy": 0.15579704195261002,
|
|
"num_tokens": 5927742.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.627440690994263,
|
|
"epoch": 2.700911350455675,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004552470487147294,
|
|
"loss": 5.5823,
|
|
"mean_token_accuracy": 0.1556493915617466,
|
|
"num_tokens": 5937025.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.70678014755249,
|
|
"epoch": 2.7050538525269263,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045505560198301443,
|
|
"loss": 5.5615,
|
|
"mean_token_accuracy": 0.15135139375925064,
|
|
"num_tokens": 5946176.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.659924745559692,
|
|
"epoch": 2.7091963545981774,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045486379205947396,
|
|
"loss": 5.5233,
|
|
"mean_token_accuracy": 0.1584411159157753,
|
|
"num_tokens": 5955550.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.594695281982422,
|
|
"epoch": 2.713338856669428,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004546716193310098,
|
|
"loss": 5.4541,
|
|
"mean_token_accuracy": 0.1651945412158966,
|
|
"num_tokens": 5964467.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.605531787872314,
|
|
"epoch": 2.717481358740679,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000454479084185255,
|
|
"loss": 5.4855,
|
|
"mean_token_accuracy": 0.1514287754893303,
|
|
"num_tokens": 5973906.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.696796035766601,
|
|
"epoch": 2.7216238608119303,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004542861870105743,
|
|
"loss": 5.5345,
|
|
"mean_token_accuracy": 0.15694239139556884,
|
|
"num_tokens": 5983701.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.623150444030761,
|
|
"epoch": 2.7257663628831814,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045409292819606224,
|
|
"loss": 5.4971,
|
|
"mean_token_accuracy": 0.16068210303783417,
|
|
"num_tokens": 5993367.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.646857404708863,
|
|
"epoch": 2.7299088649544325,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045389930813154297,
|
|
"loss": 5.4519,
|
|
"mean_token_accuracy": 0.16334805637598038,
|
|
"num_tokens": 6003752.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.675233459472656,
|
|
"epoch": 2.7340513670256836,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004537053272075693,
|
|
"loss": 5.5263,
|
|
"mean_token_accuracy": 0.1621565416455269,
|
|
"num_tokens": 6012486.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.5136041164398195,
|
|
"epoch": 2.7381938690969347,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045351098581542204,
|
|
"loss": 5.4721,
|
|
"mean_token_accuracy": 0.163838130235672,
|
|
"num_tokens": 6020742.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.646917295455933,
|
|
"epoch": 2.742336371168186,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004533162843471089,
|
|
"loss": 5.5009,
|
|
"mean_token_accuracy": 0.15474170595407485,
|
|
"num_tokens": 6030248.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.632185411453247,
|
|
"epoch": 2.7464788732394365,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004531212231953641,
|
|
"loss": 5.5436,
|
|
"mean_token_accuracy": 0.15157791674137117,
|
|
"num_tokens": 6039293.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.587828540802002,
|
|
"epoch": 2.7506213753106876,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004529258027536473,
|
|
"loss": 5.4134,
|
|
"mean_token_accuracy": 0.16378402560949326,
|
|
"num_tokens": 6048081.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.673684549331665,
|
|
"epoch": 2.7547638773819387,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004527300234161428,
|
|
"loss": 5.5753,
|
|
"mean_token_accuracy": 0.15969255268573762,
|
|
"num_tokens": 6057549.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.646554040908813,
|
|
"epoch": 2.75890637945319,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004525338855777591,
|
|
"loss": 5.5589,
|
|
"mean_token_accuracy": 0.15469738617539405,
|
|
"num_tokens": 6067372.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.595069122314453,
|
|
"epoch": 2.7630488815244405,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045233738963412746,
|
|
"loss": 5.4181,
|
|
"mean_token_accuracy": 0.1717814713716507,
|
|
"num_tokens": 6075532.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.5924467086792,
|
|
"epoch": 2.7671913835956916,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045214053598160173,
|
|
"loss": 5.4439,
|
|
"mean_token_accuracy": 0.16404369622468948,
|
|
"num_tokens": 6084389.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.598109579086303,
|
|
"epoch": 2.7713338856669427,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045194332501725725,
|
|
"loss": 5.4706,
|
|
"mean_token_accuracy": 0.16428767740726472,
|
|
"num_tokens": 6093536.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.637304496765137,
|
|
"epoch": 2.775476387738194,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045174575713889014,
|
|
"loss": 5.5167,
|
|
"mean_token_accuracy": 0.16573925465345382,
|
|
"num_tokens": 6102677.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.677533674240112,
|
|
"epoch": 2.779618889809445,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004515478327450164,
|
|
"loss": 5.5926,
|
|
"mean_token_accuracy": 0.156404609978199,
|
|
"num_tokens": 6112430.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.643227863311767,
|
|
"epoch": 2.783761391880696,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045134955223487093,
|
|
"loss": 5.4889,
|
|
"mean_token_accuracy": 0.16630716919898986,
|
|
"num_tokens": 6121408.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.57972674369812,
|
|
"epoch": 2.787903893951947,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045115091600840757,
|
|
"loss": 5.5188,
|
|
"mean_token_accuracy": 0.15800442099571227,
|
|
"num_tokens": 6131339.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.624274778366089,
|
|
"epoch": 2.7920463960231983,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00045095192446629697,
|
|
"loss": 5.4816,
|
|
"mean_token_accuracy": 0.16174803227186202,
|
|
"num_tokens": 6140590.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.704814815521241,
|
|
"epoch": 2.796188898094449,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000450752578009927,
|
|
"loss": 5.5621,
|
|
"mean_token_accuracy": 0.1653067871928215,
|
|
"num_tokens": 6149958.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.58657054901123,
|
|
"epoch": 2.8003314001657,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045055287704140115,
|
|
"loss": 5.4346,
|
|
"mean_token_accuracy": 0.15949429720640182,
|
|
"num_tokens": 6159411.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.595387983322143,
|
|
"epoch": 2.804473902236951,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004503528219635381,
|
|
"loss": 5.5138,
|
|
"mean_token_accuracy": 0.1545591250061989,
|
|
"num_tokens": 6169264.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.617677021026611,
|
|
"epoch": 2.8086164043082023,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004501524131798709,
|
|
"loss": 5.4826,
|
|
"mean_token_accuracy": 0.16332829892635345,
|
|
"num_tokens": 6178374.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.7183191776275635,
|
|
"epoch": 2.812758906379453,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000449951651094646,
|
|
"loss": 5.4697,
|
|
"mean_token_accuracy": 0.15537519603967667,
|
|
"num_tokens": 6186918.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.566522169113159,
|
|
"epoch": 2.816901408450704,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044975053611282217,
|
|
"loss": 5.5357,
|
|
"mean_token_accuracy": 0.1613329365849495,
|
|
"num_tokens": 6196400.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.514105558395386,
|
|
"epoch": 2.821043910521955,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004495490686400706,
|
|
"loss": 5.3941,
|
|
"mean_token_accuracy": 0.1649619847536087,
|
|
"num_tokens": 6204849.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.615210485458374,
|
|
"epoch": 2.8251864125932062,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044934724908277313,
|
|
"loss": 5.4834,
|
|
"mean_token_accuracy": 0.16362717300653457,
|
|
"num_tokens": 6214121.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.654720067977905,
|
|
"epoch": 2.8293289146644574,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044914507784802184,
|
|
"loss": 5.4835,
|
|
"mean_token_accuracy": 0.16509974151849746,
|
|
"num_tokens": 6223163.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.641624307632446,
|
|
"epoch": 2.8334714167357085,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004489425553436182,
|
|
"loss": 5.5084,
|
|
"mean_token_accuracy": 0.15882937759160995,
|
|
"num_tokens": 6231581.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.659383153915405,
|
|
"epoch": 2.8376139188069596,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00044873968197807213,
|
|
"loss": 5.5838,
|
|
"mean_token_accuracy": 0.1599879801273346,
|
|
"num_tokens": 6240015.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.666435050964355,
|
|
"epoch": 2.8417564208782107,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044853645816060156,
|
|
"loss": 5.5979,
|
|
"mean_token_accuracy": 0.1532062940299511,
|
|
"num_tokens": 6249083.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.6627098560333256,
|
|
"epoch": 2.8458989229494613,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044833288430113104,
|
|
"loss": 5.5496,
|
|
"mean_token_accuracy": 0.1654196172952652,
|
|
"num_tokens": 6258798.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.609176588058472,
|
|
"epoch": 2.8500414250207124,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044812896081029136,
|
|
"loss": 5.4023,
|
|
"mean_token_accuracy": 0.16617698073387147,
|
|
"num_tokens": 6267819.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.626215600967408,
|
|
"epoch": 2.8541839270919636,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004479246880994183,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.15876610353589057,
|
|
"num_tokens": 6276915.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.627827167510986,
|
|
"epoch": 2.8583264291632147,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004477200665805525,
|
|
"loss": 5.5216,
|
|
"mean_token_accuracy": 0.15861138105392455,
|
|
"num_tokens": 6285880.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.6323963642120365,
|
|
"epoch": 2.8624689312344658,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044751509666643777,
|
|
"loss": 5.6351,
|
|
"mean_token_accuracy": 0.14970673844218255,
|
|
"num_tokens": 6295408.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.6166990280151365,
|
|
"epoch": 2.8666114333057164,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044730977877052084,
|
|
"loss": 5.3753,
|
|
"mean_token_accuracy": 0.16562075018882752,
|
|
"num_tokens": 6304110.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.669659185409546,
|
|
"epoch": 2.8707539353769675,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044710411330695045,
|
|
"loss": 5.5254,
|
|
"mean_token_accuracy": 0.15842530727386475,
|
|
"num_tokens": 6312577.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.738371801376343,
|
|
"epoch": 2.8748964374482187,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044689810069057615,
|
|
"loss": 5.606,
|
|
"mean_token_accuracy": 0.15023296922445298,
|
|
"num_tokens": 6321766.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.642851161956787,
|
|
"epoch": 2.8790389395194698,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004466917413369479,
|
|
"loss": 5.5731,
|
|
"mean_token_accuracy": 0.151870658993721,
|
|
"num_tokens": 6330327.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.577242517471314,
|
|
"epoch": 2.883181441590721,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004464850356623152,
|
|
"loss": 5.4926,
|
|
"mean_token_accuracy": 0.16011006087064744,
|
|
"num_tokens": 6339596.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.613158226013184,
|
|
"epoch": 2.887323943661972,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044627798408362597,
|
|
"loss": 5.3986,
|
|
"mean_token_accuracy": 0.1670537531375885,
|
|
"num_tokens": 6348080.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.613135099411011,
|
|
"epoch": 2.891466445733223,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004460705870185258,
|
|
"loss": 5.5277,
|
|
"mean_token_accuracy": 0.15435466170310974,
|
|
"num_tokens": 6357725.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.627930736541748,
|
|
"epoch": 2.8956089478044738,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004458628448853572,
|
|
"loss": 5.4831,
|
|
"mean_token_accuracy": 0.16073092967271804,
|
|
"num_tokens": 6366821.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.729347085952758,
|
|
"epoch": 2.899751449875725,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004456547581031589,
|
|
"loss": 5.5357,
|
|
"mean_token_accuracy": 0.15560299083590506,
|
|
"num_tokens": 6375641.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.899751449875725,
|
|
"eval_entropy": 5.414937326394673,
|
|
"eval_loss": 5.899942874908447,
|
|
"eval_mean_token_accuracy": 0.15183768343569629,
|
|
"eval_num_tokens": 6375641.0,
|
|
"eval_runtime": 2.9583,
|
|
"eval_samples_per_second": 1266.605,
|
|
"eval_steps_per_second": 158.537,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.548971128463745,
|
|
"epoch": 2.903893951946976,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004454463270916646,
|
|
"loss": 5.404,
|
|
"mean_token_accuracy": 0.1680065542459488,
|
|
"num_tokens": 6384596.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.719941425323486,
|
|
"epoch": 2.908036454018227,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004452375522713023,
|
|
"loss": 5.7474,
|
|
"mean_token_accuracy": 0.14030246064066887,
|
|
"num_tokens": 6395048.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.643452119827271,
|
|
"epoch": 2.912178956089478,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004450284340631938,
|
|
"loss": 5.5762,
|
|
"mean_token_accuracy": 0.15520212948322296,
|
|
"num_tokens": 6404828.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.627139854431152,
|
|
"epoch": 2.916321458160729,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044481897288915334,
|
|
"loss": 5.4626,
|
|
"mean_token_accuracy": 0.16042221039533616,
|
|
"num_tokens": 6414670.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.474701166152954,
|
|
"epoch": 2.92046396023198,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044460916917168697,
|
|
"loss": 5.3896,
|
|
"mean_token_accuracy": 0.16453512459993364,
|
|
"num_tokens": 6423446.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.5396312236785885,
|
|
"epoch": 2.924606462303231,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004443990233339918,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.1552535504102707,
|
|
"num_tokens": 6433412.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.652051544189453,
|
|
"epoch": 2.928748964374482,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004441885357999548,
|
|
"loss": 5.444,
|
|
"mean_token_accuracy": 0.1643453508615494,
|
|
"num_tokens": 6443520.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.674185514450073,
|
|
"epoch": 2.9328914664457333,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004439777069941524,
|
|
"loss": 5.5527,
|
|
"mean_token_accuracy": 0.15431721061468123,
|
|
"num_tokens": 6452933.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.550513553619385,
|
|
"epoch": 2.9370339685169844,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044376653734184935,
|
|
"loss": 5.3908,
|
|
"mean_token_accuracy": 0.16788479536771775,
|
|
"num_tokens": 6462113.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.468146800994873,
|
|
"epoch": 2.9411764705882355,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004435550272689979,
|
|
"loss": 5.4246,
|
|
"mean_token_accuracy": 0.16583772748708725,
|
|
"num_tokens": 6471411.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.662385082244873,
|
|
"epoch": 2.945318972659486,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044334317720223706,
|
|
"loss": 5.532,
|
|
"mean_token_accuracy": 0.1599032275378704,
|
|
"num_tokens": 6480682.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.678339672088623,
|
|
"epoch": 2.9494614747307373,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044313098756889154,
|
|
"loss": 5.5754,
|
|
"mean_token_accuracy": 0.15587076246738435,
|
|
"num_tokens": 6490614.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.585533094406128,
|
|
"epoch": 2.9536039768019884,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004429184587969708,
|
|
"loss": 5.4273,
|
|
"mean_token_accuracy": 0.16405707746744155,
|
|
"num_tokens": 6499079.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.668995666503906,
|
|
"epoch": 2.9577464788732395,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004427055913151689,
|
|
"loss": 5.5561,
|
|
"mean_token_accuracy": 0.14727827310562133,
|
|
"num_tokens": 6508556.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.70610499382019,
|
|
"epoch": 2.9618889809444906,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004424923855528626,
|
|
"loss": 5.5411,
|
|
"mean_token_accuracy": 0.15901169627904893,
|
|
"num_tokens": 6517934.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.614701509475708,
|
|
"epoch": 2.9660314830157413,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004422788419401114,
|
|
"loss": 5.4321,
|
|
"mean_token_accuracy": 0.1702651560306549,
|
|
"num_tokens": 6526995.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.572284936904907,
|
|
"epoch": 2.9701739850869924,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00044206496090765595,
|
|
"loss": 5.4896,
|
|
"mean_token_accuracy": 0.1573972523212433,
|
|
"num_tokens": 6535264.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.610080623626709,
|
|
"epoch": 2.9743164871582435,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00044185074288691776,
|
|
"loss": 5.3994,
|
|
"mean_token_accuracy": 0.1746336281299591,
|
|
"num_tokens": 6544498.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.582769393920898,
|
|
"epoch": 2.9784589892294946,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044163618830999794,
|
|
"loss": 5.5098,
|
|
"mean_token_accuracy": 0.16030704230070114,
|
|
"num_tokens": 6554761.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.603819417953491,
|
|
"epoch": 2.9826014913007457,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044142129760967643,
|
|
"loss": 5.5614,
|
|
"mean_token_accuracy": 0.15714355707168579,
|
|
"num_tokens": 6564039.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.679829835891724,
|
|
"epoch": 2.986743993371997,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004412060712194115,
|
|
"loss": 5.5068,
|
|
"mean_token_accuracy": 0.15993095487356185,
|
|
"num_tokens": 6572559.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.589264869689941,
|
|
"epoch": 2.990886495443248,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004409905095733381,
|
|
"loss": 5.4931,
|
|
"mean_token_accuracy": 0.1612076446413994,
|
|
"num_tokens": 6581621.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.550649166107178,
|
|
"epoch": 2.995028997514499,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00044077461310626764,
|
|
"loss": 5.4721,
|
|
"mean_token_accuracy": 0.15403861105442046,
|
|
"num_tokens": 6590353.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.64892258644104,
|
|
"epoch": 2.9991714995857497,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000440558382253687,
|
|
"loss": 5.4575,
|
|
"mean_token_accuracy": 0.16497674435377122,
|
|
"num_tokens": 6598822.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.5967165470123295,
|
|
"epoch": 3.003314001657001,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004403418174517573,
|
|
"loss": 5.2529,
|
|
"mean_token_accuracy": 0.16683556884527206,
|
|
"num_tokens": 6606728.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.599476623535156,
|
|
"epoch": 3.007456503728252,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044012491913731353,
|
|
"loss": 5.1756,
|
|
"mean_token_accuracy": 0.1743481546640396,
|
|
"num_tokens": 6616226.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.5095573425292965,
|
|
"epoch": 3.011599005799503,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004399076877478632,
|
|
"loss": 5.0266,
|
|
"mean_token_accuracy": 0.1868550017476082,
|
|
"num_tokens": 6624801.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.60879430770874,
|
|
"epoch": 3.015741507870754,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004396901237215858,
|
|
"loss": 5.2209,
|
|
"mean_token_accuracy": 0.16688485741615294,
|
|
"num_tokens": 6633043.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.538025426864624,
|
|
"epoch": 3.019884009942005,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043947222749733173,
|
|
"loss": 5.1692,
|
|
"mean_token_accuracy": 0.17499702274799347,
|
|
"num_tokens": 6641197.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.586767101287842,
|
|
"epoch": 3.024026512013256,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043925399951462153,
|
|
"loss": 5.2064,
|
|
"mean_token_accuracy": 0.1801098555326462,
|
|
"num_tokens": 6650411.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.4191618919372555,
|
|
"epoch": 3.028169014084507,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004390354402136449,
|
|
"loss": 5.0903,
|
|
"mean_token_accuracy": 0.17804521024227143,
|
|
"num_tokens": 6659405.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.583788824081421,
|
|
"epoch": 3.032311516155758,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004388165500352597,
|
|
"loss": 5.1642,
|
|
"mean_token_accuracy": 0.17557908743619918,
|
|
"num_tokens": 6667701.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.614604902267456,
|
|
"epoch": 3.036454018227009,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004385973294209915,
|
|
"loss": 5.1881,
|
|
"mean_token_accuracy": 0.17808027267456056,
|
|
"num_tokens": 6676801.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.457578706741333,
|
|
"epoch": 3.0405965202982603,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004383777788130321,
|
|
"loss": 5.083,
|
|
"mean_token_accuracy": 0.17836775481700898,
|
|
"num_tokens": 6686005.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.479088068008423,
|
|
"epoch": 3.044739022369511,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004381578986542393,
|
|
"loss": 5.0937,
|
|
"mean_token_accuracy": 0.17960153371095658,
|
|
"num_tokens": 6694884.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.501745700836182,
|
|
"epoch": 3.048881524440762,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004379376893881352,
|
|
"loss": 5.2229,
|
|
"mean_token_accuracy": 0.17827630788087845,
|
|
"num_tokens": 6703476.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.616245269775391,
|
|
"epoch": 3.053024026512013,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004377171514589059,
|
|
"loss": 5.2531,
|
|
"mean_token_accuracy": 0.1736731931567192,
|
|
"num_tokens": 6712576.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.398316955566406,
|
|
"epoch": 3.0571665285832643,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004374962853114007,
|
|
"loss": 5.0545,
|
|
"mean_token_accuracy": 0.18062082678079605,
|
|
"num_tokens": 6721532.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.480774354934693,
|
|
"epoch": 3.0613090306545154,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004372750913911306,
|
|
"loss": 5.1028,
|
|
"mean_token_accuracy": 0.18300070911645888,
|
|
"num_tokens": 6730055.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.446919918060303,
|
|
"epoch": 3.0654515327257665,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004370535701442679,
|
|
"loss": 5.1556,
|
|
"mean_token_accuracy": 0.17969694286584853,
|
|
"num_tokens": 6739310.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.54828953742981,
|
|
"epoch": 3.069594034797017,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043683172201764537,
|
|
"loss": 5.2581,
|
|
"mean_token_accuracy": 0.16716694980859756,
|
|
"num_tokens": 6748976.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.539231872558593,
|
|
"epoch": 3.0737365368682683,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004366095474587547,
|
|
"loss": 5.1044,
|
|
"mean_token_accuracy": 0.17598106861114501,
|
|
"num_tokens": 6758043.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.500854778289795,
|
|
"epoch": 3.0778790389395194,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043638704691574646,
|
|
"loss": 5.163,
|
|
"mean_token_accuracy": 0.1786232903599739,
|
|
"num_tokens": 6767554.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.560473966598511,
|
|
"epoch": 3.0820215410107705,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004361642208374283,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.1666739672422409,
|
|
"num_tokens": 6776161.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.476630401611328,
|
|
"epoch": 3.0861640430820216,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00043594106967326495,
|
|
"loss": 5.1169,
|
|
"mean_token_accuracy": 0.17555084377527236,
|
|
"num_tokens": 6785371.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.494755983352661,
|
|
"epoch": 3.0903065451532727,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00043571759387337665,
|
|
"loss": 5.1456,
|
|
"mean_token_accuracy": 0.1808715656399727,
|
|
"num_tokens": 6795242.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.591760492324829,
|
|
"epoch": 3.0944490472245234,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004354937938885385,
|
|
"loss": 5.2235,
|
|
"mean_token_accuracy": 0.17338325083255768,
|
|
"num_tokens": 6805423.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.464981889724731,
|
|
"epoch": 3.0985915492957745,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004352696701701796,
|
|
"loss": 5.1469,
|
|
"mean_token_accuracy": 0.17880967408418655,
|
|
"num_tokens": 6814377.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.448660278320313,
|
|
"epoch": 3.1027340513670256,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004350452231703819,
|
|
"loss": 5.2118,
|
|
"mean_token_accuracy": 0.18132367432117463,
|
|
"num_tokens": 6824025.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.550149536132812,
|
|
"epoch": 3.1068765534382767,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043482045334187954,
|
|
"loss": 5.1273,
|
|
"mean_token_accuracy": 0.18649032860994338,
|
|
"num_tokens": 6833071.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.442513799667358,
|
|
"epoch": 3.111019055509528,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043459536113805785,
|
|
"loss": 5.1076,
|
|
"mean_token_accuracy": 0.17443138360977173,
|
|
"num_tokens": 6841983.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.447629022598266,
|
|
"epoch": 3.115161557580779,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00043436994701295237,
|
|
"loss": 5.1043,
|
|
"mean_token_accuracy": 0.17914267033338546,
|
|
"num_tokens": 6851436.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.461205101013183,
|
|
"epoch": 3.1193040596520296,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043414421142124804,
|
|
"loss": 5.1836,
|
|
"mean_token_accuracy": 0.1672696813941002,
|
|
"num_tokens": 6861810.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.478553342819214,
|
|
"epoch": 3.1234465617232807,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043391815481827814,
|
|
"loss": 5.0639,
|
|
"mean_token_accuracy": 0.18850475996732713,
|
|
"num_tokens": 6870418.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.41155161857605,
|
|
"epoch": 3.127589063794532,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004336917776600237,
|
|
"loss": 5.0683,
|
|
"mean_token_accuracy": 0.18221812546253205,
|
|
"num_tokens": 6879419.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.435874891281128,
|
|
"epoch": 3.131731565865783,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004334650804031122,
|
|
"loss": 5.1852,
|
|
"mean_token_accuracy": 0.17172302156686783,
|
|
"num_tokens": 6888768.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.507850646972656,
|
|
"epoch": 3.135874067937034,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004332380635048165,
|
|
"loss": 5.1296,
|
|
"mean_token_accuracy": 0.17900288552045823,
|
|
"num_tokens": 6897133.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.485491752624512,
|
|
"epoch": 3.140016570008285,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00043301072742305474,
|
|
"loss": 5.1773,
|
|
"mean_token_accuracy": 0.17168077528476716,
|
|
"num_tokens": 6904976.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.537312364578247,
|
|
"epoch": 3.1441590720795363,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004327830726163886,
|
|
"loss": 5.1738,
|
|
"mean_token_accuracy": 0.17883574068546296,
|
|
"num_tokens": 6915138.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.548703718185425,
|
|
"epoch": 3.148301574150787,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004325550995440226,
|
|
"loss": 5.1395,
|
|
"mean_token_accuracy": 0.1752033144235611,
|
|
"num_tokens": 6923962.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.386642169952393,
|
|
"epoch": 3.152444076222038,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00043232680866580356,
|
|
"loss": 5.1478,
|
|
"mean_token_accuracy": 0.17952724993228913,
|
|
"num_tokens": 6932102.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.464086627960205,
|
|
"epoch": 3.156586578293289,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00043209820044221893,
|
|
"loss": 5.153,
|
|
"mean_token_accuracy": 0.1785269558429718,
|
|
"num_tokens": 6940354.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.4963922023773195,
|
|
"epoch": 3.1607290803645403,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043186927533439654,
|
|
"loss": 5.1677,
|
|
"mean_token_accuracy": 0.1716633692383766,
|
|
"num_tokens": 6949779.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.536400413513183,
|
|
"epoch": 3.1648715824357914,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043164003380410335,
|
|
"loss": 5.2273,
|
|
"mean_token_accuracy": 0.17445194274187087,
|
|
"num_tokens": 6958760.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.486375331878662,
|
|
"epoch": 3.169014084507042,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004314104763137445,
|
|
"loss": 5.1037,
|
|
"mean_token_accuracy": 0.18134674280881882,
|
|
"num_tokens": 6967535.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.469356393814087,
|
|
"epoch": 3.173156586578293,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043118060332636264,
|
|
"loss": 5.2036,
|
|
"mean_token_accuracy": 0.17774708271026612,
|
|
"num_tokens": 6976211.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.576221227645874,
|
|
"epoch": 3.1772990886495442,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043095041530563665,
|
|
"loss": 5.2058,
|
|
"mean_token_accuracy": 0.17325948774814606,
|
|
"num_tokens": 6986001.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.420689487457276,
|
|
"epoch": 3.1814415907207954,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043071991271588104,
|
|
"loss": 5.2006,
|
|
"mean_token_accuracy": 0.18302187621593474,
|
|
"num_tokens": 6994568.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.52434401512146,
|
|
"epoch": 3.1855840927920465,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004304890960220446,
|
|
"loss": 5.2921,
|
|
"mean_token_accuracy": 0.16869494169950486,
|
|
"num_tokens": 7004744.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.520354557037353,
|
|
"epoch": 3.1897265948632976,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043025796568970975,
|
|
"loss": 5.2771,
|
|
"mean_token_accuracy": 0.1696591019630432,
|
|
"num_tokens": 7014315.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.523347425460815,
|
|
"epoch": 3.1938690969345487,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043002652218509184,
|
|
"loss": 5.1172,
|
|
"mean_token_accuracy": 0.18450311720371246,
|
|
"num_tokens": 7022258.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.522213459014893,
|
|
"epoch": 3.1980115990057993,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004297947659750376,
|
|
"loss": 5.2587,
|
|
"mean_token_accuracy": 0.1674624815583229,
|
|
"num_tokens": 7031732.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.467557621002197,
|
|
"epoch": 3.2021541010770505,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004295626975270246,
|
|
"loss": 5.1818,
|
|
"mean_token_accuracy": 0.1766605108976364,
|
|
"num_tokens": 7040478.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.490935039520264,
|
|
"epoch": 3.2062966031483016,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042933031730916024,
|
|
"loss": 5.1395,
|
|
"mean_token_accuracy": 0.18396904766559602,
|
|
"num_tokens": 7049360.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.5688018798828125,
|
|
"epoch": 3.2104391052195527,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000429097625790181,
|
|
"loss": 5.2267,
|
|
"mean_token_accuracy": 0.17498258501291275,
|
|
"num_tokens": 7059271.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.398471450805664,
|
|
"epoch": 3.2145816072908038,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004288646234394508,
|
|
"loss": 5.2177,
|
|
"mean_token_accuracy": 0.17296054661273957,
|
|
"num_tokens": 7069038.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.5407984256744385,
|
|
"epoch": 3.218724109362055,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042863131072696115,
|
|
"loss": 5.1617,
|
|
"mean_token_accuracy": 0.17704827934503556,
|
|
"num_tokens": 7077971.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.4559561252594,
|
|
"epoch": 3.2228666114333056,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00042839768812332905,
|
|
"loss": 5.1612,
|
|
"mean_token_accuracy": 0.17406544089317322,
|
|
"num_tokens": 7086821.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.419722366333008,
|
|
"epoch": 3.2270091135045567,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000428163756099797,
|
|
"loss": 5.1859,
|
|
"mean_token_accuracy": 0.17225482910871506,
|
|
"num_tokens": 7095407.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.48249921798706,
|
|
"epoch": 3.2311516155758078,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004279295151282314,
|
|
"loss": 5.1116,
|
|
"mean_token_accuracy": 0.18240531533956528,
|
|
"num_tokens": 7103964.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.564240264892578,
|
|
"epoch": 3.235294117647059,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004276949656811217,
|
|
"loss": 5.251,
|
|
"mean_token_accuracy": 0.17079487890005113,
|
|
"num_tokens": 7113571.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.522033739089966,
|
|
"epoch": 3.23943661971831,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00042746010823158006,
|
|
"loss": 5.2738,
|
|
"mean_token_accuracy": 0.17067040055990218,
|
|
"num_tokens": 7122844.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.4324219703674315,
|
|
"epoch": 3.243579121789561,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042722494325333945,
|
|
"loss": 5.1326,
|
|
"mean_token_accuracy": 0.176889306306839,
|
|
"num_tokens": 7132123.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.458636379241943,
|
|
"epoch": 3.2477216238608118,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004269894712207533,
|
|
"loss": 5.1819,
|
|
"mean_token_accuracy": 0.18051923513412477,
|
|
"num_tokens": 7141068.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.43212571144104,
|
|
"epoch": 3.251864125932063,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004267536926087947,
|
|
"loss": 5.1925,
|
|
"mean_token_accuracy": 0.1708641156554222,
|
|
"num_tokens": 7150829.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.458031702041626,
|
|
"epoch": 3.256006628003314,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004265176078930545,
|
|
"loss": 5.1703,
|
|
"mean_token_accuracy": 0.17345706522464752,
|
|
"num_tokens": 7160814.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.498150825500488,
|
|
"epoch": 3.260149130074565,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00042628121754974166,
|
|
"loss": 5.2634,
|
|
"mean_token_accuracy": 0.17203340232372283,
|
|
"num_tokens": 7170085.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.590339422225952,
|
|
"epoch": 3.264291632145816,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004260445220556811,
|
|
"loss": 5.2102,
|
|
"mean_token_accuracy": 0.1736685037612915,
|
|
"num_tokens": 7179366.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.52174072265625,
|
|
"epoch": 3.2684341342170673,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00042580752188831376,
|
|
"loss": 5.2819,
|
|
"mean_token_accuracy": 0.17067586481571198,
|
|
"num_tokens": 7188342.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.556540441513062,
|
|
"epoch": 3.272576636288318,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004255702175256946,
|
|
"loss": 5.3095,
|
|
"mean_token_accuracy": 0.17531315684318544,
|
|
"num_tokens": 7197466.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.42687840461731,
|
|
"epoch": 3.276719138359569,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00042533260944649273,
|
|
"loss": 5.1579,
|
|
"mean_token_accuracy": 0.17852312475442886,
|
|
"num_tokens": 7205462.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.424286890029907,
|
|
"epoch": 3.28086164043082,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004250946981299893,
|
|
"loss": 5.1838,
|
|
"mean_token_accuracy": 0.17682615667581558,
|
|
"num_tokens": 7215115.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.459038209915161,
|
|
"epoch": 3.2850041425020713,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042485648405607776,
|
|
"loss": 5.1379,
|
|
"mean_token_accuracy": 0.17029522657394408,
|
|
"num_tokens": 7224909.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.392039203643799,
|
|
"epoch": 3.2891466445733224,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004246179677052617,
|
|
"loss": 5.0642,
|
|
"mean_token_accuracy": 0.18698266595602037,
|
|
"num_tokens": 7232946.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.362467432022095,
|
|
"epoch": 3.2932891466445735,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004243791495586547,
|
|
"loss": 5.1736,
|
|
"mean_token_accuracy": 0.17996304631233215,
|
|
"num_tokens": 7241104.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.567073392868042,
|
|
"epoch": 3.297431648715824,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004241400300979791,
|
|
"loss": 5.2257,
|
|
"mean_token_accuracy": 0.1786801502108574,
|
|
"num_tokens": 7250749.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.447740983963013,
|
|
"epoch": 3.3015741507870753,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004239006098055651,
|
|
"loss": 5.124,
|
|
"mean_token_accuracy": 0.1787547677755356,
|
|
"num_tokens": 7258687.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.41381049156189,
|
|
"epoch": 3.3057166528583264,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004236608891643494,
|
|
"loss": 5.1572,
|
|
"mean_token_accuracy": 0.17702401280403138,
|
|
"num_tokens": 7267316.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.448715782165527,
|
|
"epoch": 3.3098591549295775,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004234208686578748,
|
|
"loss": 5.1777,
|
|
"mean_token_accuracy": 0.17505329996347427,
|
|
"num_tokens": 7276734.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.440592193603516,
|
|
"epoch": 3.3140016570008286,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00042318054877028876,
|
|
"loss": 5.1579,
|
|
"mean_token_accuracy": 0.17840310037136078,
|
|
"num_tokens": 7285916.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.3140016570008286,
|
|
"eval_entropy": 5.363256025416002,
|
|
"eval_loss": 5.893955707550049,
|
|
"eval_mean_token_accuracy": 0.15349840453819935,
|
|
"eval_num_tokens": 7285916.0,
|
|
"eval_runtime": 2.7471,
|
|
"eval_samples_per_second": 1363.961,
|
|
"eval_steps_per_second": 170.723,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.487448406219483,
|
|
"epoch": 3.3181441590720797,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042293992998634314,
|
|
"loss": 5.1667,
|
|
"mean_token_accuracy": 0.17779863625764847,
|
|
"num_tokens": 7294184.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.414953899383545,
|
|
"epoch": 3.3222866611433304,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004226990127913918,
|
|
"loss": 5.177,
|
|
"mean_token_accuracy": 0.1768763244152069,
|
|
"num_tokens": 7302647.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.442982530593872,
|
|
"epoch": 3.3264291632145815,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004224577976713915,
|
|
"loss": 5.2506,
|
|
"mean_token_accuracy": 0.16832198798656464,
|
|
"num_tokens": 7311455.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.609448575973511,
|
|
"epoch": 3.3305716652858326,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042221628511289927,
|
|
"loss": 5.3136,
|
|
"mean_token_accuracy": 0.16625809520483018,
|
|
"num_tokens": 7320978.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.612923622131348,
|
|
"epoch": 3.3347141673570837,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004219744756030723,
|
|
"loss": 5.3273,
|
|
"mean_token_accuracy": 0.16872513592243193,
|
|
"num_tokens": 7330202.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.4015247344970705,
|
|
"epoch": 3.338856669428335,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00042173236962966706,
|
|
"loss": 5.1649,
|
|
"mean_token_accuracy": 0.17969190180301667,
|
|
"num_tokens": 7339121.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.492907905578614,
|
|
"epoch": 3.342999171499586,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004214899676810375,
|
|
"loss": 5.2251,
|
|
"mean_token_accuracy": 0.1665240079164505,
|
|
"num_tokens": 7348631.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.469464063644409,
|
|
"epoch": 3.347141673570837,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004212472702461348,
|
|
"loss": 5.1843,
|
|
"mean_token_accuracy": 0.1820875510573387,
|
|
"num_tokens": 7358354.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.454071807861328,
|
|
"epoch": 3.3512841756420877,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004210042778145064,
|
|
"loss": 5.2038,
|
|
"mean_token_accuracy": 0.17285584658384323,
|
|
"num_tokens": 7367031.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.436822319030762,
|
|
"epoch": 3.355426677713339,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00042076099087629443,
|
|
"loss": 5.2025,
|
|
"mean_token_accuracy": 0.1784903883934021,
|
|
"num_tokens": 7375996.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.552537536621093,
|
|
"epoch": 3.35956917978459,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00042051740992223534,
|
|
"loss": 5.2567,
|
|
"mean_token_accuracy": 0.16439318358898164,
|
|
"num_tokens": 7385355.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.402252960205078,
|
|
"epoch": 3.363711681855841,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004202735354436586,
|
|
"loss": 5.0999,
|
|
"mean_token_accuracy": 0.1792566418647766,
|
|
"num_tokens": 7394384.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.511481237411499,
|
|
"epoch": 3.367854183927092,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00042002936793248546,
|
|
"loss": 5.2458,
|
|
"mean_token_accuracy": 0.1770971804857254,
|
|
"num_tokens": 7403141.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.508409786224365,
|
|
"epoch": 3.371996685998343,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004197849078812286,
|
|
"loss": 5.199,
|
|
"mean_token_accuracy": 0.1739865079522133,
|
|
"num_tokens": 7413331.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.471702003479004,
|
|
"epoch": 3.376139188069594,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004195401557829907,
|
|
"loss": 5.2368,
|
|
"mean_token_accuracy": 0.17052597552537918,
|
|
"num_tokens": 7422362.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.4563977241516115,
|
|
"epoch": 3.380281690140845,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00041929511213146343,
|
|
"loss": 5.1661,
|
|
"mean_token_accuracy": 0.18033864200115204,
|
|
"num_tokens": 7431350.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.4815953254699705,
|
|
"epoch": 3.384424192212096,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004190497774209266,
|
|
"loss": 5.1532,
|
|
"mean_token_accuracy": 0.1800685152411461,
|
|
"num_tokens": 7440382.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.51438536643982,
|
|
"epoch": 3.3885666942833472,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041880415214624715,
|
|
"loss": 5.2355,
|
|
"mean_token_accuracy": 0.17847414761781694,
|
|
"num_tokens": 7449357.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.463125705718994,
|
|
"epoch": 3.3927091963545983,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004185582368028781,
|
|
"loss": 5.1468,
|
|
"mean_token_accuracy": 0.1853232800960541,
|
|
"num_tokens": 7457544.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.409591102600098,
|
|
"epoch": 3.3968516984258494,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004183120318868577,
|
|
"loss": 5.1616,
|
|
"mean_token_accuracy": 0.18774343580007552,
|
|
"num_tokens": 7465915.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.435869741439819,
|
|
"epoch": 3.4009942004971,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00041806553789480787,
|
|
"loss": 5.1161,
|
|
"mean_token_accuracy": 0.1833360970020294,
|
|
"num_tokens": 7474551.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.507905626296997,
|
|
"epoch": 3.405136702568351,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00041781875532393423,
|
|
"loss": 5.2287,
|
|
"mean_token_accuracy": 0.17473389357328414,
|
|
"num_tokens": 7483271.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.474280261993409,
|
|
"epoch": 3.4092792046396023,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00041757168467202404,
|
|
"loss": 5.2371,
|
|
"mean_token_accuracy": 0.17051266878843307,
|
|
"num_tokens": 7493179.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.47375521659851,
|
|
"epoch": 3.4134217067108534,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00041732432643744576,
|
|
"loss": 5.2132,
|
|
"mean_token_accuracy": 0.17562344074249267,
|
|
"num_tokens": 7501858.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.562008953094482,
|
|
"epoch": 3.4175642087821045,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004170766811191481,
|
|
"loss": 5.282,
|
|
"mean_token_accuracy": 0.16895801872014998,
|
|
"num_tokens": 7511817.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.3949895858764645,
|
|
"epoch": 3.421706710853355,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004168287492166587,
|
|
"loss": 5.2513,
|
|
"mean_token_accuracy": 0.175115604698658,
|
|
"num_tokens": 7521655.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.478516006469727,
|
|
"epoch": 3.4258492129246063,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00041658053123008325,
|
|
"loss": 5.2289,
|
|
"mean_token_accuracy": 0.17174260020256044,
|
|
"num_tokens": 7531321.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.469200515747071,
|
|
"epoch": 3.4299917149958574,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004163320276601046,
|
|
"loss": 5.2104,
|
|
"mean_token_accuracy": 0.17538849860429764,
|
|
"num_tokens": 7540613.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.478423690795898,
|
|
"epoch": 3.4341342170671085,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004160832390079816,
|
|
"loss": 5.2823,
|
|
"mean_token_accuracy": 0.17514234632253647,
|
|
"num_tokens": 7549598.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.449503993988037,
|
|
"epoch": 3.4382767191383596,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041583416577554815,
|
|
"loss": 5.1509,
|
|
"mean_token_accuracy": 0.18169643133878707,
|
|
"num_tokens": 7558738.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.454873323440552,
|
|
"epoch": 3.4424192212096107,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00041558480846521213,
|
|
"loss": 5.1996,
|
|
"mean_token_accuracy": 0.17627429366111755,
|
|
"num_tokens": 7566873.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.413073778152466,
|
|
"epoch": 3.446561723280862,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041533516757995455,
|
|
"loss": 5.18,
|
|
"mean_token_accuracy": 0.1794169783592224,
|
|
"num_tokens": 7576645.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.449489498138428,
|
|
"epoch": 3.4507042253521125,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004150852436233283,
|
|
"loss": 5.1844,
|
|
"mean_token_accuracy": 0.17338148206472398,
|
|
"num_tokens": 7584854.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.500524044036865,
|
|
"epoch": 3.4548467274233636,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004148350370994574,
|
|
"loss": 5.3142,
|
|
"mean_token_accuracy": 0.16781232208013536,
|
|
"num_tokens": 7594022.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.458495187759399,
|
|
"epoch": 3.4589892294946147,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004145845485130357,
|
|
"loss": 5.1757,
|
|
"mean_token_accuracy": 0.17461147904396057,
|
|
"num_tokens": 7603259.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.453242301940918,
|
|
"epoch": 3.463131731565866,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041433377836932594,
|
|
"loss": 5.2853,
|
|
"mean_token_accuracy": 0.1680028423666954,
|
|
"num_tokens": 7613451.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.533868455886841,
|
|
"epoch": 3.467274233637117,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00041408272717415913,
|
|
"loss": 5.245,
|
|
"mean_token_accuracy": 0.17225596755743028,
|
|
"num_tokens": 7622218.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.351213121414185,
|
|
"epoch": 3.4714167357083676,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004138313954339327,
|
|
"loss": 5.104,
|
|
"mean_token_accuracy": 0.17463669180870056,
|
|
"num_tokens": 7630508.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.488496780395508,
|
|
"epoch": 3.4755592377796187,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004135797836556106,
|
|
"loss": 5.1982,
|
|
"mean_token_accuracy": 0.181541608273983,
|
|
"num_tokens": 7640141.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.459699392318726,
|
|
"epoch": 3.47970173985087,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041332789234672094,
|
|
"loss": 5.193,
|
|
"mean_token_accuracy": 0.1721531167626381,
|
|
"num_tokens": 7648764.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.406872129440307,
|
|
"epoch": 3.483844241922121,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004130757220153561,
|
|
"loss": 5.1952,
|
|
"mean_token_accuracy": 0.18173792064189911,
|
|
"num_tokens": 7658642.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.544135618209839,
|
|
"epoch": 3.487986743993372,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004128232731701712,
|
|
"loss": 5.377,
|
|
"mean_token_accuracy": 0.16519460082054138,
|
|
"num_tokens": 7667524.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.488169002532959,
|
|
"epoch": 3.492129246064623,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004125705463203833,
|
|
"loss": 5.1928,
|
|
"mean_token_accuracy": 0.18297115415334703,
|
|
"num_tokens": 7675800.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.490040969848633,
|
|
"epoch": 3.4962717481358743,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00041231754197576986,
|
|
"loss": 5.2252,
|
|
"mean_token_accuracy": 0.1757526621222496,
|
|
"num_tokens": 7683742.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.468280076980591,
|
|
"epoch": 3.5004142502071254,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004120642606466685,
|
|
"loss": 5.2278,
|
|
"mean_token_accuracy": 0.1796238511800766,
|
|
"num_tokens": 7692379.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.494374942779541,
|
|
"epoch": 3.504556752278376,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004118107028439751,
|
|
"loss": 5.2138,
|
|
"mean_token_accuracy": 0.17588699012994766,
|
|
"num_tokens": 7702156.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.467664003372192,
|
|
"epoch": 3.508699254349627,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041155686907914365,
|
|
"loss": 5.2275,
|
|
"mean_token_accuracy": 0.1754150867462158,
|
|
"num_tokens": 7712135.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.451332426071167,
|
|
"epoch": 3.5128417564208783,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004113027598641845,
|
|
"loss": 5.1785,
|
|
"mean_token_accuracy": 0.1752738744020462,
|
|
"num_tokens": 7721347.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.471200275421142,
|
|
"epoch": 3.5169842584921294,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004110483757116637,
|
|
"loss": 5.2058,
|
|
"mean_token_accuracy": 0.1789921909570694,
|
|
"num_tokens": 7731351.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.433224248886108,
|
|
"epoch": 3.52112676056338,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041079371713470195,
|
|
"loss": 5.0996,
|
|
"mean_token_accuracy": 0.18778069019317628,
|
|
"num_tokens": 7740413.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.416608476638794,
|
|
"epoch": 3.525269262634631,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00041053878464697343,
|
|
"loss": 5.1977,
|
|
"mean_token_accuracy": 0.18009850829839708,
|
|
"num_tokens": 7749319.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.519042015075684,
|
|
"epoch": 3.5294117647058822,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004102835787627047,
|
|
"loss": 5.2446,
|
|
"mean_token_accuracy": 0.17312991321086885,
|
|
"num_tokens": 7757936.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.542943048477173,
|
|
"epoch": 3.5335542667771334,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004100280999966741,
|
|
"loss": 5.3083,
|
|
"mean_token_accuracy": 0.17331433892250062,
|
|
"num_tokens": 7767349.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.4042651653289795,
|
|
"epoch": 3.5376967688483845,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040977234886421007,
|
|
"loss": 5.1133,
|
|
"mean_token_accuracy": 0.17731781899929047,
|
|
"num_tokens": 7776448.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.4222177028656,
|
|
"epoch": 3.5418392709196356,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040951632588119064,
|
|
"loss": 5.209,
|
|
"mean_token_accuracy": 0.17401482313871383,
|
|
"num_tokens": 7785768.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.600345802307129,
|
|
"epoch": 3.5459817729908867,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004092600315640422,
|
|
"loss": 5.2999,
|
|
"mean_token_accuracy": 0.1715576618909836,
|
|
"num_tokens": 7794803.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.411063814163208,
|
|
"epoch": 3.550124275062138,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004090034664297382,
|
|
"loss": 5.157,
|
|
"mean_token_accuracy": 0.17963247746229172,
|
|
"num_tokens": 7803533.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.421855831146241,
|
|
"epoch": 3.5542667771333885,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00040874663099579883,
|
|
"loss": 5.2246,
|
|
"mean_token_accuracy": 0.1706953912973404,
|
|
"num_tokens": 7812854.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.49954047203064,
|
|
"epoch": 3.5584092792046396,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040848952578028915,
|
|
"loss": 5.1628,
|
|
"mean_token_accuracy": 0.174958436191082,
|
|
"num_tokens": 7821683.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.501424264907837,
|
|
"epoch": 3.5625517812758907,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004082321513018183,
|
|
"loss": 5.1738,
|
|
"mean_token_accuracy": 0.18335361480712892,
|
|
"num_tokens": 7830967.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.48747296333313,
|
|
"epoch": 3.566694283347142,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00040797450807953897,
|
|
"loss": 5.2159,
|
|
"mean_token_accuracy": 0.17444179207086563,
|
|
"num_tokens": 7841179.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.489819192886353,
|
|
"epoch": 3.5708367854183924,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004077165966331454,
|
|
"loss": 5.1587,
|
|
"mean_token_accuracy": 0.1852751389145851,
|
|
"num_tokens": 7849841.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.301797342300415,
|
|
"epoch": 3.5749792874896436,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004074584174828735,
|
|
"loss": 5.2184,
|
|
"mean_token_accuracy": 0.1747626096010208,
|
|
"num_tokens": 7859390.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.493851947784424,
|
|
"epoch": 3.5791217895608947,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00040719997114949856,
|
|
"loss": 5.1913,
|
|
"mean_token_accuracy": 0.17565421015024185,
|
|
"num_tokens": 7868848.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.506833124160766,
|
|
"epoch": 3.5832642916321458,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00040694125815433533,
|
|
"loss": 5.2498,
|
|
"mean_token_accuracy": 0.18058759868144988,
|
|
"num_tokens": 7877521.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.488465976715088,
|
|
"epoch": 3.587406793703397,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004066822790192359,
|
|
"loss": 5.2169,
|
|
"mean_token_accuracy": 0.17181889414787294,
|
|
"num_tokens": 7886355.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.437413024902344,
|
|
"epoch": 3.591549295774648,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00040642303426658967,
|
|
"loss": 5.2518,
|
|
"mean_token_accuracy": 0.17224401384592056,
|
|
"num_tokens": 7895028.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.379445266723633,
|
|
"epoch": 3.595691797845899,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040616352441932163,
|
|
"loss": 5.1282,
|
|
"mean_token_accuracy": 0.17695483118295668,
|
|
"num_tokens": 7904192.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.439553546905517,
|
|
"epoch": 3.59983429991715,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00040590375000089146,
|
|
"loss": 5.2066,
|
|
"mean_token_accuracy": 0.17032651007175445,
|
|
"num_tokens": 7912328.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.5121315002441404,
|
|
"epoch": 3.603976801988401,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040564371153529256,
|
|
"loss": 5.2517,
|
|
"mean_token_accuracy": 0.1720314145088196,
|
|
"num_tokens": 7922594.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.450373315811158,
|
|
"epoch": 3.608119304059652,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004053834095470509,
|
|
"loss": 5.2201,
|
|
"mean_token_accuracy": 0.16222715824842454,
|
|
"num_tokens": 7932345.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.505369853973389,
|
|
"epoch": 3.612261806130903,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040512284456122407,
|
|
"loss": 5.2236,
|
|
"mean_token_accuracy": 0.17660728394985198,
|
|
"num_tokens": 7942800.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.410261583328247,
|
|
"epoch": 3.616404308202154,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00040486201710340016,
|
|
"loss": 5.1287,
|
|
"mean_token_accuracy": 0.1859680861234665,
|
|
"num_tokens": 7951094.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.429905796051026,
|
|
"epoch": 3.6205468102734053,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040460092769969654,
|
|
"loss": 5.1947,
|
|
"mean_token_accuracy": 0.17089750617742538,
|
|
"num_tokens": 7960652.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.518308162689209,
|
|
"epoch": 3.624689312344656,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004043395768767592,
|
|
"loss": 5.2977,
|
|
"mean_token_accuracy": 0.16983983814716339,
|
|
"num_tokens": 7969350.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.475090789794922,
|
|
"epoch": 3.628831814415907,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004040779651617612,
|
|
"loss": 5.1831,
|
|
"mean_token_accuracy": 0.1762356474995613,
|
|
"num_tokens": 7978969.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.422111892700196,
|
|
"epoch": 3.632974316487158,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040381609308240207,
|
|
"loss": 5.2118,
|
|
"mean_token_accuracy": 0.1777448311448097,
|
|
"num_tokens": 7988531.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.423072242736817,
|
|
"epoch": 3.6371168185584093,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00040355396116690634,
|
|
"loss": 5.1864,
|
|
"mean_token_accuracy": 0.17186058908700944,
|
|
"num_tokens": 7996576.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.435235500335693,
|
|
"epoch": 3.6412593206296604,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004032915699440227,
|
|
"loss": 5.2216,
|
|
"mean_token_accuracy": 0.17088689655065536,
|
|
"num_tokens": 8006279.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.496079921722412,
|
|
"epoch": 3.6454018227009115,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000403028919943023,
|
|
"loss": 5.2049,
|
|
"mean_token_accuracy": 0.17314011603593826,
|
|
"num_tokens": 8015476.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.474205350875854,
|
|
"epoch": 3.6495443247721626,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004027660116937009,
|
|
"loss": 5.2326,
|
|
"mean_token_accuracy": 0.17077937126159667,
|
|
"num_tokens": 8025450.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.4402251720428465,
|
|
"epoch": 3.6536868268434133,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004025028457263712,
|
|
"loss": 5.2238,
|
|
"mean_token_accuracy": 0.17810273319482803,
|
|
"num_tokens": 8034850.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.3980237483978275,
|
|
"epoch": 3.6578293289146644,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004022394225718684,
|
|
"loss": 5.1709,
|
|
"mean_token_accuracy": 0.17368220686912536,
|
|
"num_tokens": 8044526.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.446775197982788,
|
|
"epoch": 3.6619718309859155,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004019757427615456,
|
|
"loss": 5.1341,
|
|
"mean_token_accuracy": 0.17317759096622468,
|
|
"num_tokens": 8054033.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.478370094299317,
|
|
"epoch": 3.6661143330571666,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004017118068272739,
|
|
"loss": 5.2298,
|
|
"mean_token_accuracy": 0.17926038503646852,
|
|
"num_tokens": 8063535.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.495908880233765,
|
|
"epoch": 3.6702568351284177,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040144761530144096,
|
|
"loss": 5.2181,
|
|
"mean_token_accuracy": 0.18870080262422562,
|
|
"num_tokens": 8073079.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.367459678649903,
|
|
"epoch": 3.6743993371996684,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040118316871695,
|
|
"loss": 5.1505,
|
|
"mean_token_accuracy": 0.18230827152729034,
|
|
"num_tokens": 8082925.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.457099437713623,
|
|
"epoch": 3.6785418392709195,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00040091846760721857,
|
|
"loss": 5.2466,
|
|
"mean_token_accuracy": 0.17278454601764678,
|
|
"num_tokens": 8092038.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.460009002685547,
|
|
"epoch": 3.6826843413421706,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040065351250617774,
|
|
"loss": 5.2137,
|
|
"mean_token_accuracy": 0.1679331600666046,
|
|
"num_tokens": 8101581.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.392482471466065,
|
|
"epoch": 3.6868268434134217,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040038830394827107,
|
|
"loss": 5.2771,
|
|
"mean_token_accuracy": 0.17032679468393325,
|
|
"num_tokens": 8111392.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.5369532108306885,
|
|
"epoch": 3.690969345484673,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000400122842468453,
|
|
"loss": 5.2742,
|
|
"mean_token_accuracy": 0.16916199624538422,
|
|
"num_tokens": 8120941.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.403111267089844,
|
|
"epoch": 3.695111847555924,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003998571286021884,
|
|
"loss": 5.1822,
|
|
"mean_token_accuracy": 0.1721450611948967,
|
|
"num_tokens": 8129983.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.482623481750489,
|
|
"epoch": 3.699254349627175,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003995911628854512,
|
|
"loss": 5.3064,
|
|
"mean_token_accuracy": 0.1740827202796936,
|
|
"num_tokens": 8139428.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.509563970565796,
|
|
"epoch": 3.7033968516984257,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003993249458547234,
|
|
"loss": 5.2302,
|
|
"mean_token_accuracy": 0.16980735808610917,
|
|
"num_tokens": 8148520.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.3746521949768065,
|
|
"epoch": 3.707539353769677,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039905847804699363,
|
|
"loss": 5.0676,
|
|
"mean_token_accuracy": 0.18766331672668457,
|
|
"num_tokens": 8156908.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.446433591842651,
|
|
"epoch": 3.711681855840928,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039879175999975677,
|
|
"loss": 5.2156,
|
|
"mean_token_accuracy": 0.18001207262277602,
|
|
"num_tokens": 8165677.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.386814975738526,
|
|
"epoch": 3.715824357912179,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00039852479225101217,
|
|
"loss": 5.1883,
|
|
"mean_token_accuracy": 0.17356531620025634,
|
|
"num_tokens": 8174786.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.454575872421264,
|
|
"epoch": 3.71996685998343,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00039825757533926305,
|
|
"loss": 5.2484,
|
|
"mean_token_accuracy": 0.16340016275644303,
|
|
"num_tokens": 8184150.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.4400050163269045,
|
|
"epoch": 3.724109362054681,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00039799010980351494,
|
|
"loss": 5.1455,
|
|
"mean_token_accuracy": 0.1879016414284706,
|
|
"num_tokens": 8192257.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.452324247360229,
|
|
"epoch": 3.728251864125932,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00039772239618327535,
|
|
"loss": 5.2273,
|
|
"mean_token_accuracy": 0.1748780906200409,
|
|
"num_tokens": 8201041.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.728251864125932,
|
|
"eval_entropy": 5.297962182620441,
|
|
"eval_loss": 5.854063987731934,
|
|
"eval_mean_token_accuracy": 0.15576711011085428,
|
|
"eval_num_tokens": 8201041.0,
|
|
"eval_runtime": 2.7449,
|
|
"eval_samples_per_second": 1365.057,
|
|
"eval_steps_per_second": 170.86,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.430630874633789,
|
|
"epoch": 3.732394366197183,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00039745443501855156,
|
|
"loss": 5.1896,
|
|
"mean_token_accuracy": 0.18020173758268357,
|
|
"num_tokens": 8209881.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.426507759094238,
|
|
"epoch": 3.736536868268434,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039718622684985075,
|
|
"loss": 5.2198,
|
|
"mean_token_accuracy": 0.1857784554362297,
|
|
"num_tokens": 8218504.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.427436780929566,
|
|
"epoch": 3.7406793703396852,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000396917772218178,
|
|
"loss": 5.0397,
|
|
"mean_token_accuracy": 0.19483556151390075,
|
|
"num_tokens": 8226785.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.463036251068115,
|
|
"epoch": 3.7448218724109363,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003966490716650357,
|
|
"loss": 5.1879,
|
|
"mean_token_accuracy": 0.1808652698993683,
|
|
"num_tokens": 8236092.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.398115777969361,
|
|
"epoch": 3.7489643744821874,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00039638012573242225,
|
|
"loss": 5.1896,
|
|
"mean_token_accuracy": 0.18426763266324997,
|
|
"num_tokens": 8245297.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.498414611816406,
|
|
"epoch": 3.7531068765534386,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003961109349628309,
|
|
"loss": 5.2544,
|
|
"mean_token_accuracy": 0.16994525790214537,
|
|
"num_tokens": 8254772.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.488787031173706,
|
|
"epoch": 3.757249378624689,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000395841499899249,
|
|
"loss": 5.3168,
|
|
"mean_token_accuracy": 0.1684611275792122,
|
|
"num_tokens": 8264310.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.4844201564788815,
|
|
"epoch": 3.7613918806959403,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039557182108515624,
|
|
"loss": 5.2197,
|
|
"mean_token_accuracy": 0.1802253693342209,
|
|
"num_tokens": 8274184.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.40406289100647,
|
|
"epoch": 3.7655343827671914,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039530189906452457,
|
|
"loss": 5.1556,
|
|
"mean_token_accuracy": 0.18066346347332002,
|
|
"num_tokens": 8283151.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.490919494628907,
|
|
"epoch": 3.7696768848384425,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00039503173438181605,
|
|
"loss": 5.2375,
|
|
"mean_token_accuracy": 0.16958207488059998,
|
|
"num_tokens": 8292574.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.37688570022583,
|
|
"epoch": 3.773819386909693,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00039476132758198236,
|
|
"loss": 5.1589,
|
|
"mean_token_accuracy": 0.17906700521707536,
|
|
"num_tokens": 8301281.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.404768991470337,
|
|
"epoch": 3.7779618889809443,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039449067921046353,
|
|
"loss": 5.2169,
|
|
"mean_token_accuracy": 0.184201979637146,
|
|
"num_tokens": 8310201.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.5153802871704105,
|
|
"epoch": 3.7821043910521954,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039421978981318703,
|
|
"loss": 5.243,
|
|
"mean_token_accuracy": 0.17428653091192245,
|
|
"num_tokens": 8320369.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.490174674987793,
|
|
"epoch": 3.7862468931234465,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039394865993656625,
|
|
"loss": 5.2332,
|
|
"mean_token_accuracy": 0.17829314470291138,
|
|
"num_tokens": 8329519.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.4183516025543215,
|
|
"epoch": 3.7903893951946976,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003936772901274999,
|
|
"loss": 5.2233,
|
|
"mean_token_accuracy": 0.16786337196826934,
|
|
"num_tokens": 8338282.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.422275066375732,
|
|
"epoch": 3.7945318972659487,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003934056809333705,
|
|
"loss": 5.2367,
|
|
"mean_token_accuracy": 0.1760738953948021,
|
|
"num_tokens": 8346861.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.493864250183106,
|
|
"epoch": 3.7986743993372,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039313383290204333,
|
|
"loss": 5.2689,
|
|
"mean_token_accuracy": 0.16675272732973098,
|
|
"num_tokens": 8356078.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.449766826629639,
|
|
"epoch": 3.802816901408451,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003928617465818658,
|
|
"loss": 5.2279,
|
|
"mean_token_accuracy": 0.16728012412786483,
|
|
"num_tokens": 8365487.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.477828598022461,
|
|
"epoch": 3.8069594034797016,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039258942252166554,
|
|
"loss": 5.2102,
|
|
"mean_token_accuracy": 0.17344195693731307,
|
|
"num_tokens": 8374622.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.467281007766724,
|
|
"epoch": 3.8111019055509527,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003923168612707502,
|
|
"loss": 5.1437,
|
|
"mean_token_accuracy": 0.1838704138994217,
|
|
"num_tokens": 8383312.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.375951862335205,
|
|
"epoch": 3.815244407622204,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003920440633789053,
|
|
"loss": 5.1668,
|
|
"mean_token_accuracy": 0.17426021844148637,
|
|
"num_tokens": 8391987.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.411648225784302,
|
|
"epoch": 3.819386909693455,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00039177102939639424,
|
|
"loss": 5.1782,
|
|
"mean_token_accuracy": 0.17309755831956863,
|
|
"num_tokens": 8401263.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.520602798461914,
|
|
"epoch": 3.8235294117647056,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003914977598739563,
|
|
"loss": 5.1998,
|
|
"mean_token_accuracy": 0.18663258850574493,
|
|
"num_tokens": 8410110.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.505752420425415,
|
|
"epoch": 3.8276719138359567,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003912242553628059,
|
|
"loss": 5.2827,
|
|
"mean_token_accuracy": 0.16937017440795898,
|
|
"num_tokens": 8419321.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.470319938659668,
|
|
"epoch": 3.831814415907208,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003909505164146318,
|
|
"loss": 5.2512,
|
|
"mean_token_accuracy": 0.17635022550821305,
|
|
"num_tokens": 8428376.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.446829080581665,
|
|
"epoch": 3.835956917978459,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000390676543581595,
|
|
"loss": 5.1826,
|
|
"mean_token_accuracy": 0.1743323564529419,
|
|
"num_tokens": 8437263.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.361465311050415,
|
|
"epoch": 3.84009942004971,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003904023374163289,
|
|
"loss": 5.1943,
|
|
"mean_token_accuracy": 0.17949013262987137,
|
|
"num_tokens": 8445889.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.503735971450806,
|
|
"epoch": 3.844241922120961,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003901278984719372,
|
|
"loss": 5.3739,
|
|
"mean_token_accuracy": 0.16415481567382811,
|
|
"num_tokens": 8456086.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.492273569107056,
|
|
"epoch": 3.8483844241922123,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00038985322730199326,
|
|
"loss": 5.2018,
|
|
"mean_token_accuracy": 0.1795567825436592,
|
|
"num_tokens": 8465150.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.431647396087646,
|
|
"epoch": 3.8525269262634634,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003895783244605389,
|
|
"loss": 5.2783,
|
|
"mean_token_accuracy": 0.1720671087503433,
|
|
"num_tokens": 8474172.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.404905462265015,
|
|
"epoch": 3.856669428334714,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00038930319050208315,
|
|
"loss": 5.1532,
|
|
"mean_token_accuracy": 0.18708953112363816,
|
|
"num_tokens": 8483636.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.408392810821534,
|
|
"epoch": 3.860811930405965,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038902782598160127,
|
|
"loss": 5.1091,
|
|
"mean_token_accuracy": 0.1881578117609024,
|
|
"num_tokens": 8491560.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.484521150588989,
|
|
"epoch": 3.8649544324772163,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00038875223145453346,
|
|
"loss": 5.4046,
|
|
"mean_token_accuracy": 0.17662398070096968,
|
|
"num_tokens": 8501685.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.551066589355469,
|
|
"epoch": 3.8690969345484674,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003884764074767842,
|
|
"loss": 5.2498,
|
|
"mean_token_accuracy": 0.17176942974328996,
|
|
"num_tokens": 8510580.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.477302551269531,
|
|
"epoch": 3.873239436619718,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003882003546047205,
|
|
"loss": 5.2013,
|
|
"mean_token_accuracy": 0.17522621154785156,
|
|
"num_tokens": 8519595.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.506098508834839,
|
|
"epoch": 3.877381938690969,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038792407339517123,
|
|
"loss": 5.2786,
|
|
"mean_token_accuracy": 0.17400375306606292,
|
|
"num_tokens": 8529118.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.507658243179321,
|
|
"epoch": 3.8815244407622203,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038764756440542574,
|
|
"loss": 5.2477,
|
|
"mean_token_accuracy": 0.17185091674327851,
|
|
"num_tokens": 8539322.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.454025650024414,
|
|
"epoch": 3.8856669428334714,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00038737082819323285,
|
|
"loss": 5.1677,
|
|
"mean_token_accuracy": 0.17740404605865479,
|
|
"num_tokens": 8548707.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.427890157699585,
|
|
"epoch": 3.8898094449047225,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00038709386531679974,
|
|
"loss": 5.2576,
|
|
"mean_token_accuracy": 0.1718251511454582,
|
|
"num_tokens": 8557500.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.436252593994141,
|
|
"epoch": 3.8939519469759736,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038681667633479086,
|
|
"loss": 5.2621,
|
|
"mean_token_accuracy": 0.1682343602180481,
|
|
"num_tokens": 8567465.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.381856250762939,
|
|
"epoch": 3.8980944490472247,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00038653926180632665,
|
|
"loss": 5.1165,
|
|
"mean_token_accuracy": 0.18321423530578612,
|
|
"num_tokens": 8576680.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.495894527435302,
|
|
"epoch": 3.902236951118476,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038626162229098257,
|
|
"loss": 5.2631,
|
|
"mean_token_accuracy": 0.16617788672447203,
|
|
"num_tokens": 8585539.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.400326347351074,
|
|
"epoch": 3.9063794531897265,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003859837583487877,
|
|
"loss": 5.1007,
|
|
"mean_token_accuracy": 0.18613590002059938,
|
|
"num_tokens": 8593598.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.4148589134216305,
|
|
"epoch": 3.9105219552609776,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003857056705402241,
|
|
"loss": 5.236,
|
|
"mean_token_accuracy": 0.1721765860915184,
|
|
"num_tokens": 8603440.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.384281969070434,
|
|
"epoch": 3.9146644573322287,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003854273594262253,
|
|
"loss": 5.0591,
|
|
"mean_token_accuracy": 0.1858549162745476,
|
|
"num_tokens": 8612490.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.387153816223145,
|
|
"epoch": 3.91880695940348,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003851488255681752,
|
|
"loss": 5.1508,
|
|
"mean_token_accuracy": 0.17980385273694993,
|
|
"num_tokens": 8621994.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.462166833877563,
|
|
"epoch": 3.922949461474731,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000384870069527907,
|
|
"loss": 5.2269,
|
|
"mean_token_accuracy": 0.16994049549102783,
|
|
"num_tokens": 8631185.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.445327711105347,
|
|
"epoch": 3.9270919635459816,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00038459109186770213,
|
|
"loss": 5.1201,
|
|
"mean_token_accuracy": 0.18937034606933595,
|
|
"num_tokens": 8639018.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.365725660324097,
|
|
"epoch": 3.9312344656172327,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038431189315028905,
|
|
"loss": 5.1293,
|
|
"mean_token_accuracy": 0.1805494785308838,
|
|
"num_tokens": 8647672.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.42286114692688,
|
|
"epoch": 3.9353769676884838,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003840324739388419,
|
|
"loss": 5.2659,
|
|
"mean_token_accuracy": 0.16662627309560776,
|
|
"num_tokens": 8656638.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.4803203582763675,
|
|
"epoch": 3.939519469759735,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038375283479698,
|
|
"loss": 5.2229,
|
|
"mean_token_accuracy": 0.18231208473443986,
|
|
"num_tokens": 8666167.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.505121231079102,
|
|
"epoch": 3.943661971830986,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003834729762887659,
|
|
"loss": 5.3067,
|
|
"mean_token_accuracy": 0.16447765678167342,
|
|
"num_tokens": 8676435.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.413615751266479,
|
|
"epoch": 3.947804473902237,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003831928989787049,
|
|
"loss": 5.1808,
|
|
"mean_token_accuracy": 0.18012969940900803,
|
|
"num_tokens": 8686011.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.463970613479614,
|
|
"epoch": 3.951946975973488,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003829126034317435,
|
|
"loss": 5.1959,
|
|
"mean_token_accuracy": 0.17213619947433473,
|
|
"num_tokens": 8695063.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.517300796508789,
|
|
"epoch": 3.956089478044739,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003826320902132684,
|
|
"loss": 5.299,
|
|
"mean_token_accuracy": 0.16880902647972107,
|
|
"num_tokens": 8704585.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.469470262527466,
|
|
"epoch": 3.96023198011599,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038235135988910545,
|
|
"loss": 5.2111,
|
|
"mean_token_accuracy": 0.17501715570688248,
|
|
"num_tokens": 8714097.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.433980226516724,
|
|
"epoch": 3.964374482187241,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00038207041302551844,
|
|
"loss": 5.1327,
|
|
"mean_token_accuracy": 0.1802523225545883,
|
|
"num_tokens": 8722890.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.36038613319397,
|
|
"epoch": 3.968516984258492,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003817892501892078,
|
|
"loss": 5.1474,
|
|
"mean_token_accuracy": 0.18226009160280227,
|
|
"num_tokens": 8732046.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.385818529129028,
|
|
"epoch": 3.9726594863297433,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038150787194730977,
|
|
"loss": 5.1867,
|
|
"mean_token_accuracy": 0.18098812699317932,
|
|
"num_tokens": 8741750.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.49769434928894,
|
|
"epoch": 3.976801988400994,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000381226278867395,
|
|
"loss": 5.1585,
|
|
"mean_token_accuracy": 0.17722393125295638,
|
|
"num_tokens": 8750141.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.3994584560394285,
|
|
"epoch": 3.980944490472245,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00038094447151746746,
|
|
"loss": 5.0956,
|
|
"mean_token_accuracy": 0.18723244071006775,
|
|
"num_tokens": 8758567.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.376461982727051,
|
|
"epoch": 3.985086992543496,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003806624504659635,
|
|
"loss": 5.193,
|
|
"mean_token_accuracy": 0.1782562255859375,
|
|
"num_tokens": 8767829.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.416551733016968,
|
|
"epoch": 3.9892294946147473,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003803802162817503,
|
|
"loss": 5.0891,
|
|
"mean_token_accuracy": 0.18413031548261644,
|
|
"num_tokens": 8776911.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.403548002243042,
|
|
"epoch": 3.9933719966859984,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000380097769534125,
|
|
"loss": 5.2011,
|
|
"mean_token_accuracy": 0.18309975266456605,
|
|
"num_tokens": 8786117.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.401311016082763,
|
|
"epoch": 3.9975144987572495,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003798151107928136,
|
|
"loss": 5.1436,
|
|
"mean_token_accuracy": 0.18381998240947722,
|
|
"num_tokens": 8795647.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.363515567779541,
|
|
"epoch": 4.001657000828501,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00037953224062796967,
|
|
"loss": 5.1114,
|
|
"mean_token_accuracy": 0.18405449539422988,
|
|
"num_tokens": 8803480.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.504069137573242,
|
|
"epoch": 4.005799502899752,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003792491596101733,
|
|
"loss": 4.9454,
|
|
"mean_token_accuracy": 0.19255731254816055,
|
|
"num_tokens": 8812825.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.45196304321289,
|
|
"epoch": 4.009942004971003,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003789658683104298,
|
|
"loss": 5.0053,
|
|
"mean_token_accuracy": 0.19097152799367906,
|
|
"num_tokens": 8821911.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.429521989822388,
|
|
"epoch": 4.014084507042254,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00037868236730016866,
|
|
"loss": 4.9518,
|
|
"mean_token_accuracy": 0.18948255032300948,
|
|
"num_tokens": 8830691.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.400455188751221,
|
|
"epoch": 4.018227009113504,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037839865715124234,
|
|
"loss": 4.9281,
|
|
"mean_token_accuracy": 0.19010053426027299,
|
|
"num_tokens": 8840410.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.374143838882446,
|
|
"epoch": 4.022369511184755,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00037811473843592534,
|
|
"loss": 4.9042,
|
|
"mean_token_accuracy": 0.19410175532102586,
|
|
"num_tokens": 8849662.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.4465690612792965,
|
|
"epoch": 4.026512013256006,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00037783061172691273,
|
|
"loss": 4.8985,
|
|
"mean_token_accuracy": 0.19412659406661986,
|
|
"num_tokens": 8858666.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.497456216812134,
|
|
"epoch": 4.0306545153272575,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003775462775973191,
|
|
"loss": 4.9611,
|
|
"mean_token_accuracy": 0.1847592905163765,
|
|
"num_tokens": 8867885.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.351856470108032,
|
|
"epoch": 4.034797017398509,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003772617366206773,
|
|
"loss": 4.9175,
|
|
"mean_token_accuracy": 0.19424951672554017,
|
|
"num_tokens": 8877458.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.363556432723999,
|
|
"epoch": 4.03893951946976,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037697698937093783,
|
|
"loss": 4.9251,
|
|
"mean_token_accuracy": 0.18704084753990174,
|
|
"num_tokens": 8886995.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.463095617294312,
|
|
"epoch": 4.043082021541011,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00037669203642246685,
|
|
"loss": 5.026,
|
|
"mean_token_accuracy": 0.18413490056991577,
|
|
"num_tokens": 8895224.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.4529016494750975,
|
|
"epoch": 4.047224523612262,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037640687835004567,
|
|
"loss": 4.9189,
|
|
"mean_token_accuracy": 0.1925043895840645,
|
|
"num_tokens": 8905216.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.426698875427246,
|
|
"epoch": 4.051367025683513,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003761215157288692,
|
|
"loss": 4.9558,
|
|
"mean_token_accuracy": 0.19299548715353013,
|
|
"num_tokens": 8915493.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.362393617630005,
|
|
"epoch": 4.055509527754764,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037583594913454514,
|
|
"loss": 4.925,
|
|
"mean_token_accuracy": 0.1959809884428978,
|
|
"num_tokens": 8924740.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.42564959526062,
|
|
"epoch": 4.059652029826015,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003755501791430925,
|
|
"loss": 4.946,
|
|
"mean_token_accuracy": 0.18527257889509202,
|
|
"num_tokens": 8934726.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.448423719406128,
|
|
"epoch": 4.063794531897266,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037526420633094044,
|
|
"loss": 4.9581,
|
|
"mean_token_accuracy": 0.18748125433921814,
|
|
"num_tokens": 8944301.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.449982643127441,
|
|
"epoch": 4.067937033968517,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037497803127492757,
|
|
"loss": 4.9881,
|
|
"mean_token_accuracy": 0.19187170565128325,
|
|
"num_tokens": 8953967.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.326543855667114,
|
|
"epoch": 4.072079536039768,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003746916545523001,
|
|
"loss": 4.8948,
|
|
"mean_token_accuracy": 0.19177066087722777,
|
|
"num_tokens": 8962983.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.381871175765991,
|
|
"epoch": 4.076222038111019,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003744050767407113,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.18154184073209761,
|
|
"num_tokens": 8972019.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.40556058883667,
|
|
"epoch": 4.08036454018227,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003741182984182198,
|
|
"loss": 4.901,
|
|
"mean_token_accuracy": 0.20234887301921844,
|
|
"num_tokens": 8981751.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.3616517066955565,
|
|
"epoch": 4.084507042253521,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003738313201632889,
|
|
"loss": 4.8923,
|
|
"mean_token_accuracy": 0.19780989587306977,
|
|
"num_tokens": 8990330.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.314311408996582,
|
|
"epoch": 4.088649544324772,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003735441425547851,
|
|
"loss": 4.905,
|
|
"mean_token_accuracy": 0.19072338491678237,
|
|
"num_tokens": 8998366.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.290538501739502,
|
|
"epoch": 4.092792046396023,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 0.00037325676617197697,
|
|
"loss": 4.8317,
|
|
"mean_token_accuracy": 0.20186650305986403,
|
|
"num_tokens": 9006881.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.323595571517944,
|
|
"epoch": 4.096934548467274,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003729691915945342,
|
|
"loss": 4.8944,
|
|
"mean_token_accuracy": 0.19853760153055192,
|
|
"num_tokens": 9015415.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.40374059677124,
|
|
"epoch": 4.1010770505385254,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000372681419402526,
|
|
"loss": 4.9483,
|
|
"mean_token_accuracy": 0.19546929001808167,
|
|
"num_tokens": 9024578.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.381409311294556,
|
|
"epoch": 4.105219552609777,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003723934501764203,
|
|
"loss": 4.9597,
|
|
"mean_token_accuracy": 0.1867024227976799,
|
|
"num_tokens": 9032637.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.395318984985352,
|
|
"epoch": 4.109362054681028,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003721052844970826,
|
|
"loss": 4.8913,
|
|
"mean_token_accuracy": 0.18718749284744263,
|
|
"num_tokens": 9041513.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.394489192962647,
|
|
"epoch": 4.113504556752279,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00037181692294577457,
|
|
"loss": 4.9155,
|
|
"mean_token_accuracy": 0.1854037255048752,
|
|
"num_tokens": 9051024.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.33848295211792,
|
|
"epoch": 4.117647058823529,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00037152836610415284,
|
|
"loss": 4.8895,
|
|
"mean_token_accuracy": 0.19274405539035797,
|
|
"num_tokens": 9059494.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.31979193687439,
|
|
"epoch": 4.12178956089478,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037123961455426836,
|
|
"loss": 4.7988,
|
|
"mean_token_accuracy": 0.19865831583738328,
|
|
"num_tokens": 9068166.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.391095542907715,
|
|
"epoch": 4.125932062966031,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037095066887856434,
|
|
"loss": 4.907,
|
|
"mean_token_accuracy": 0.19333944320678711,
|
|
"num_tokens": 9077283.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.3755443572998045,
|
|
"epoch": 4.130074565037282,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003706615296598758,
|
|
"loss": 4.9332,
|
|
"mean_token_accuracy": 0.19493403136730195,
|
|
"num_tokens": 9086605.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.301020622253418,
|
|
"epoch": 4.134217067108533,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003703721974814282,
|
|
"loss": 4.9169,
|
|
"mean_token_accuracy": 0.19399153292179108,
|
|
"num_tokens": 9095567.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.282594537734985,
|
|
"epoch": 4.1383595691797845,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037008267292683615,
|
|
"loss": 4.9012,
|
|
"mean_token_accuracy": 0.19265301078557967,
|
|
"num_tokens": 9104457.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.398877716064453,
|
|
"epoch": 4.142502071251036,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036979295658010227,
|
|
"loss": 4.9533,
|
|
"mean_token_accuracy": 0.20162091553211212,
|
|
"num_tokens": 9113945.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.142502071251036,
|
|
"eval_entropy": 5.273210916946184,
|
|
"eval_loss": 5.83656120300293,
|
|
"eval_mean_token_accuracy": 0.15772452866281272,
|
|
"eval_num_tokens": 9113945.0,
|
|
"eval_runtime": 2.9415,
|
|
"eval_samples_per_second": 1273.844,
|
|
"eval_steps_per_second": 159.443,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.424594402313232,
|
|
"epoch": 4.146644573322287,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003695030490256161,
|
|
"loss": 5.0303,
|
|
"mean_token_accuracy": 0.18944208920001984,
|
|
"num_tokens": 9122996.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.355675983428955,
|
|
"epoch": 4.150787075393538,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000369212950848153,
|
|
"loss": 4.8793,
|
|
"mean_token_accuracy": 0.1962927296757698,
|
|
"num_tokens": 9132079.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.384340238571167,
|
|
"epoch": 4.154929577464789,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003689226626328725,
|
|
"loss": 5.0058,
|
|
"mean_token_accuracy": 0.19004727751016617,
|
|
"num_tokens": 9141335.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.442283582687378,
|
|
"epoch": 4.15907207953604,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00036863218496531767,
|
|
"loss": 5.0264,
|
|
"mean_token_accuracy": 0.18990386426448821,
|
|
"num_tokens": 9150049.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.409811115264892,
|
|
"epoch": 4.163214581607291,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003683415184314139,
|
|
"loss": 4.9057,
|
|
"mean_token_accuracy": 0.19072521328926087,
|
|
"num_tokens": 9159245.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.281745195388794,
|
|
"epoch": 4.167357083678542,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036805066361746716,
|
|
"loss": 4.9432,
|
|
"mean_token_accuracy": 0.19057238101959229,
|
|
"num_tokens": 9167892.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.262811374664307,
|
|
"epoch": 4.1714995857497925,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036775962111016346,
|
|
"loss": 4.9189,
|
|
"mean_token_accuracy": 0.19675376862287522,
|
|
"num_tokens": 9176067.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.335107898712158,
|
|
"epoch": 4.175642087821044,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036746839149656734,
|
|
"loss": 4.9032,
|
|
"mean_token_accuracy": 0.19636716991662978,
|
|
"num_tokens": 9185318.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.338614749908447,
|
|
"epoch": 4.179784589892295,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003671769753641207,
|
|
"loss": 4.9605,
|
|
"mean_token_accuracy": 0.19530149400234223,
|
|
"num_tokens": 9194885.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.349650192260742,
|
|
"epoch": 4.183927091963546,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00036688537330064187,
|
|
"loss": 4.987,
|
|
"mean_token_accuracy": 0.19436514377593994,
|
|
"num_tokens": 9203906.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.318757629394531,
|
|
"epoch": 4.188069594034797,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000366593585894324,
|
|
"loss": 4.8507,
|
|
"mean_token_accuracy": 0.2013680413365364,
|
|
"num_tokens": 9213351.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.347932624816894,
|
|
"epoch": 4.192212096106048,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003663016137337341,
|
|
"loss": 4.9278,
|
|
"mean_token_accuracy": 0.18784158080816268,
|
|
"num_tokens": 9222191.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.372864198684693,
|
|
"epoch": 4.196354598177299,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00036600945740781197,
|
|
"loss": 4.9668,
|
|
"mean_token_accuracy": 0.1919815331697464,
|
|
"num_tokens": 9231517.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.393643712997436,
|
|
"epoch": 4.20049710024855,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036571711750586896,
|
|
"loss": 4.9891,
|
|
"mean_token_accuracy": 0.19960682690143586,
|
|
"num_tokens": 9241208.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.391959381103516,
|
|
"epoch": 4.204639602319801,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003654245946175864,
|
|
"loss": 4.9079,
|
|
"mean_token_accuracy": 0.19324131309986115,
|
|
"num_tokens": 9250237.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.37893590927124,
|
|
"epoch": 4.2087821043910525,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036513188933301504,
|
|
"loss": 5.0399,
|
|
"mean_token_accuracy": 0.18783097416162492,
|
|
"num_tokens": 9260273.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.378773164749146,
|
|
"epoch": 4.212924606462304,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003648390022425734,
|
|
"loss": 4.8996,
|
|
"mean_token_accuracy": 0.19140231609344482,
|
|
"num_tokens": 9268878.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.325588417053223,
|
|
"epoch": 4.217067108533554,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003645459339370468,
|
|
"loss": 4.9262,
|
|
"mean_token_accuracy": 0.19706592857837676,
|
|
"num_tokens": 9276324.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.312953901290894,
|
|
"epoch": 4.221209610604805,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036425268500758594,
|
|
"loss": 4.9725,
|
|
"mean_token_accuracy": 0.1936806783080101,
|
|
"num_tokens": 9285052.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.354986000061035,
|
|
"epoch": 4.225352112676056,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00036395925604570614,
|
|
"loss": 4.8763,
|
|
"mean_token_accuracy": 0.20042537599802018,
|
|
"num_tokens": 9293324.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.3900477409362795,
|
|
"epoch": 4.229494614747307,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003636656476432854,
|
|
"loss": 4.9262,
|
|
"mean_token_accuracy": 0.19263672083616257,
|
|
"num_tokens": 9302131.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.263408470153808,
|
|
"epoch": 4.233637116818558,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003633718603925642,
|
|
"loss": 4.862,
|
|
"mean_token_accuracy": 0.19388986378908157,
|
|
"num_tokens": 9311128.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.420498418807983,
|
|
"epoch": 4.237779618889809,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003630778948861434,
|
|
"loss": 4.9352,
|
|
"mean_token_accuracy": 0.1895424723625183,
|
|
"num_tokens": 9319849.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.336805248260498,
|
|
"epoch": 4.2419221209610605,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003627837517169836,
|
|
"loss": 4.9458,
|
|
"mean_token_accuracy": 0.18560263216495515,
|
|
"num_tokens": 9328355.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.352609777450562,
|
|
"epoch": 4.246064623032312,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003624894314784037,
|
|
"loss": 4.9236,
|
|
"mean_token_accuracy": 0.18928787410259246,
|
|
"num_tokens": 9337562.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.323793935775757,
|
|
"epoch": 4.250207125103563,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003621949347640797,
|
|
"loss": 4.9315,
|
|
"mean_token_accuracy": 0.196185165643692,
|
|
"num_tokens": 9347851.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.398552274703979,
|
|
"epoch": 4.254349627174814,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003619002621680439,
|
|
"loss": 4.9325,
|
|
"mean_token_accuracy": 0.19673411697149276,
|
|
"num_tokens": 9356524.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.4241108894348145,
|
|
"epoch": 4.258492129246065,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003616054142846829,
|
|
"loss": 5.0307,
|
|
"mean_token_accuracy": 0.1882170706987381,
|
|
"num_tokens": 9365521.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.301144218444824,
|
|
"epoch": 4.262634631317316,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003613103917087371,
|
|
"loss": 4.9262,
|
|
"mean_token_accuracy": 0.19470559656620026,
|
|
"num_tokens": 9374088.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.345515966415405,
|
|
"epoch": 4.266777133388567,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003610151950352994,
|
|
"loss": 4.9794,
|
|
"mean_token_accuracy": 0.19499585926532745,
|
|
"num_tokens": 9382683.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.306409788131714,
|
|
"epoch": 4.270919635459817,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003607198248598137,
|
|
"loss": 4.9063,
|
|
"mean_token_accuracy": 0.194849793612957,
|
|
"num_tokens": 9391790.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.414876556396484,
|
|
"epoch": 4.2750621375310685,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003604242817780739,
|
|
"loss": 4.9901,
|
|
"mean_token_accuracy": 0.19321189671754838,
|
|
"num_tokens": 9400234.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.350374364852906,
|
|
"epoch": 4.27920463960232,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00036012856638622273,
|
|
"loss": 4.9079,
|
|
"mean_token_accuracy": 0.19192621409893035,
|
|
"num_tokens": 9409513.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.292260932922363,
|
|
"epoch": 4.283347141673571,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003598326792807504,
|
|
"loss": 4.9821,
|
|
"mean_token_accuracy": 0.18258902430534363,
|
|
"num_tokens": 9418608.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.427756977081299,
|
|
"epoch": 4.287489643744822,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00035953662105849357,
|
|
"loss": 5.0092,
|
|
"mean_token_accuracy": 0.19452760964632035,
|
|
"num_tokens": 9428705.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.3185462474823,
|
|
"epoch": 4.291632145816073,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000359240392316634,
|
|
"loss": 4.9177,
|
|
"mean_token_accuracy": 0.18542011678218842,
|
|
"num_tokens": 9438149.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.330511379241943,
|
|
"epoch": 4.295774647887324,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003589439936526974,
|
|
"loss": 4.8709,
|
|
"mean_token_accuracy": 0.19628295600414275,
|
|
"num_tokens": 9447566.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.303903436660766,
|
|
"epoch": 4.299917149958575,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003586474256645525,
|
|
"loss": 4.9641,
|
|
"mean_token_accuracy": 0.1915314257144928,
|
|
"num_tokens": 9458156.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.410308837890625,
|
|
"epoch": 4.304059652029826,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00035835068895040905,
|
|
"loss": 5.0014,
|
|
"mean_token_accuracy": 0.19137390553951264,
|
|
"num_tokens": 9467388.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.388073253631592,
|
|
"epoch": 4.308202154101077,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035805378410881753,
|
|
"loss": 4.9642,
|
|
"mean_token_accuracy": 0.1940509021282196,
|
|
"num_tokens": 9477515.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.389194107055664,
|
|
"epoch": 4.312344656172328,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003577567117386675,
|
|
"loss": 4.9854,
|
|
"mean_token_accuracy": 0.1937498316168785,
|
|
"num_tokens": 9485851.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.377271747589111,
|
|
"epoch": 4.3164871582435795,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003574594724391863,
|
|
"loss": 4.9773,
|
|
"mean_token_accuracy": 0.18631901144981383,
|
|
"num_tokens": 9495523.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.303675985336303,
|
|
"epoch": 4.32062966031483,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003571620668099382,
|
|
"loss": 4.9104,
|
|
"mean_token_accuracy": 0.19740363359451293,
|
|
"num_tokens": 9504195.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.377063655853272,
|
|
"epoch": 4.324772162386081,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003568644954508227,
|
|
"loss": 4.9616,
|
|
"mean_token_accuracy": 0.1940310388803482,
|
|
"num_tokens": 9513313.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.400073862075805,
|
|
"epoch": 4.328914664457332,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003565667589620739,
|
|
"loss": 5.0313,
|
|
"mean_token_accuracy": 0.18615816980600358,
|
|
"num_tokens": 9522387.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.380652332305909,
|
|
"epoch": 4.333057166528583,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003562688579442588,
|
|
"loss": 4.9582,
|
|
"mean_token_accuracy": 0.1893748164176941,
|
|
"num_tokens": 9531522.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.349949216842651,
|
|
"epoch": 4.337199668599834,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00035597079299827625,
|
|
"loss": 4.9875,
|
|
"mean_token_accuracy": 0.18508986830711366,
|
|
"num_tokens": 9540248.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.3083008289337155,
|
|
"epoch": 4.341342170671085,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035567256472535585,
|
|
"loss": 4.9192,
|
|
"mean_token_accuracy": 0.19538334757089615,
|
|
"num_tokens": 9549850.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.325525283813477,
|
|
"epoch": 4.345484672742336,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003553741737270566,
|
|
"loss": 4.9167,
|
|
"mean_token_accuracy": 0.19581695050001144,
|
|
"num_tokens": 9558929.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.291470670700074,
|
|
"epoch": 4.3496271748135875,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035507562060526574,
|
|
"loss": 4.9163,
|
|
"mean_token_accuracy": 0.19085376411676408,
|
|
"num_tokens": 9567627.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.419774055480957,
|
|
"epoch": 4.353769676884839,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00035477690596219763,
|
|
"loss": 5.0513,
|
|
"mean_token_accuracy": 0.18172826021909713,
|
|
"num_tokens": 9576223.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.422041416168213,
|
|
"epoch": 4.35791217895609,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00035447803040039225,
|
|
"loss": 5.0178,
|
|
"mean_token_accuracy": 0.1885324940085411,
|
|
"num_tokens": 9585510.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.344962453842163,
|
|
"epoch": 4.362054681027341,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003541789945227143,
|
|
"loss": 4.9361,
|
|
"mean_token_accuracy": 0.1920488640666008,
|
|
"num_tokens": 9594843.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.401527786254883,
|
|
"epoch": 4.366197183098592,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00035387979893235184,
|
|
"loss": 4.9467,
|
|
"mean_token_accuracy": 0.19816664457321168,
|
|
"num_tokens": 9604142.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.339557313919068,
|
|
"epoch": 4.370339685169842,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003535804442328151,
|
|
"loss": 4.994,
|
|
"mean_token_accuracy": 0.197935588657856,
|
|
"num_tokens": 9613669.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.395236444473267,
|
|
"epoch": 4.374482187241093,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00035328093102793506,
|
|
"loss": 4.9843,
|
|
"mean_token_accuracy": 0.19234263002872468,
|
|
"num_tokens": 9623502.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.377492380142212,
|
|
"epoch": 4.378624689312344,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035298125992186283,
|
|
"loss": 4.9784,
|
|
"mean_token_accuracy": 0.1910512089729309,
|
|
"num_tokens": 9632041.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.310422706604004,
|
|
"epoch": 4.3827671913835955,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035268143151906764,
|
|
"loss": 4.9853,
|
|
"mean_token_accuracy": 0.18939480930566788,
|
|
"num_tokens": 9641442.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.356794643402099,
|
|
"epoch": 4.386909693454847,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00035238144642433616,
|
|
"loss": 4.956,
|
|
"mean_token_accuracy": 0.19045386761426925,
|
|
"num_tokens": 9650857.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.461606073379516,
|
|
"epoch": 4.391052195526098,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000352081305242771,
|
|
"loss": 4.9906,
|
|
"mean_token_accuracy": 0.1938766911625862,
|
|
"num_tokens": 9659173.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.355851411819458,
|
|
"epoch": 4.395194697597349,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00035178100857978994,
|
|
"loss": 4.9344,
|
|
"mean_token_accuracy": 0.19574909955263137,
|
|
"num_tokens": 9668114.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.261717939376831,
|
|
"epoch": 4.3993371996686,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00035148055704112405,
|
|
"loss": 4.9927,
|
|
"mean_token_accuracy": 0.1930756837129593,
|
|
"num_tokens": 9676628.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.305881071090698,
|
|
"epoch": 4.403479701739851,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003511799512328168,
|
|
"loss": 4.8897,
|
|
"mean_token_accuracy": 0.19723279178142547,
|
|
"num_tokens": 9685801.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.3678816795349125,
|
|
"epoch": 4.407622203811102,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003508791917612232,
|
|
"loss": 4.9677,
|
|
"mean_token_accuracy": 0.18847441226243972,
|
|
"num_tokens": 9694237.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.357159948348999,
|
|
"epoch": 4.411764705882353,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00035057827923300785,
|
|
"loss": 4.887,
|
|
"mean_token_accuracy": 0.1993523895740509,
|
|
"num_tokens": 9701904.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.308569765090942,
|
|
"epoch": 4.415907207953604,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00035027721425514414,
|
|
"loss": 4.8661,
|
|
"mean_token_accuracy": 0.20139306634664536,
|
|
"num_tokens": 9710784.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.310255718231201,
|
|
"epoch": 4.4200497100248555,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003499759974349131,
|
|
"loss": 4.9969,
|
|
"mean_token_accuracy": 0.19186950027942656,
|
|
"num_tokens": 9720452.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.335639381408692,
|
|
"epoch": 4.424192212096106,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00034967462937990194,
|
|
"loss": 4.9619,
|
|
"mean_token_accuracy": 0.18502581119537354,
|
|
"num_tokens": 9729122.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.414971780776978,
|
|
"epoch": 4.428334714167357,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000349373110698003,
|
|
"loss": 5.0216,
|
|
"mean_token_accuracy": 0.18420466929674148,
|
|
"num_tokens": 9738756.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.406450796127319,
|
|
"epoch": 4.432477216238608,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003490714419974124,
|
|
"loss": 5.0113,
|
|
"mean_token_accuracy": 0.1904602199792862,
|
|
"num_tokens": 9748547.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.358873558044434,
|
|
"epoch": 4.436619718309859,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00034876962388662885,
|
|
"loss": 5.0582,
|
|
"mean_token_accuracy": 0.18162461519241332,
|
|
"num_tokens": 9758421.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.277571201324463,
|
|
"epoch": 4.44076222038111,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003484676569744525,
|
|
"loss": 4.8563,
|
|
"mean_token_accuracy": 0.19333565086126328,
|
|
"num_tokens": 9766855.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.348770189285278,
|
|
"epoch": 4.444904722452361,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003481655418699836,
|
|
"loss": 4.9589,
|
|
"mean_token_accuracy": 0.1960444375872612,
|
|
"num_tokens": 9776157.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.279411029815674,
|
|
"epoch": 4.449047224523612,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034786327918262126,
|
|
"loss": 4.8489,
|
|
"mean_token_accuracy": 0.2087496116757393,
|
|
"num_tokens": 9784386.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.354473733901978,
|
|
"epoch": 4.4531897265948635,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00034756086952206255,
|
|
"loss": 4.9009,
|
|
"mean_token_accuracy": 0.1958460658788681,
|
|
"num_tokens": 9793838.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.353402185440063,
|
|
"epoch": 4.457332228666115,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003472583134983005,
|
|
"loss": 4.9971,
|
|
"mean_token_accuracy": 0.19221619367599488,
|
|
"num_tokens": 9802595.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.384884357452393,
|
|
"epoch": 4.461474730737366,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034695561172162396,
|
|
"loss": 4.9775,
|
|
"mean_token_accuracy": 0.19024742841720582,
|
|
"num_tokens": 9811309.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.2829090595245365,
|
|
"epoch": 4.465617232808617,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00034665276480261526,
|
|
"loss": 4.9139,
|
|
"mean_token_accuracy": 0.1915873184800148,
|
|
"num_tokens": 9820327.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.351649951934815,
|
|
"epoch": 4.469759734879867,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003463497733521499,
|
|
"loss": 4.9852,
|
|
"mean_token_accuracy": 0.18354866653680801,
|
|
"num_tokens": 9829588.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.42580213546753,
|
|
"epoch": 4.473902236951118,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003460466379813947,
|
|
"loss": 4.9887,
|
|
"mean_token_accuracy": 0.18973219394683838,
|
|
"num_tokens": 9838548.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.335280847549439,
|
|
"epoch": 4.478044739022369,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003457433593018068,
|
|
"loss": 5.0018,
|
|
"mean_token_accuracy": 0.18838239908218385,
|
|
"num_tokens": 9847925.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.396121597290039,
|
|
"epoch": 4.48218724109362,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00034543993792513243,
|
|
"loss": 5.0169,
|
|
"mean_token_accuracy": 0.18753209412097932,
|
|
"num_tokens": 9856839.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.37349534034729,
|
|
"epoch": 4.486329743164871,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003451363744634057,
|
|
"loss": 4.978,
|
|
"mean_token_accuracy": 0.19382779747247697,
|
|
"num_tokens": 9866519.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.430073976516724,
|
|
"epoch": 4.4904722452361225,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00034483266952894725,
|
|
"loss": 5.0582,
|
|
"mean_token_accuracy": 0.19110478311777115,
|
|
"num_tokens": 9876866.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.440963220596314,
|
|
"epoch": 4.494614747307374,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003445288237343632,
|
|
"loss": 5.0042,
|
|
"mean_token_accuracy": 0.1870105817914009,
|
|
"num_tokens": 9884639.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.332541465759277,
|
|
"epoch": 4.498757249378625,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003442248376925436,
|
|
"loss": 4.9715,
|
|
"mean_token_accuracy": 0.1879585549235344,
|
|
"num_tokens": 9893819.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.251445341110229,
|
|
"epoch": 4.502899751449876,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00034392071201666167,
|
|
"loss": 4.9164,
|
|
"mean_token_accuracy": 0.19130610078573226,
|
|
"num_tokens": 9902419.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.390727472305298,
|
|
"epoch": 4.507042253521127,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000343616447320172,
|
|
"loss": 4.9565,
|
|
"mean_token_accuracy": 0.19780462980270386,
|
|
"num_tokens": 9912434.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.366470289230347,
|
|
"epoch": 4.511184755592378,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003433120442168097,
|
|
"loss": 4.9114,
|
|
"mean_token_accuracy": 0.193550705909729,
|
|
"num_tokens": 9921800.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.316132354736328,
|
|
"epoch": 4.515327257663629,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00034300750332058924,
|
|
"loss": 4.9803,
|
|
"mean_token_accuracy": 0.18861733824014665,
|
|
"num_tokens": 9930863.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.340768337249756,
|
|
"epoch": 4.51946975973488,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003427028252458028,
|
|
"loss": 4.995,
|
|
"mean_token_accuracy": 0.1959969326853752,
|
|
"num_tokens": 9940524.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.398335361480713,
|
|
"epoch": 4.5236122618061305,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003423980106070194,
|
|
"loss": 4.98,
|
|
"mean_token_accuracy": 0.18993111699819565,
|
|
"num_tokens": 9948949.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.363409519195557,
|
|
"epoch": 4.527754763877382,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003420930600190834,
|
|
"loss": 5.0007,
|
|
"mean_token_accuracy": 0.18487946093082427,
|
|
"num_tokens": 9958154.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.353360748291015,
|
|
"epoch": 4.531897265948633,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003417879740971135,
|
|
"loss": 4.8917,
|
|
"mean_token_accuracy": 0.19643958359956742,
|
|
"num_tokens": 9967023.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.392623519897461,
|
|
"epoch": 4.536039768019884,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003414827534565014,
|
|
"loss": 4.9908,
|
|
"mean_token_accuracy": 0.18982938379049302,
|
|
"num_tokens": 9976596.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.358148193359375,
|
|
"epoch": 4.540182270091135,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034117739871291036,
|
|
"loss": 4.8992,
|
|
"mean_token_accuracy": 0.19934887886047364,
|
|
"num_tokens": 9985330.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.35659613609314,
|
|
"epoch": 4.544324772162386,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034087191048227445,
|
|
"loss": 5.032,
|
|
"mean_token_accuracy": 0.1861831620335579,
|
|
"num_tokens": 9996053.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.335864496231079,
|
|
"epoch": 4.548467274233637,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003405662893807967,
|
|
"loss": 4.9445,
|
|
"mean_token_accuracy": 0.1877502605319023,
|
|
"num_tokens": 10005517.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.389253044128418,
|
|
"epoch": 4.552609776304888,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003402605360249482,
|
|
"loss": 5.0416,
|
|
"mean_token_accuracy": 0.18830080330371857,
|
|
"num_tokens": 10014354.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.3128773212432865,
|
|
"epoch": 4.556752278376139,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000339954651031467,
|
|
"loss": 4.9238,
|
|
"mean_token_accuracy": 0.19178629368543626,
|
|
"num_tokens": 10023194.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.556752278376139,
|
|
"eval_entropy": 5.195640559898003,
|
|
"eval_loss": 5.80400276184082,
|
|
"eval_mean_token_accuracy": 0.15982486544323882,
|
|
"eval_num_tokens": 10023194.0,
|
|
"eval_runtime": 2.7416,
|
|
"eval_samples_per_second": 1366.696,
|
|
"eval_steps_per_second": 171.065,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.267048501968384,
|
|
"epoch": 4.5608947804473905,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033964863501735647,
|
|
"loss": 4.9188,
|
|
"mean_token_accuracy": 0.19228280782699586,
|
|
"num_tokens": 10032490.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.392655229568481,
|
|
"epoch": 4.565037282518642,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033934248859988453,
|
|
"loss": 4.9346,
|
|
"mean_token_accuracy": 0.19580908715724946,
|
|
"num_tokens": 10041613.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.262086534500122,
|
|
"epoch": 4.569179784589892,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003390362123965817,
|
|
"loss": 4.8733,
|
|
"mean_token_accuracy": 0.1911102056503296,
|
|
"num_tokens": 10050500.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.36342945098877,
|
|
"epoch": 4.573322286661144,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003387298070252407,
|
|
"loss": 4.9492,
|
|
"mean_token_accuracy": 0.1935834139585495,
|
|
"num_tokens": 10059650.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.337366104125977,
|
|
"epoch": 4.577464788732394,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003384232731039146,
|
|
"loss": 4.9844,
|
|
"mean_token_accuracy": 0.19709572941064835,
|
|
"num_tokens": 10067882.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.300627326965332,
|
|
"epoch": 4.581607290803645,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033811661125091584,
|
|
"loss": 4.9253,
|
|
"mean_token_accuracy": 0.19670316576957703,
|
|
"num_tokens": 10077147.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.369974613189697,
|
|
"epoch": 4.585749792874896,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00033780982208481494,
|
|
"loss": 4.9829,
|
|
"mean_token_accuracy": 0.1897311642765999,
|
|
"num_tokens": 10086149.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.303711175918579,
|
|
"epoch": 4.589892294946147,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003375029062244392,
|
|
"loss": 4.9027,
|
|
"mean_token_accuracy": 0.20062701553106307,
|
|
"num_tokens": 10094997.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.461194086074829,
|
|
"epoch": 4.5940347970173985,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033719586428887125,
|
|
"loss": 4.9635,
|
|
"mean_token_accuracy": 0.19236611425876618,
|
|
"num_tokens": 10103585.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.276572418212891,
|
|
"epoch": 4.59817729908865,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033688869689744836,
|
|
"loss": 4.8715,
|
|
"mean_token_accuracy": 0.1956779330968857,
|
|
"num_tokens": 10113877.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.333914184570313,
|
|
"epoch": 4.602319801159901,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003365814046697608,
|
|
"loss": 4.9676,
|
|
"mean_token_accuracy": 0.1832288920879364,
|
|
"num_tokens": 10123127.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.39079098701477,
|
|
"epoch": 4.606462303231152,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003362739882256504,
|
|
"loss": 5.0347,
|
|
"mean_token_accuracy": 0.186669659614563,
|
|
"num_tokens": 10132301.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.369095230102539,
|
|
"epoch": 4.610604805302403,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003359664481852099,
|
|
"loss": 5.0411,
|
|
"mean_token_accuracy": 0.18307175636291503,
|
|
"num_tokens": 10140985.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.309534072875977,
|
|
"epoch": 4.614747307373654,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033565878516878103,
|
|
"loss": 4.9435,
|
|
"mean_token_accuracy": 0.186006298661232,
|
|
"num_tokens": 10149668.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.324404716491699,
|
|
"epoch": 4.618889809444905,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003353509997969537,
|
|
"loss": 4.9982,
|
|
"mean_token_accuracy": 0.1883559986948967,
|
|
"num_tokens": 10158584.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.366722822189331,
|
|
"epoch": 4.623032311516155,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033504309269056484,
|
|
"loss": 4.968,
|
|
"mean_token_accuracy": 0.19045166224241256,
|
|
"num_tokens": 10167657.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.274179601669312,
|
|
"epoch": 4.6271748135874065,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003347350644706965,
|
|
"loss": 4.8462,
|
|
"mean_token_accuracy": 0.1975993499159813,
|
|
"num_tokens": 10176657.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.3866353034973145,
|
|
"epoch": 4.631317315658658,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003344269157586754,
|
|
"loss": 4.9669,
|
|
"mean_token_accuracy": 0.19621438384056092,
|
|
"num_tokens": 10185132.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.304204607009888,
|
|
"epoch": 4.635459817729909,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033411864717607114,
|
|
"loss": 4.8497,
|
|
"mean_token_accuracy": 0.19386404007673264,
|
|
"num_tokens": 10193428.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.324200677871704,
|
|
"epoch": 4.63960231980116,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003338102593446951,
|
|
"loss": 5.0163,
|
|
"mean_token_accuracy": 0.18838184028863908,
|
|
"num_tokens": 10202549.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.341615533828735,
|
|
"epoch": 4.643744821872411,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00033350175288659925,
|
|
"loss": 4.8885,
|
|
"mean_token_accuracy": 0.1946927636861801,
|
|
"num_tokens": 10211161.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.309947538375854,
|
|
"epoch": 4.647887323943662,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033319312842407483,
|
|
"loss": 4.9923,
|
|
"mean_token_accuracy": 0.18832400739192962,
|
|
"num_tokens": 10220988.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.2807288646698,
|
|
"epoch": 4.652029826014913,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003328843865796511,
|
|
"loss": 4.8873,
|
|
"mean_token_accuracy": 0.19748056083917617,
|
|
"num_tokens": 10229383.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.421126461029052,
|
|
"epoch": 4.656172328086164,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033257552797609426,
|
|
"loss": 4.9686,
|
|
"mean_token_accuracy": 0.18525946289300918,
|
|
"num_tokens": 10238622.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.506065034866333,
|
|
"epoch": 4.660314830157415,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033226655323640553,
|
|
"loss": 5.0677,
|
|
"mean_token_accuracy": 0.18611674308776854,
|
|
"num_tokens": 10248201.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.445924186706543,
|
|
"epoch": 4.664457332228666,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000331957462983821,
|
|
"loss": 5.0837,
|
|
"mean_token_accuracy": 0.17612475752830506,
|
|
"num_tokens": 10258239.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.377671623229981,
|
|
"epoch": 4.668599834299917,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003316482578418094,
|
|
"loss": 4.9675,
|
|
"mean_token_accuracy": 0.1935265153646469,
|
|
"num_tokens": 10266915.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.348722076416015,
|
|
"epoch": 4.672742336371169,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033133893843407126,
|
|
"loss": 5.0783,
|
|
"mean_token_accuracy": 0.18859633952379226,
|
|
"num_tokens": 10276826.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.327623510360718,
|
|
"epoch": 4.676884838442419,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00033102950538453776,
|
|
"loss": 4.9912,
|
|
"mean_token_accuracy": 0.19163501411676406,
|
|
"num_tokens": 10286041.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.367205619812012,
|
|
"epoch": 4.68102734051367,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033071995931736897,
|
|
"loss": 5.0151,
|
|
"mean_token_accuracy": 0.19023409783840178,
|
|
"num_tokens": 10295258.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.404719638824463,
|
|
"epoch": 4.685169842584921,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00033041030085695323,
|
|
"loss": 4.9423,
|
|
"mean_token_accuracy": 0.19237637221813203,
|
|
"num_tokens": 10304726.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.34141435623169,
|
|
"epoch": 4.689312344656172,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003301005306279055,
|
|
"loss": 4.8745,
|
|
"mean_token_accuracy": 0.2018929526209831,
|
|
"num_tokens": 10312912.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.300086307525635,
|
|
"epoch": 4.693454846727423,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032979064925506627,
|
|
"loss": 4.9743,
|
|
"mean_token_accuracy": 0.18896473944187164,
|
|
"num_tokens": 10322046.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.290521335601807,
|
|
"epoch": 4.697597348798674,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003294806573634999,
|
|
"loss": 4.9263,
|
|
"mean_token_accuracy": 0.206088125705719,
|
|
"num_tokens": 10331375.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.353716802597046,
|
|
"epoch": 4.7017398508699255,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003291705555784942,
|
|
"loss": 5.0624,
|
|
"mean_token_accuracy": 0.18850265592336654,
|
|
"num_tokens": 10341465.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.466339874267578,
|
|
"epoch": 4.705882352941177,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00032886034452555814,
|
|
"loss": 5.052,
|
|
"mean_token_accuracy": 0.18991907685995102,
|
|
"num_tokens": 10350137.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.3118266582489015,
|
|
"epoch": 4.710024855012428,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003285500248304214,
|
|
"loss": 4.9502,
|
|
"mean_token_accuracy": 0.1877942368388176,
|
|
"num_tokens": 10358666.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.319189500808716,
|
|
"epoch": 4.714167357083679,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003282395971190328,
|
|
"loss": 4.8375,
|
|
"mean_token_accuracy": 0.19829211980104447,
|
|
"num_tokens": 10366910.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.377291250228882,
|
|
"epoch": 4.71830985915493,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032792906201755876,
|
|
"loss": 5.0053,
|
|
"mean_token_accuracy": 0.19080002307891847,
|
|
"num_tokens": 10376435.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.293179368972778,
|
|
"epoch": 4.72245236122618,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00032761842015238256,
|
|
"loss": 4.8965,
|
|
"mean_token_accuracy": 0.2024861231446266,
|
|
"num_tokens": 10384670.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.261117172241211,
|
|
"epoch": 4.726594863297431,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003273076721501029,
|
|
"loss": 4.8845,
|
|
"mean_token_accuracy": 0.1939824104309082,
|
|
"num_tokens": 10393731.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.386179351806641,
|
|
"epoch": 4.730737365368682,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003269968186375325,
|
|
"loss": 4.9916,
|
|
"mean_token_accuracy": 0.19829274713993073,
|
|
"num_tokens": 10402369.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.379951667785645,
|
|
"epoch": 4.7348798674399335,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003266858602416967,
|
|
"loss": 4.9801,
|
|
"mean_token_accuracy": 0.1936841771006584,
|
|
"num_tokens": 10411566.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.384341192245484,
|
|
"epoch": 4.739022369511185,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032637479758983255,
|
|
"loss": 5.0294,
|
|
"mean_token_accuracy": 0.18491634279489516,
|
|
"num_tokens": 10420086.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.409085655212403,
|
|
"epoch": 4.743164871582436,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032606363130938766,
|
|
"loss": 5.0704,
|
|
"mean_token_accuracy": 0.18897949457168578,
|
|
"num_tokens": 10429746.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.349277639389038,
|
|
"epoch": 4.747307373653687,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003257523620280181,
|
|
"loss": 4.9498,
|
|
"mean_token_accuracy": 0.19129680693149567,
|
|
"num_tokens": 10438536.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.348873805999756,
|
|
"epoch": 4.751449875724938,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00032544099037358834,
|
|
"loss": 4.9497,
|
|
"mean_token_accuracy": 0.19692206382751465,
|
|
"num_tokens": 10447471.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.33362455368042,
|
|
"epoch": 4.755592377796189,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003251295169741688,
|
|
"loss": 4.962,
|
|
"mean_token_accuracy": 0.1995266228914261,
|
|
"num_tokens": 10457179.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.327010250091552,
|
|
"epoch": 4.75973487986744,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00032481794245803536,
|
|
"loss": 5.0295,
|
|
"mean_token_accuracy": 0.18461760133504868,
|
|
"num_tokens": 10466291.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.323912620544434,
|
|
"epoch": 4.763877381938691,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032450626745366806,
|
|
"loss": 4.9102,
|
|
"mean_token_accuracy": 0.19402227252721788,
|
|
"num_tokens": 10475452.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.306818199157715,
|
|
"epoch": 4.768019884009942,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032419449258974924,
|
|
"loss": 4.8753,
|
|
"mean_token_accuracy": 0.20045212656259537,
|
|
"num_tokens": 10484381.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.342190933227539,
|
|
"epoch": 4.7721623860811935,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000323882618495163,
|
|
"loss": 5.0665,
|
|
"mean_token_accuracy": 0.1831491023302078,
|
|
"num_tokens": 10495132.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.351349258422852,
|
|
"epoch": 4.776304888152444,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003235706457989934,
|
|
"loss": 5.0246,
|
|
"mean_token_accuracy": 0.1850237250328064,
|
|
"num_tokens": 10504519.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.387107896804809,
|
|
"epoch": 4.780447390223695,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032325857513052354,
|
|
"loss": 4.9654,
|
|
"mean_token_accuracy": 0.19212980419397355,
|
|
"num_tokens": 10513575.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.396086692810059,
|
|
"epoch": 4.784589892294946,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000322946407119234,
|
|
"loss": 5.0554,
|
|
"mean_token_accuracy": 0.18998173326253892,
|
|
"num_tokens": 10522148.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.361077404022216,
|
|
"epoch": 4.788732394366197,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032263414239480187,
|
|
"loss": 4.9891,
|
|
"mean_token_accuracy": 0.18589809089899062,
|
|
"num_tokens": 10531183.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.342947483062744,
|
|
"epoch": 4.792874896437448,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000322321781587099,
|
|
"loss": 4.9298,
|
|
"mean_token_accuracy": 0.19391343146562576,
|
|
"num_tokens": 10540677.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.276329088211059,
|
|
"epoch": 4.797017398508699,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003220093253261915,
|
|
"loss": 4.9342,
|
|
"mean_token_accuracy": 0.18589555770158767,
|
|
"num_tokens": 10550009.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.315787076950073,
|
|
"epoch": 4.80115990057995,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003216967742423377,
|
|
"loss": 4.9621,
|
|
"mean_token_accuracy": 0.19312297254800798,
|
|
"num_tokens": 10559694.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.339448022842407,
|
|
"epoch": 4.8053024026512015,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003213841289659872,
|
|
"loss": 4.9613,
|
|
"mean_token_accuracy": 0.20080769211053848,
|
|
"num_tokens": 10567965.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.312955284118653,
|
|
"epoch": 4.809444904722453,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003210713901277799,
|
|
"loss": 4.9597,
|
|
"mean_token_accuracy": 0.19690640568733214,
|
|
"num_tokens": 10577653.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.359428977966308,
|
|
"epoch": 4.813587406793704,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000320758558358544,
|
|
"loss": 4.9728,
|
|
"mean_token_accuracy": 0.18951950818300248,
|
|
"num_tokens": 10586903.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.302858638763428,
|
|
"epoch": 4.817729908864955,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003204456342892955,
|
|
"loss": 4.8829,
|
|
"mean_token_accuracy": 0.19500194638967513,
|
|
"num_tokens": 10595411.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.321697616577149,
|
|
"epoch": 4.821872410936205,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032013261855123637,
|
|
"loss": 4.9475,
|
|
"mean_token_accuracy": 0.1907385230064392,
|
|
"num_tokens": 10604147.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.368891286849975,
|
|
"epoch": 4.826014913007457,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00031981951177575366,
|
|
"loss": 5.0445,
|
|
"mean_token_accuracy": 0.1850367307662964,
|
|
"num_tokens": 10614048.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.374803781509399,
|
|
"epoch": 4.830157415078707,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003195063145944178,
|
|
"loss": 5.0779,
|
|
"mean_token_accuracy": 0.18388603180646895,
|
|
"num_tokens": 10623399.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.389354228973389,
|
|
"epoch": 4.834299917149958,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003191930276389819,
|
|
"loss": 4.9183,
|
|
"mean_token_accuracy": 0.19506800919771194,
|
|
"num_tokens": 10632627.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.401239061355591,
|
|
"epoch": 4.838442419221209,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00031887965154137987,
|
|
"loss": 4.9726,
|
|
"mean_token_accuracy": 0.1816769301891327,
|
|
"num_tokens": 10641155.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.354168939590454,
|
|
"epoch": 4.8425849212924605,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003185661869337255,
|
|
"loss": 5.0118,
|
|
"mean_token_accuracy": 0.18452974855899812,
|
|
"num_tokens": 10649661.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.265790605545044,
|
|
"epoch": 4.846727423363712,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031825263444831137,
|
|
"loss": 4.9151,
|
|
"mean_token_accuracy": 0.19235364347696304,
|
|
"num_tokens": 10658652.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.3790277481079105,
|
|
"epoch": 4.850869925434963,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003179389947176069,
|
|
"loss": 5.0445,
|
|
"mean_token_accuracy": 0.1886226385831833,
|
|
"num_tokens": 10668703.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.385872554779053,
|
|
"epoch": 4.855012427506214,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003176252683742578,
|
|
"loss": 4.9484,
|
|
"mean_token_accuracy": 0.19056980460882186,
|
|
"num_tokens": 10677740.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.347926044464112,
|
|
"epoch": 4.859154929577465,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003173114560510843,
|
|
"loss": 4.9447,
|
|
"mean_token_accuracy": 0.18929830342531204,
|
|
"num_tokens": 10686429.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.332098054885864,
|
|
"epoch": 4.863297431648716,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031699755838108027,
|
|
"loss": 5.0442,
|
|
"mean_token_accuracy": 0.1941297546029091,
|
|
"num_tokens": 10696367.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.310936546325683,
|
|
"epoch": 4.867439933719967,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003166835759974115,
|
|
"loss": 4.9902,
|
|
"mean_token_accuracy": 0.1873769447207451,
|
|
"num_tokens": 10705712.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.371630382537842,
|
|
"epoch": 4.871582435791218,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031636950953341486,
|
|
"loss": 5.0345,
|
|
"mean_token_accuracy": 0.18638068735599517,
|
|
"num_tokens": 10715897.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.3334348678588865,
|
|
"epoch": 4.8757249378624685,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00031605535962259666,
|
|
"loss": 4.9268,
|
|
"mean_token_accuracy": 0.1887778237462044,
|
|
"num_tokens": 10724810.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.353308916091919,
|
|
"epoch": 4.87986743993372,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003157411268986316,
|
|
"loss": 4.9771,
|
|
"mean_token_accuracy": 0.1932398796081543,
|
|
"num_tokens": 10734661.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.287876415252685,
|
|
"epoch": 4.884009942004971,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003154268119953615,
|
|
"loss": 4.9586,
|
|
"mean_token_accuracy": 0.19356854259967804,
|
|
"num_tokens": 10743642.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.344345998764038,
|
|
"epoch": 4.888152444076222,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003151124155467938,
|
|
"loss": 4.9607,
|
|
"mean_token_accuracy": 0.19625648856163025,
|
|
"num_tokens": 10752812.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.3407337188720705,
|
|
"epoch": 4.892294946147473,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00031479793818710056,
|
|
"loss": 4.9983,
|
|
"mean_token_accuracy": 0.18306475579738618,
|
|
"num_tokens": 10762117.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.299752902984619,
|
|
"epoch": 4.896437448218724,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031448338055061685,
|
|
"loss": 4.9018,
|
|
"mean_token_accuracy": 0.19901909530162812,
|
|
"num_tokens": 10771515.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.309285831451416,
|
|
"epoch": 4.900579950289975,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031416874327183994,
|
|
"loss": 4.9486,
|
|
"mean_token_accuracy": 0.19302618503570557,
|
|
"num_tokens": 10780610.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.22410569190979,
|
|
"epoch": 4.904722452361226,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031385402698542753,
|
|
"loss": 4.9151,
|
|
"mean_token_accuracy": 0.19276074767112733,
|
|
"num_tokens": 10789169.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.30161485671997,
|
|
"epoch": 4.908864954432477,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00031353923232619674,
|
|
"loss": 4.9078,
|
|
"mean_token_accuracy": 0.19779694378376006,
|
|
"num_tokens": 10798636.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.3571271896362305,
|
|
"epoch": 4.9130074565037285,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003132243599291229,
|
|
"loss": 4.9753,
|
|
"mean_token_accuracy": 0.19644105434417725,
|
|
"num_tokens": 10808245.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.384200191497802,
|
|
"epoch": 4.91714995857498,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00031290941042933797,
|
|
"loss": 4.9679,
|
|
"mean_token_accuracy": 0.18787674605846405,
|
|
"num_tokens": 10817164.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.287544202804566,
|
|
"epoch": 4.92129246064623,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00031259438446212953,
|
|
"loss": 5.0007,
|
|
"mean_token_accuracy": 0.18373542726039888,
|
|
"num_tokens": 10826098.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.33493857383728,
|
|
"epoch": 4.925434962717482,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031227928266293933,
|
|
"loss": 5.0115,
|
|
"mean_token_accuracy": 0.18100205808877945,
|
|
"num_tokens": 10836427.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.345344829559326,
|
|
"epoch": 4.929577464788732,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00031196410566736215,
|
|
"loss": 4.9689,
|
|
"mean_token_accuracy": 0.19277302771806717,
|
|
"num_tokens": 10845804.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.3519449710845945,
|
|
"epoch": 4.933719966859983,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031164885411114453,
|
|
"loss": 4.946,
|
|
"mean_token_accuracy": 0.1908810630440712,
|
|
"num_tokens": 10855951.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.301464605331421,
|
|
"epoch": 4.937862468931234,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00031133352863018315,
|
|
"loss": 5.079,
|
|
"mean_token_accuracy": 0.18923453092575074,
|
|
"num_tokens": 10865528.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.369193601608276,
|
|
"epoch": 4.942004971002485,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031101812986052406,
|
|
"loss": 4.9936,
|
|
"mean_token_accuracy": 0.19839225262403487,
|
|
"num_tokens": 10874365.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.310976552963257,
|
|
"epoch": 4.9461474730737365,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00031070265843836107,
|
|
"loss": 4.9184,
|
|
"mean_token_accuracy": 0.1972976267337799,
|
|
"num_tokens": 10883034.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.323369884490967,
|
|
"epoch": 4.950289975144988,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003103871150000345,
|
|
"loss": 4.9768,
|
|
"mean_token_accuracy": 0.1961338758468628,
|
|
"num_tokens": 10892307.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.354686069488525,
|
|
"epoch": 4.954432477216239,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00031007150018202983,
|
|
"loss": 5.057,
|
|
"mean_token_accuracy": 0.18536237627267838,
|
|
"num_tokens": 10902074.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.372721862792969,
|
|
"epoch": 4.95857497928749,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030975581462097673,
|
|
"loss": 4.8961,
|
|
"mean_token_accuracy": 0.19350670725107194,
|
|
"num_tokens": 10911205.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.278085422515869,
|
|
"epoch": 4.962717481358741,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00030944005895364754,
|
|
"loss": 4.9233,
|
|
"mean_token_accuracy": 0.19045489132404328,
|
|
"num_tokens": 10920371.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.219862222671509,
|
|
"epoch": 4.966859983429992,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00030912423381695584,
|
|
"loss": 4.9124,
|
|
"mean_token_accuracy": 0.20015972554683686,
|
|
"num_tokens": 10928741.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.3216063499450685,
|
|
"epoch": 4.971002485501243,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00030880833984795554,
|
|
"loss": 4.9251,
|
|
"mean_token_accuracy": 0.20140086114406586,
|
|
"num_tokens": 10938074.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 4.971002485501243,
|
|
"eval_entropy": 5.223882782942196,
|
|
"eval_loss": 5.772702217102051,
|
|
"eval_mean_token_accuracy": 0.16291698060437307,
|
|
"eval_num_tokens": 10938074.0,
|
|
"eval_runtime": 2.7438,
|
|
"eval_samples_per_second": 1365.624,
|
|
"eval_steps_per_second": 170.931,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.304221105575562,
|
|
"epoch": 4.975144987572493,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003084923776838392,
|
|
"loss": 4.9428,
|
|
"mean_token_accuracy": 0.1906152918934822,
|
|
"num_tokens": 10946756.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.28564419746399,
|
|
"epoch": 4.9792874896437445,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00030817634796193705,
|
|
"loss": 4.9192,
|
|
"mean_token_accuracy": 0.19711493700742722,
|
|
"num_tokens": 10956313.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.420253419876099,
|
|
"epoch": 4.983429991714996,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00030786025131971555,
|
|
"loss": 4.9823,
|
|
"mean_token_accuracy": 0.19457563310861586,
|
|
"num_tokens": 10964333.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.329161500930786,
|
|
"epoch": 4.987572493786247,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003075440883947764,
|
|
"loss": 4.8999,
|
|
"mean_token_accuracy": 0.19664130955934525,
|
|
"num_tokens": 10974110.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.248602771759034,
|
|
"epoch": 4.991714995857498,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003072278598248544,
|
|
"loss": 4.9529,
|
|
"mean_token_accuracy": 0.19656815081834794,
|
|
"num_tokens": 10983186.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.307998704910278,
|
|
"epoch": 4.995857497928749,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00030691156624781734,
|
|
"loss": 4.9377,
|
|
"mean_token_accuracy": 0.19932860583066941,
|
|
"num_tokens": 10992244.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.353039836883545,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.000306595208301664,
|
|
"loss": 4.927,
|
|
"mean_token_accuracy": 0.20540354549884796,
|
|
"num_tokens": 11000005.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.336960649490356,
|
|
"epoch": 5.004142502071251,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00030627878662452273,
|
|
"loss": 4.7717,
|
|
"mean_token_accuracy": 0.20522954761981965,
|
|
"num_tokens": 11009346.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.356115531921387,
|
|
"epoch": 5.008285004142502,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003059623018546508,
|
|
"loss": 4.7197,
|
|
"mean_token_accuracy": 0.2092713624238968,
|
|
"num_tokens": 11017844.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.362998294830322,
|
|
"epoch": 5.012427506213753,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003056457546304326,
|
|
"loss": 4.7531,
|
|
"mean_token_accuracy": 0.21363187432289124,
|
|
"num_tokens": 11026840.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.250718975067139,
|
|
"epoch": 5.016570008285004,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003053291455903782,
|
|
"loss": 4.7186,
|
|
"mean_token_accuracy": 0.2076275572180748,
|
|
"num_tokens": 11034906.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.263190746307373,
|
|
"epoch": 5.0207125103562555,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00030501247537312293,
|
|
"loss": 4.7802,
|
|
"mean_token_accuracy": 0.20651286244392394,
|
|
"num_tokens": 11044403.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.360515308380127,
|
|
"epoch": 5.024855012427507,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003046957446174251,
|
|
"loss": 4.7387,
|
|
"mean_token_accuracy": 0.21264788061380385,
|
|
"num_tokens": 11052685.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.316996622085571,
|
|
"epoch": 5.028997514498757,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00030437895396216524,
|
|
"loss": 4.591,
|
|
"mean_token_accuracy": 0.21709324419498444,
|
|
"num_tokens": 11060830.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.315813970565796,
|
|
"epoch": 5.033140016570008,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00030406210404634475,
|
|
"loss": 4.7708,
|
|
"mean_token_accuracy": 0.20373892933130264,
|
|
"num_tokens": 11070220.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.294347095489502,
|
|
"epoch": 5.037282518641259,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003037451955090846,
|
|
"loss": 4.6847,
|
|
"mean_token_accuracy": 0.20420003384351731,
|
|
"num_tokens": 11079171.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.297221279144287,
|
|
"epoch": 5.04142502071251,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030342822898962377,
|
|
"loss": 4.7697,
|
|
"mean_token_accuracy": 0.20524269640445708,
|
|
"num_tokens": 11089205.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.273457336425781,
|
|
"epoch": 5.045567522783761,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003031112051273185,
|
|
"loss": 4.7429,
|
|
"mean_token_accuracy": 0.21156702935695648,
|
|
"num_tokens": 11098339.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.299424123764038,
|
|
"epoch": 5.049710024855012,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003027941245616405,
|
|
"loss": 4.6524,
|
|
"mean_token_accuracy": 0.21203886568546296,
|
|
"num_tokens": 11107218.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.3286974906921385,
|
|
"epoch": 5.0538525269262635,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003024769879321761,
|
|
"loss": 4.7702,
|
|
"mean_token_accuracy": 0.2051073744893074,
|
|
"num_tokens": 11116457.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.300696182250976,
|
|
"epoch": 5.057995028997515,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003021597958786243,
|
|
"loss": 4.7449,
|
|
"mean_token_accuracy": 0.20762935131788254,
|
|
"num_tokens": 11124834.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.288244676589966,
|
|
"epoch": 5.062137531068766,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003018425490407962,
|
|
"loss": 4.76,
|
|
"mean_token_accuracy": 0.20868725776672364,
|
|
"num_tokens": 11133920.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.387529897689819,
|
|
"epoch": 5.066280033140017,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030152524805861344,
|
|
"loss": 4.7597,
|
|
"mean_token_accuracy": 0.20800695270299913,
|
|
"num_tokens": 11143573.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.3629701137542725,
|
|
"epoch": 5.070422535211268,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00030120789357210667,
|
|
"loss": 4.8139,
|
|
"mean_token_accuracy": 0.20669919848442078,
|
|
"num_tokens": 11154255.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.300123262405395,
|
|
"epoch": 5.074565037282519,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00030089048622141453,
|
|
"loss": 4.7241,
|
|
"mean_token_accuracy": 0.2095050945878029,
|
|
"num_tokens": 11163306.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.368470001220703,
|
|
"epoch": 5.078707539353769,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00030057302664678243,
|
|
"loss": 4.7569,
|
|
"mean_token_accuracy": 0.20853464156389237,
|
|
"num_tokens": 11172763.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.33392744064331,
|
|
"epoch": 5.08285004142502,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00030025551548856105,
|
|
"loss": 4.7413,
|
|
"mean_token_accuracy": 0.20951523035764694,
|
|
"num_tokens": 11182196.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.287980794906616,
|
|
"epoch": 5.0869925434962715,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002999379533872051,
|
|
"loss": 4.7905,
|
|
"mean_token_accuracy": 0.19804110527038574,
|
|
"num_tokens": 11192462.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.281665325164795,
|
|
"epoch": 5.091135045567523,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002996203409832719,
|
|
"loss": 4.7519,
|
|
"mean_token_accuracy": 0.2107569918036461,
|
|
"num_tokens": 11201805.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.31660532951355,
|
|
"epoch": 5.095277547638774,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002993026789174205,
|
|
"loss": 4.7322,
|
|
"mean_token_accuracy": 0.21754502654075622,
|
|
"num_tokens": 11210114.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.335706996917724,
|
|
"epoch": 5.099420049710025,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002989849678304099,
|
|
"loss": 4.8147,
|
|
"mean_token_accuracy": 0.20241696387529373,
|
|
"num_tokens": 11219725.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.308135366439819,
|
|
"epoch": 5.103562551781276,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00029866720836309817,
|
|
"loss": 4.7283,
|
|
"mean_token_accuracy": 0.20736500322818757,
|
|
"num_tokens": 11228583.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.3292882442474365,
|
|
"epoch": 5.107705053852527,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029834940115644086,
|
|
"loss": 4.7472,
|
|
"mean_token_accuracy": 0.20635622441768647,
|
|
"num_tokens": 11237787.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.2917111873626705,
|
|
"epoch": 5.111847555923778,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002980315468514898,
|
|
"loss": 4.7704,
|
|
"mean_token_accuracy": 0.20057149082422257,
|
|
"num_tokens": 11247136.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.239889097213745,
|
|
"epoch": 5.115990057995029,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00029771364608939196,
|
|
"loss": 4.6994,
|
|
"mean_token_accuracy": 0.20981708765029908,
|
|
"num_tokens": 11256931.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.272354030609131,
|
|
"epoch": 5.12013256006628,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002973956995113878,
|
|
"loss": 4.6811,
|
|
"mean_token_accuracy": 0.20489206910133362,
|
|
"num_tokens": 11266457.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.272558641433716,
|
|
"epoch": 5.1242750621375315,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002970777077588103,
|
|
"loss": 4.6443,
|
|
"mean_token_accuracy": 0.2170465260744095,
|
|
"num_tokens": 11274239.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.282349872589111,
|
|
"epoch": 5.128417564208782,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002967596714730837,
|
|
"loss": 4.7752,
|
|
"mean_token_accuracy": 0.20363858938217164,
|
|
"num_tokens": 11283239.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.285331869125367,
|
|
"epoch": 5.132560066280033,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00029644159129572185,
|
|
"loss": 4.7121,
|
|
"mean_token_accuracy": 0.2075791761279106,
|
|
"num_tokens": 11291662.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.2614960193634035,
|
|
"epoch": 5.136702568351284,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002961234678683272,
|
|
"loss": 4.7453,
|
|
"mean_token_accuracy": 0.21309943348169327,
|
|
"num_tokens": 11300165.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.323043155670166,
|
|
"epoch": 5.140845070422535,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002958053018325896,
|
|
"loss": 4.7132,
|
|
"mean_token_accuracy": 0.21612811386585234,
|
|
"num_tokens": 11309729.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.375712585449219,
|
|
"epoch": 5.144987572493786,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002954870938302846,
|
|
"loss": 4.7657,
|
|
"mean_token_accuracy": 0.2033802881836891,
|
|
"num_tokens": 11318674.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.292187595367432,
|
|
"epoch": 5.149130074565037,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002951688445032726,
|
|
"loss": 4.7309,
|
|
"mean_token_accuracy": 0.21144932359457017,
|
|
"num_tokens": 11327583.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.326848030090332,
|
|
"epoch": 5.153272576636288,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002948505544934971,
|
|
"loss": 4.7639,
|
|
"mean_token_accuracy": 0.2049189493060112,
|
|
"num_tokens": 11338132.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.31851224899292,
|
|
"epoch": 5.1574150787075395,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000294532224442984,
|
|
"loss": 4.7911,
|
|
"mean_token_accuracy": 0.20397177040576936,
|
|
"num_tokens": 11347210.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.275168132781983,
|
|
"epoch": 5.161557580778791,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00029421385499383976,
|
|
"loss": 4.6742,
|
|
"mean_token_accuracy": 0.22349916100502015,
|
|
"num_tokens": 11355479.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.348278999328613,
|
|
"epoch": 5.165700082850042,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029389544678825035,
|
|
"loss": 4.8403,
|
|
"mean_token_accuracy": 0.19372698813676834,
|
|
"num_tokens": 11364518.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.247139596939087,
|
|
"epoch": 5.169842584921293,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00029357700046847985,
|
|
"loss": 4.737,
|
|
"mean_token_accuracy": 0.2046540141105652,
|
|
"num_tokens": 11373589.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.368101263046265,
|
|
"epoch": 5.173985086992544,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002932585166768694,
|
|
"loss": 4.7938,
|
|
"mean_token_accuracy": 0.20394597351551055,
|
|
"num_tokens": 11383209.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.315500497817993,
|
|
"epoch": 5.178127589063794,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00029293999605583545,
|
|
"loss": 4.7294,
|
|
"mean_token_accuracy": 0.20444581657648087,
|
|
"num_tokens": 11392012.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.282355690002442,
|
|
"epoch": 5.182270091135045,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029262143924786905,
|
|
"loss": 4.7761,
|
|
"mean_token_accuracy": 0.20456164181232453,
|
|
"num_tokens": 11400489.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.2667319774627686,
|
|
"epoch": 5.186412593206296,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00029230284689553404,
|
|
"loss": 4.6817,
|
|
"mean_token_accuracy": 0.21175160706043245,
|
|
"num_tokens": 11409337.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.286681175231934,
|
|
"epoch": 5.190555095277547,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029198421964146605,
|
|
"loss": 4.7382,
|
|
"mean_token_accuracy": 0.21787525415420533,
|
|
"num_tokens": 11418650.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.2383523941040036,
|
|
"epoch": 5.1946975973487985,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002916655581283709,
|
|
"loss": 4.747,
|
|
"mean_token_accuracy": 0.20590767711400987,
|
|
"num_tokens": 11427516.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.225810289382935,
|
|
"epoch": 5.19884009942005,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00029134686299902397,
|
|
"loss": 4.7223,
|
|
"mean_token_accuracy": 0.20668105334043502,
|
|
"num_tokens": 11435676.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.336339044570923,
|
|
"epoch": 5.202982601491301,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002910281348962679,
|
|
"loss": 4.7825,
|
|
"mean_token_accuracy": 0.1937339723110199,
|
|
"num_tokens": 11444520.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.342389440536499,
|
|
"epoch": 5.207125103562552,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00029070937446301225,
|
|
"loss": 4.7611,
|
|
"mean_token_accuracy": 0.20350439995527267,
|
|
"num_tokens": 11453759.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.377904796600342,
|
|
"epoch": 5.211267605633803,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002903905823422317,
|
|
"loss": 4.9078,
|
|
"mean_token_accuracy": 0.1901518851518631,
|
|
"num_tokens": 11463409.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.323945093154907,
|
|
"epoch": 5.215410107705054,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002900717591769646,
|
|
"loss": 4.7966,
|
|
"mean_token_accuracy": 0.20510322898626326,
|
|
"num_tokens": 11473440.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.27484130859375,
|
|
"epoch": 5.219552609776305,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002897529056103122,
|
|
"loss": 4.6336,
|
|
"mean_token_accuracy": 0.2183629482984543,
|
|
"num_tokens": 11482353.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.297479438781738,
|
|
"epoch": 5.223695111847556,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002894340222854371,
|
|
"loss": 4.7712,
|
|
"mean_token_accuracy": 0.20640946179628372,
|
|
"num_tokens": 11491331.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.215849542617798,
|
|
"epoch": 5.2278376139188065,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002891151098455617,
|
|
"loss": 4.7107,
|
|
"mean_token_accuracy": 0.20350281596183778,
|
|
"num_tokens": 11500553.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.170541191101075,
|
|
"epoch": 5.231980115990058,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002887961689339672,
|
|
"loss": 4.6526,
|
|
"mean_token_accuracy": 0.21051572561264037,
|
|
"num_tokens": 11509146.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.332199335098267,
|
|
"epoch": 5.236122618061309,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002884772001939924,
|
|
"loss": 4.7771,
|
|
"mean_token_accuracy": 0.20190449655056,
|
|
"num_tokens": 11518172.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.30375485420227,
|
|
"epoch": 5.24026512013256,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000288158204269032,
|
|
"loss": 4.7764,
|
|
"mean_token_accuracy": 0.19927115142345428,
|
|
"num_tokens": 11527291.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.285002279281616,
|
|
"epoch": 5.244407622203811,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00028783918180253567,
|
|
"loss": 4.7763,
|
|
"mean_token_accuracy": 0.20877851247787477,
|
|
"num_tokens": 11536306.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.325639486312866,
|
|
"epoch": 5.248550124275062,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002875201334380066,
|
|
"loss": 4.801,
|
|
"mean_token_accuracy": 0.20768691897392272,
|
|
"num_tokens": 11545879.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.260626125335693,
|
|
"epoch": 5.252692626346313,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002872010598190002,
|
|
"loss": 4.7203,
|
|
"mean_token_accuracy": 0.21331494748592378,
|
|
"num_tokens": 11555067.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.273142910003662,
|
|
"epoch": 5.256835128417564,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002868819615891229,
|
|
"loss": 4.699,
|
|
"mean_token_accuracy": 0.211579430103302,
|
|
"num_tokens": 11563419.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.256633758544922,
|
|
"epoch": 5.260977630488815,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002865628393920305,
|
|
"loss": 4.7696,
|
|
"mean_token_accuracy": 0.207663132250309,
|
|
"num_tokens": 11573018.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.237578821182251,
|
|
"epoch": 5.2651201325600665,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002862436938714276,
|
|
"loss": 4.7003,
|
|
"mean_token_accuracy": 0.20862672179937364,
|
|
"num_tokens": 11581996.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.355519008636475,
|
|
"epoch": 5.269262634631318,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002859245256710654,
|
|
"loss": 4.7534,
|
|
"mean_token_accuracy": 0.20775581747293473,
|
|
"num_tokens": 11591706.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.319948673248291,
|
|
"epoch": 5.273405136702569,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002856053354347411,
|
|
"loss": 4.7813,
|
|
"mean_token_accuracy": 0.2074725329875946,
|
|
"num_tokens": 11600616.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.243834018707275,
|
|
"epoch": 5.27754763877382,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002852861238062963,
|
|
"loss": 4.6584,
|
|
"mean_token_accuracy": 0.21568063348531724,
|
|
"num_tokens": 11609650.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.26465950012207,
|
|
"epoch": 5.28169014084507,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002849668914296157,
|
|
"loss": 4.6998,
|
|
"mean_token_accuracy": 0.21142313480377198,
|
|
"num_tokens": 11618785.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.305418300628662,
|
|
"epoch": 5.285832642916321,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000284647638948626,
|
|
"loss": 4.7576,
|
|
"mean_token_accuracy": 0.20556087791919708,
|
|
"num_tokens": 11628089.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.268257570266724,
|
|
"epoch": 5.289975144987572,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002843283670072942,
|
|
"loss": 4.8658,
|
|
"mean_token_accuracy": 0.19545172303915023,
|
|
"num_tokens": 11636954.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.275126600265503,
|
|
"epoch": 5.294117647058823,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002840090762496268,
|
|
"loss": 4.7803,
|
|
"mean_token_accuracy": 0.20498874336481093,
|
|
"num_tokens": 11646719.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.340854024887085,
|
|
"epoch": 5.2982601491300745,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002836897673196681,
|
|
"loss": 4.7015,
|
|
"mean_token_accuracy": 0.20719585418701172,
|
|
"num_tokens": 11655589.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.227226972579956,
|
|
"epoch": 5.302402651201326,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028337044086149913,
|
|
"loss": 4.7283,
|
|
"mean_token_accuracy": 0.20560554265975953,
|
|
"num_tokens": 11663703.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.296469211578369,
|
|
"epoch": 5.306545153272577,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00028305109751923616,
|
|
"loss": 4.8177,
|
|
"mean_token_accuracy": 0.20388006865978242,
|
|
"num_tokens": 11672643.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.270711994171142,
|
|
"epoch": 5.310687655343828,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00028273173793702967,
|
|
"loss": 4.673,
|
|
"mean_token_accuracy": 0.2078327417373657,
|
|
"num_tokens": 11681449.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.301759338378906,
|
|
"epoch": 5.314830157415079,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002824123627590629,
|
|
"loss": 4.7568,
|
|
"mean_token_accuracy": 0.19922166019678117,
|
|
"num_tokens": 11690820.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.291653823852539,
|
|
"epoch": 5.31897265948633,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002820929726295504,
|
|
"loss": 4.7858,
|
|
"mean_token_accuracy": 0.2068773627281189,
|
|
"num_tokens": 11699705.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.257635831832886,
|
|
"epoch": 5.323115161557581,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002817735681927371,
|
|
"loss": 4.7267,
|
|
"mean_token_accuracy": 0.21190525144338607,
|
|
"num_tokens": 11709417.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.305531692504883,
|
|
"epoch": 5.327257663628831,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028145415009289644,
|
|
"loss": 4.7662,
|
|
"mean_token_accuracy": 0.2035451725125313,
|
|
"num_tokens": 11718661.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.240819025039673,
|
|
"epoch": 5.3314001657000825,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002811347189743297,
|
|
"loss": 4.7448,
|
|
"mean_token_accuracy": 0.21049206256866454,
|
|
"num_tokens": 11727568.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.250747394561768,
|
|
"epoch": 5.335542667771334,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028081527548136446,
|
|
"loss": 4.6853,
|
|
"mean_token_accuracy": 0.2216053858399391,
|
|
"num_tokens": 11736615.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.348236608505249,
|
|
"epoch": 5.339685169842585,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002804958202583531,
|
|
"loss": 4.8181,
|
|
"mean_token_accuracy": 0.19986798763275146,
|
|
"num_tokens": 11747458.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.277511358261108,
|
|
"epoch": 5.343827671913836,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002801763539496718,
|
|
"loss": 4.7878,
|
|
"mean_token_accuracy": 0.2081502467393875,
|
|
"num_tokens": 11756279.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.279990530014038,
|
|
"epoch": 5.347970173985087,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002798568771997188,
|
|
"loss": 4.7942,
|
|
"mean_token_accuracy": 0.2071585923433304,
|
|
"num_tokens": 11765022.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.2339942932128904,
|
|
"epoch": 5.352112676056338,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002795373906529137,
|
|
"loss": 4.7376,
|
|
"mean_token_accuracy": 0.20560838282108307,
|
|
"num_tokens": 11774704.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.248803615570068,
|
|
"epoch": 5.356255178127589,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002792178949536959,
|
|
"loss": 4.6784,
|
|
"mean_token_accuracy": 0.21611350923776626,
|
|
"num_tokens": 11782309.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.28652811050415,
|
|
"epoch": 5.36039768019884,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000278898390746523,
|
|
"loss": 4.7467,
|
|
"mean_token_accuracy": 0.20095686316490174,
|
|
"num_tokens": 11792595.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.2047077178955075,
|
|
"epoch": 5.364540182270091,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00027857887867587005,
|
|
"loss": 4.7269,
|
|
"mean_token_accuracy": 0.21140239089727403,
|
|
"num_tokens": 11801536.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.271708583831787,
|
|
"epoch": 5.368682684341342,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002782593593862276,
|
|
"loss": 4.8229,
|
|
"mean_token_accuracy": 0.19779271334409715,
|
|
"num_tokens": 11810222.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.317666244506836,
|
|
"epoch": 5.3728251864125935,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002779398335221011,
|
|
"loss": 4.8066,
|
|
"mean_token_accuracy": 0.20721431225538253,
|
|
"num_tokens": 11819310.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.305458688735962,
|
|
"epoch": 5.376967688483845,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002776203017280092,
|
|
"loss": 4.7703,
|
|
"mean_token_accuracy": 0.19982181042432784,
|
|
"num_tokens": 11827916.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.217341518402099,
|
|
"epoch": 5.381110190555095,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027730076464848225,
|
|
"loss": 4.7654,
|
|
"mean_token_accuracy": 0.20862893313169478,
|
|
"num_tokens": 11836770.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.276216888427735,
|
|
"epoch": 5.385252692626346,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00027698122292806174,
|
|
"loss": 4.7006,
|
|
"mean_token_accuracy": 0.21164953708648682,
|
|
"num_tokens": 11845917.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.385252692626346,
|
|
"eval_entropy": 5.213927641097925,
|
|
"eval_loss": 5.769200801849365,
|
|
"eval_mean_token_accuracy": 0.16407465213524508,
|
|
"eval_num_tokens": 11845917.0,
|
|
"eval_runtime": 2.9494,
|
|
"eval_samples_per_second": 1270.426,
|
|
"eval_steps_per_second": 159.015,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.290333843231201,
|
|
"epoch": 5.389395194697597,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000276661677211298,
|
|
"loss": 4.7403,
|
|
"mean_token_accuracy": 0.20341738760471345,
|
|
"num_tokens": 11855118.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.278094148635864,
|
|
"epoch": 5.393537696768848,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002763421281427497,
|
|
"loss": 4.6794,
|
|
"mean_token_accuracy": 0.21257141679525376,
|
|
"num_tokens": 11863793.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.368717813491822,
|
|
"epoch": 5.397680198840099,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002760225763669823,
|
|
"loss": 4.8171,
|
|
"mean_token_accuracy": 0.197501078248024,
|
|
"num_tokens": 11873194.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.341726636886596,
|
|
"epoch": 5.40182270091135,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00027570302252856666,
|
|
"loss": 4.7747,
|
|
"mean_token_accuracy": 0.20519590526819229,
|
|
"num_tokens": 11882566.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.2676441192626955,
|
|
"epoch": 5.4059652029826015,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027538346727207767,
|
|
"loss": 4.7475,
|
|
"mean_token_accuracy": 0.21314287781715394,
|
|
"num_tokens": 11892267.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.272220420837402,
|
|
"epoch": 5.410107705053853,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002750639112420934,
|
|
"loss": 4.7716,
|
|
"mean_token_accuracy": 0.21124217808246612,
|
|
"num_tokens": 11901440.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.270874547958374,
|
|
"epoch": 5.414250207125104,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00027474435508319304,
|
|
"loss": 4.8396,
|
|
"mean_token_accuracy": 0.20556266158819197,
|
|
"num_tokens": 11911259.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.265343093872071,
|
|
"epoch": 5.418392709196355,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027442479943995645,
|
|
"loss": 4.7941,
|
|
"mean_token_accuracy": 0.2035036563873291,
|
|
"num_tokens": 11920467.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.283535575866699,
|
|
"epoch": 5.422535211267606,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002741052449569624,
|
|
"loss": 4.8089,
|
|
"mean_token_accuracy": 0.19865555614233016,
|
|
"num_tokens": 11929971.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.330445003509522,
|
|
"epoch": 5.426677713338857,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002737856922787869,
|
|
"loss": 4.8088,
|
|
"mean_token_accuracy": 0.19926221221685408,
|
|
"num_tokens": 11938585.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.301173543930053,
|
|
"epoch": 5.430820215410108,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00027346614205000277,
|
|
"loss": 4.7884,
|
|
"mean_token_accuracy": 0.20320066064596176,
|
|
"num_tokens": 11948264.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.329202175140381,
|
|
"epoch": 5.434962717481358,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027314659491517777,
|
|
"loss": 4.7827,
|
|
"mean_token_accuracy": 0.2079697236418724,
|
|
"num_tokens": 11958459.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.296626996994019,
|
|
"epoch": 5.4391052195526095,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002728270515188733,
|
|
"loss": 4.7728,
|
|
"mean_token_accuracy": 0.20527158677577972,
|
|
"num_tokens": 11967994.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.325187683105469,
|
|
"epoch": 5.443247721623861,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002725075125056435,
|
|
"loss": 4.8235,
|
|
"mean_token_accuracy": 0.21111733317375184,
|
|
"num_tokens": 11976564.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.296580266952515,
|
|
"epoch": 5.447390223695112,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002721879785200333,
|
|
"loss": 4.7541,
|
|
"mean_token_accuracy": 0.21026979386806488,
|
|
"num_tokens": 11985682.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.34701886177063,
|
|
"epoch": 5.451532725766363,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002718684502065777,
|
|
"loss": 4.8138,
|
|
"mean_token_accuracy": 0.20367322415113448,
|
|
"num_tokens": 11995098.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.3261590003967285,
|
|
"epoch": 5.455675227837614,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00027154892820980024,
|
|
"loss": 4.8096,
|
|
"mean_token_accuracy": 0.206008443236351,
|
|
"num_tokens": 12004538.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.262662792205811,
|
|
"epoch": 5.459817729908865,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00027122941317421175,
|
|
"loss": 4.741,
|
|
"mean_token_accuracy": 0.21086378544569015,
|
|
"num_tokens": 12012568.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.287980222702027,
|
|
"epoch": 5.463960231980116,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000270909905744309,
|
|
"loss": 4.776,
|
|
"mean_token_accuracy": 0.20741475820541383,
|
|
"num_tokens": 12021934.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.313226222991943,
|
|
"epoch": 5.468102734051367,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00027059040656457316,
|
|
"loss": 4.8104,
|
|
"mean_token_accuracy": 0.2010733589529991,
|
|
"num_tokens": 12031158.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.3314722061157225,
|
|
"epoch": 5.472245236122618,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002702709162794692,
|
|
"loss": 4.8078,
|
|
"mean_token_accuracy": 0.20088913142681122,
|
|
"num_tokens": 12040392.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.248980331420898,
|
|
"epoch": 5.4763877381938695,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026995143553344394,
|
|
"loss": 4.6481,
|
|
"mean_token_accuracy": 0.21935920119285585,
|
|
"num_tokens": 12048412.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.258455896377564,
|
|
"epoch": 5.48053024026512,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026963196497092483,
|
|
"loss": 4.7625,
|
|
"mean_token_accuracy": 0.20619691014289857,
|
|
"num_tokens": 12057681.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.290913534164429,
|
|
"epoch": 5.484672742336371,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00026931250523631905,
|
|
"loss": 4.7451,
|
|
"mean_token_accuracy": 0.20713003873825073,
|
|
"num_tokens": 12066243.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.215484523773194,
|
|
"epoch": 5.488815244407622,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026899305697401175,
|
|
"loss": 4.7545,
|
|
"mean_token_accuracy": 0.20806889086961747,
|
|
"num_tokens": 12075312.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.231590509414673,
|
|
"epoch": 5.492957746478873,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002686736208283649,
|
|
"loss": 4.7629,
|
|
"mean_token_accuracy": 0.20990005433559417,
|
|
"num_tokens": 12083990.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.376721620559692,
|
|
"epoch": 5.497100248550124,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026835419744371627,
|
|
"loss": 4.8732,
|
|
"mean_token_accuracy": 0.19839346706867217,
|
|
"num_tokens": 12093301.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.241463851928711,
|
|
"epoch": 5.501242750621375,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002680347874643775,
|
|
"loss": 4.6882,
|
|
"mean_token_accuracy": 0.21418087482452391,
|
|
"num_tokens": 12102054.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.262872982025146,
|
|
"epoch": 5.505385252692626,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002677153915346336,
|
|
"loss": 4.7708,
|
|
"mean_token_accuracy": 0.20341249853372573,
|
|
"num_tokens": 12111432.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.237018823623657,
|
|
"epoch": 5.5095277547638775,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002673960102987412,
|
|
"loss": 4.7032,
|
|
"mean_token_accuracy": 0.21349260658025743,
|
|
"num_tokens": 12120424.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.301366519927979,
|
|
"epoch": 5.513670256835129,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026707664440092676,
|
|
"loss": 4.8377,
|
|
"mean_token_accuracy": 0.20626042485237123,
|
|
"num_tokens": 12130210.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.414148330688477,
|
|
"epoch": 5.51781275890638,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026675729448538655,
|
|
"loss": 4.8269,
|
|
"mean_token_accuracy": 0.21095569282770157,
|
|
"num_tokens": 12139366.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.2740867137908936,
|
|
"epoch": 5.521955260977631,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00026643796119628404,
|
|
"loss": 4.6432,
|
|
"mean_token_accuracy": 0.21321782022714614,
|
|
"num_tokens": 12148385.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.290845537185669,
|
|
"epoch": 5.526097763048882,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002661186451777495,
|
|
"loss": 4.7822,
|
|
"mean_token_accuracy": 0.20492887496948242,
|
|
"num_tokens": 12156640.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.280693674087525,
|
|
"epoch": 5.530240265120133,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002657993470738781,
|
|
"loss": 4.8587,
|
|
"mean_token_accuracy": 0.19673615247011184,
|
|
"num_tokens": 12166532.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.24079852104187,
|
|
"epoch": 5.534382767191383,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026548006752872915,
|
|
"loss": 4.6934,
|
|
"mean_token_accuracy": 0.2146509751677513,
|
|
"num_tokens": 12175239.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.243715524673462,
|
|
"epoch": 5.538525269262634,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002651608071863244,
|
|
"loss": 4.7611,
|
|
"mean_token_accuracy": 0.20688466280698775,
|
|
"num_tokens": 12184588.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.333131456375122,
|
|
"epoch": 5.542667771333885,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026484156669064665,
|
|
"loss": 4.8177,
|
|
"mean_token_accuracy": 0.20230551064014435,
|
|
"num_tokens": 12194140.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.310734224319458,
|
|
"epoch": 5.5468102734051365,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000264522346685639,
|
|
"loss": 4.8193,
|
|
"mean_token_accuracy": 0.20696787536144257,
|
|
"num_tokens": 12203504.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.33381724357605,
|
|
"epoch": 5.550952775476388,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002642031478152031,
|
|
"loss": 4.7782,
|
|
"mean_token_accuracy": 0.207833394408226,
|
|
"num_tokens": 12212594.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.3383890151977536,
|
|
"epoch": 5.555095277547639,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026388397072319786,
|
|
"loss": 4.8162,
|
|
"mean_token_accuracy": 0.2045178234577179,
|
|
"num_tokens": 12222101.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.29089732170105,
|
|
"epoch": 5.55923777961889,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002635648160534385,
|
|
"loss": 4.8118,
|
|
"mean_token_accuracy": 0.20435990989208222,
|
|
"num_tokens": 12232118.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.352822351455688,
|
|
"epoch": 5.563380281690141,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026324568444969473,
|
|
"loss": 4.7439,
|
|
"mean_token_accuracy": 0.20706239342689514,
|
|
"num_tokens": 12241275.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.245583724975586,
|
|
"epoch": 5.567522783761392,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026292657655569,
|
|
"loss": 4.7375,
|
|
"mean_token_accuracy": 0.21279312521219254,
|
|
"num_tokens": 12251125.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.229356813430786,
|
|
"epoch": 5.571665285832643,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026260749301509954,
|
|
"loss": 4.7241,
|
|
"mean_token_accuracy": 0.21198077648878097,
|
|
"num_tokens": 12260613.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.324583339691162,
|
|
"epoch": 5.575807787903894,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026228843447155005,
|
|
"loss": 4.733,
|
|
"mean_token_accuracy": 0.2107634127140045,
|
|
"num_tokens": 12269878.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.270222854614258,
|
|
"epoch": 5.5799502899751445,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00026196940156861736,
|
|
"loss": 4.7819,
|
|
"mean_token_accuracy": 0.20616225749254227,
|
|
"num_tokens": 12278616.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.194797945022583,
|
|
"epoch": 5.584092792046396,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026165039494982555,
|
|
"loss": 4.7068,
|
|
"mean_token_accuracy": 0.2110671728849411,
|
|
"num_tokens": 12287104.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.187378549575806,
|
|
"epoch": 5.588235294117647,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026133141525864605,
|
|
"loss": 4.7357,
|
|
"mean_token_accuracy": 0.20822987705469131,
|
|
"num_tokens": 12296503.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.294446992874145,
|
|
"epoch": 5.592377796188898,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026101246313849557,
|
|
"loss": 4.6989,
|
|
"mean_token_accuracy": 0.2123148426413536,
|
|
"num_tokens": 12306298.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.2862865924835205,
|
|
"epoch": 5.596520298260149,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002606935392327356,
|
|
"loss": 4.7664,
|
|
"mean_token_accuracy": 0.20611962229013442,
|
|
"num_tokens": 12314826.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.258157396316529,
|
|
"epoch": 5.6006628003314,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026037464418467035,
|
|
"loss": 4.7079,
|
|
"mean_token_accuracy": 0.21357058137655258,
|
|
"num_tokens": 12323776.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.312329053878784,
|
|
"epoch": 5.604805302402651,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002600557786375459,
|
|
"loss": 4.7819,
|
|
"mean_token_accuracy": 0.20854859054088593,
|
|
"num_tokens": 12332374.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.217779874801636,
|
|
"epoch": 5.608947804473902,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025973694323454916,
|
|
"loss": 4.6354,
|
|
"mean_token_accuracy": 0.21756067872047424,
|
|
"num_tokens": 12341007.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.175305271148682,
|
|
"epoch": 5.613090306545153,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025941813861880575,
|
|
"loss": 4.7114,
|
|
"mean_token_accuracy": 0.20966241657733917,
|
|
"num_tokens": 12350330.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.267894315719604,
|
|
"epoch": 5.6172328086164045,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025909936543337957,
|
|
"loss": 4.7519,
|
|
"mean_token_accuracy": 0.1993374451994896,
|
|
"num_tokens": 12358592.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.290813016891479,
|
|
"epoch": 5.621375310687656,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002587806243212709,
|
|
"loss": 4.7692,
|
|
"mean_token_accuracy": 0.20917081385850905,
|
|
"num_tokens": 12368195.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.241605234146118,
|
|
"epoch": 5.625517812758907,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002584619159254154,
|
|
"loss": 4.7124,
|
|
"mean_token_accuracy": 0.20834194272756576,
|
|
"num_tokens": 12377206.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.247827816009521,
|
|
"epoch": 5.629660314830158,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025814324088868276,
|
|
"loss": 4.7593,
|
|
"mean_token_accuracy": 0.21013118624687194,
|
|
"num_tokens": 12386595.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.175970602035522,
|
|
"epoch": 5.633802816901408,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00025782459985387526,
|
|
"loss": 4.6569,
|
|
"mean_token_accuracy": 0.2180004358291626,
|
|
"num_tokens": 12395923.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.269517803192139,
|
|
"epoch": 5.637945318972659,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002575059934637268,
|
|
"loss": 4.7985,
|
|
"mean_token_accuracy": 0.20669761300086975,
|
|
"num_tokens": 12405094.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.268765878677368,
|
|
"epoch": 5.64208782104391,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002571874223609012,
|
|
"loss": 4.7972,
|
|
"mean_token_accuracy": 0.20435070544481276,
|
|
"num_tokens": 12414834.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.294848012924194,
|
|
"epoch": 5.646230323115161,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025686888718799136,
|
|
"loss": 4.7513,
|
|
"mean_token_accuracy": 0.20469613671302794,
|
|
"num_tokens": 12423485.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.368587017059326,
|
|
"epoch": 5.6503728251864125,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002565503885875173,
|
|
"loss": 4.7935,
|
|
"mean_token_accuracy": 0.20704300552606583,
|
|
"num_tokens": 12432149.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.2622401237487795,
|
|
"epoch": 5.654515327257664,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00025623192720192564,
|
|
"loss": 4.7433,
|
|
"mean_token_accuracy": 0.2119109019637108,
|
|
"num_tokens": 12441272.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.190871524810791,
|
|
"epoch": 5.658657829328915,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002559135036735877,
|
|
"loss": 4.7388,
|
|
"mean_token_accuracy": 0.20155471414327622,
|
|
"num_tokens": 12449318.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.168833065032959,
|
|
"epoch": 5.662800331400166,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025559511864479864,
|
|
"loss": 4.6898,
|
|
"mean_token_accuracy": 0.21231105625629426,
|
|
"num_tokens": 12458473.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.313192129135132,
|
|
"epoch": 5.666942833471417,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002552767727577758,
|
|
"loss": 4.8113,
|
|
"mean_token_accuracy": 0.20445662438869477,
|
|
"num_tokens": 12467703.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.292496681213379,
|
|
"epoch": 5.671085335542668,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00025495846665465774,
|
|
"loss": 4.6938,
|
|
"mean_token_accuracy": 0.21795116513967513,
|
|
"num_tokens": 12476166.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.235640621185302,
|
|
"epoch": 5.675227837613919,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002546402009775024,
|
|
"loss": 4.7403,
|
|
"mean_token_accuracy": 0.20330073386430741,
|
|
"num_tokens": 12484880.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.31177544593811,
|
|
"epoch": 5.679370339685169,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025432197636828663,
|
|
"loss": 4.8103,
|
|
"mean_token_accuracy": 0.20381423085927963,
|
|
"num_tokens": 12494208.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.334811210632324,
|
|
"epoch": 5.683512841756421,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002540037934689042,
|
|
"loss": 4.8024,
|
|
"mean_token_accuracy": 0.20098040848970414,
|
|
"num_tokens": 12503647.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.295432138442993,
|
|
"epoch": 5.687655343827672,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00025368565292116475,
|
|
"loss": 4.8225,
|
|
"mean_token_accuracy": 0.20177613645792009,
|
|
"num_tokens": 12512869.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.28051609992981,
|
|
"epoch": 5.691797845898923,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002533675553667927,
|
|
"loss": 4.8747,
|
|
"mean_token_accuracy": 0.19059171974658967,
|
|
"num_tokens": 12523222.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.24097170829773,
|
|
"epoch": 5.695940347970174,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00025304950144742523,
|
|
"loss": 4.7456,
|
|
"mean_token_accuracy": 0.20784133225679396,
|
|
"num_tokens": 12532343.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.255422782897949,
|
|
"epoch": 5.700082850041425,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002527314918046122,
|
|
"loss": 4.7467,
|
|
"mean_token_accuracy": 0.21441746652126312,
|
|
"num_tokens": 12541508.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.222242641448974,
|
|
"epoch": 5.704225352112676,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002524135270798136,
|
|
"loss": 4.7731,
|
|
"mean_token_accuracy": 0.20265263617038726,
|
|
"num_tokens": 12550837.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.259101152420044,
|
|
"epoch": 5.708367854183927,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00025209560791439914,
|
|
"loss": 4.7223,
|
|
"mean_token_accuracy": 0.21348682641983033,
|
|
"num_tokens": 12559859.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.279586315155029,
|
|
"epoch": 5.712510356255178,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002517777349496465,
|
|
"loss": 4.8096,
|
|
"mean_token_accuracy": 0.20708946287631988,
|
|
"num_tokens": 12569137.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.332545948028565,
|
|
"epoch": 5.716652858326429,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002514599088267401,
|
|
"loss": 4.8391,
|
|
"mean_token_accuracy": 0.20370080918073655,
|
|
"num_tokens": 12578111.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.30065131187439,
|
|
"epoch": 5.72079536039768,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00025114213018677004,
|
|
"loss": 4.7826,
|
|
"mean_token_accuracy": 0.21398544907569886,
|
|
"num_tokens": 12587604.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.328654146194458,
|
|
"epoch": 5.7249378624689315,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002508243996707305,
|
|
"loss": 4.7982,
|
|
"mean_token_accuracy": 0.2033779114484787,
|
|
"num_tokens": 12596620.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.252767944335938,
|
|
"epoch": 5.729080364540183,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00025050671791951887,
|
|
"loss": 4.7682,
|
|
"mean_token_accuracy": 0.20568756759166718,
|
|
"num_tokens": 12605452.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.195128536224365,
|
|
"epoch": 5.733222866611433,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002501890855739337,
|
|
"loss": 4.6654,
|
|
"mean_token_accuracy": 0.22127037048339843,
|
|
"num_tokens": 12614009.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.287421989440918,
|
|
"epoch": 5.737365368682684,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024987150327467414,
|
|
"loss": 4.8201,
|
|
"mean_token_accuracy": 0.2010796323418617,
|
|
"num_tokens": 12624135.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.221401596069336,
|
|
"epoch": 5.741507870753935,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024955397166233855,
|
|
"loss": 4.6474,
|
|
"mean_token_accuracy": 0.22327282428741455,
|
|
"num_tokens": 12632870.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.304728364944458,
|
|
"epoch": 5.745650372825186,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024923649137742276,
|
|
"loss": 4.8549,
|
|
"mean_token_accuracy": 0.1950896918773651,
|
|
"num_tokens": 12643606.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.385456705093384,
|
|
"epoch": 5.749792874896437,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002489190630603194,
|
|
"loss": 4.8834,
|
|
"mean_token_accuracy": 0.20502226054668427,
|
|
"num_tokens": 12652926.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.331468009948731,
|
|
"epoch": 5.753935376967688,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002486016873513159,
|
|
"loss": 4.8219,
|
|
"mean_token_accuracy": 0.20297979861497878,
|
|
"num_tokens": 12661841.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.257536315917969,
|
|
"epoch": 5.7580778790389395,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002482843648905937,
|
|
"loss": 4.7409,
|
|
"mean_token_accuracy": 0.2069641500711441,
|
|
"num_tokens": 12670527.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.216205167770386,
|
|
"epoch": 5.762220381110191,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002479670963182271,
|
|
"loss": 4.7365,
|
|
"mean_token_accuracy": 0.19854751229286194,
|
|
"num_tokens": 12680346.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.315979433059693,
|
|
"epoch": 5.766362883181442,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024764988227418114,
|
|
"loss": 4.8273,
|
|
"mean_token_accuracy": 0.20946504771709443,
|
|
"num_tokens": 12691088.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.343829536437989,
|
|
"epoch": 5.770505385252693,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002473327233983116,
|
|
"loss": 4.7894,
|
|
"mean_token_accuracy": 0.20746862143278122,
|
|
"num_tokens": 12700296.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.241484355926514,
|
|
"epoch": 5.774647887323944,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002470156203303622,
|
|
"loss": 4.7882,
|
|
"mean_token_accuracy": 0.20888512581586838,
|
|
"num_tokens": 12709595.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.276812028884888,
|
|
"epoch": 5.778790389395195,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002466985737099649,
|
|
"loss": 4.838,
|
|
"mean_token_accuracy": 0.20352406352758406,
|
|
"num_tokens": 12719052.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.237840700149536,
|
|
"epoch": 5.782932891466446,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002463815841766371,
|
|
"loss": 4.7033,
|
|
"mean_token_accuracy": 0.21847177892923356,
|
|
"num_tokens": 12727287.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.252689790725708,
|
|
"epoch": 5.787075393537696,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002460646523697813,
|
|
"loss": 4.7649,
|
|
"mean_token_accuracy": 0.2055829644203186,
|
|
"num_tokens": 12736511.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.295902395248413,
|
|
"epoch": 5.7912178956089475,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024574777892868386,
|
|
"loss": 4.7758,
|
|
"mean_token_accuracy": 0.20809024274349214,
|
|
"num_tokens": 12745597.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.260382986068725,
|
|
"epoch": 5.795360397680199,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024543096449251286,
|
|
"loss": 4.6362,
|
|
"mean_token_accuracy": 0.221256822347641,
|
|
"num_tokens": 12753492.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.3082959175109865,
|
|
"epoch": 5.79950289975145,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002451142097003177,
|
|
"loss": 4.7765,
|
|
"mean_token_accuracy": 0.21173641979694366,
|
|
"num_tokens": 12761999.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 5.79950289975145,
|
|
"eval_entropy": 5.159527343473455,
|
|
"eval_loss": 5.7473931312561035,
|
|
"eval_mean_token_accuracy": 0.1656088415525361,
|
|
"eval_num_tokens": 12761999.0,
|
|
"eval_runtime": 2.7553,
|
|
"eval_samples_per_second": 1359.941,
|
|
"eval_steps_per_second": 170.219,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.250266122817993,
|
|
"epoch": 5.803645401822701,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002447975151910276,
|
|
"loss": 4.6884,
|
|
"mean_token_accuracy": 0.20662099421024321,
|
|
"num_tokens": 12770922.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.225958204269409,
|
|
"epoch": 5.807787903893952,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024448088160344974,
|
|
"loss": 4.72,
|
|
"mean_token_accuracy": 0.2077726110816002,
|
|
"num_tokens": 12779539.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.266984176635742,
|
|
"epoch": 5.811930405965203,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002441643095762688,
|
|
"loss": 4.7539,
|
|
"mean_token_accuracy": 0.20866499096155167,
|
|
"num_tokens": 12788121.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.206212854385376,
|
|
"epoch": 5.816072908036454,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000243847799748045,
|
|
"loss": 4.6301,
|
|
"mean_token_accuracy": 0.21100328713655472,
|
|
"num_tokens": 12797390.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.249676036834717,
|
|
"epoch": 5.820215410107705,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00024353135275721356,
|
|
"loss": 4.8644,
|
|
"mean_token_accuracy": 0.19687919616699218,
|
|
"num_tokens": 12807876.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.184282016754151,
|
|
"epoch": 5.824357912178956,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024321496924208235,
|
|
"loss": 4.6805,
|
|
"mean_token_accuracy": 0.2075910523533821,
|
|
"num_tokens": 12816496.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.282375478744507,
|
|
"epoch": 5.8285004142502075,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024289864984083166,
|
|
"loss": 4.7571,
|
|
"mean_token_accuracy": 0.21421819925308228,
|
|
"num_tokens": 12825925.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.330373096466064,
|
|
"epoch": 5.832642916321458,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002425823951915122,
|
|
"loss": 4.776,
|
|
"mean_token_accuracy": 0.20406203866004943,
|
|
"num_tokens": 12835479.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.227119112014771,
|
|
"epoch": 5.836785418392709,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002422662059320442,
|
|
"loss": 4.6764,
|
|
"mean_token_accuracy": 0.2194245144724846,
|
|
"num_tokens": 12844024.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.242313861846924,
|
|
"epoch": 5.84092792046396,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024195008270021595,
|
|
"loss": 4.6767,
|
|
"mean_token_accuracy": 0.21089302152395248,
|
|
"num_tokens": 12852673.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.168107366561889,
|
|
"epoch": 5.845070422535211,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024163402613368252,
|
|
"loss": 4.7042,
|
|
"mean_token_accuracy": 0.21519682556390762,
|
|
"num_tokens": 12861191.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.273849678039551,
|
|
"epoch": 5.849212924606462,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024131803686996457,
|
|
"loss": 4.8483,
|
|
"mean_token_accuracy": 0.20502860397100447,
|
|
"num_tokens": 12870874.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.314899635314942,
|
|
"epoch": 5.853355426677713,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00024100211554644696,
|
|
"loss": 4.7741,
|
|
"mean_token_accuracy": 0.21026002764701843,
|
|
"num_tokens": 12879896.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.297897291183472,
|
|
"epoch": 5.857497928748964,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00024068626280037753,
|
|
"loss": 4.786,
|
|
"mean_token_accuracy": 0.20807335823774337,
|
|
"num_tokens": 12888665.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.261636590957641,
|
|
"epoch": 5.8616404308202155,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024037047926886578,
|
|
"loss": 4.7416,
|
|
"mean_token_accuracy": 0.20277187824249268,
|
|
"num_tokens": 12897477.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.251101160049439,
|
|
"epoch": 5.865782932891467,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024005476558888172,
|
|
"loss": 4.794,
|
|
"mean_token_accuracy": 0.2007831260561943,
|
|
"num_tokens": 12905934.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.255609798431396,
|
|
"epoch": 5.869925434962718,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002397391223972541,
|
|
"loss": 4.7929,
|
|
"mean_token_accuracy": 0.20374300181865693,
|
|
"num_tokens": 12915092.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.2630064487457275,
|
|
"epoch": 5.874067937033969,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023942355033066993,
|
|
"loss": 4.8721,
|
|
"mean_token_accuracy": 0.19899580031633377,
|
|
"num_tokens": 12924775.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.329388093948364,
|
|
"epoch": 5.87821043910522,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023910805002567244,
|
|
"loss": 4.7789,
|
|
"mean_token_accuracy": 0.2048063576221466,
|
|
"num_tokens": 12933677.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.261180973052978,
|
|
"epoch": 5.882352941176471,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002387926221186603,
|
|
"loss": 4.8017,
|
|
"mean_token_accuracy": 0.2079474300146103,
|
|
"num_tokens": 12942416.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.231311798095703,
|
|
"epoch": 5.886495443247721,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023847726724588608,
|
|
"loss": 4.7718,
|
|
"mean_token_accuracy": 0.21066755801439285,
|
|
"num_tokens": 12952284.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.278716087341309,
|
|
"epoch": 5.890637945318972,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002381619860434549,
|
|
"loss": 4.8123,
|
|
"mean_token_accuracy": 0.19911128133535386,
|
|
"num_tokens": 12960694.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.215971040725708,
|
|
"epoch": 5.894780447390223,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023784677914732356,
|
|
"loss": 4.727,
|
|
"mean_token_accuracy": 0.20931341648101806,
|
|
"num_tokens": 12969138.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.292342042922973,
|
|
"epoch": 5.8989229494614746,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023753164719329868,
|
|
"loss": 4.766,
|
|
"mean_token_accuracy": 0.20421497970819474,
|
|
"num_tokens": 12978518.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.309631299972534,
|
|
"epoch": 5.903065451532726,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023721659081703595,
|
|
"loss": 4.7856,
|
|
"mean_token_accuracy": 0.2164353221654892,
|
|
"num_tokens": 12987506.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.2704308986663815,
|
|
"epoch": 5.907207953603977,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023690161065403838,
|
|
"loss": 4.7667,
|
|
"mean_token_accuracy": 0.20384967178106309,
|
|
"num_tokens": 12996259.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.284243488311768,
|
|
"epoch": 5.911350455675228,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002365867073396555,
|
|
"loss": 4.7914,
|
|
"mean_token_accuracy": 0.20621232688426971,
|
|
"num_tokens": 13005430.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.2192955493927,
|
|
"epoch": 5.915492957746479,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023627188150908174,
|
|
"loss": 4.7414,
|
|
"mean_token_accuracy": 0.21523537784814833,
|
|
"num_tokens": 13014915.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.267714023590088,
|
|
"epoch": 5.91963545981773,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023595713379735494,
|
|
"loss": 4.7483,
|
|
"mean_token_accuracy": 0.20747244358062744,
|
|
"num_tokens": 13023471.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.29262957572937,
|
|
"epoch": 5.923777961888981,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023564246483935602,
|
|
"loss": 4.7364,
|
|
"mean_token_accuracy": 0.20276205390691757,
|
|
"num_tokens": 13033270.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.26424388885498,
|
|
"epoch": 5.927920463960232,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002353278752698063,
|
|
"loss": 4.7029,
|
|
"mean_token_accuracy": 0.2120548114180565,
|
|
"num_tokens": 13043300.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.261445045471191,
|
|
"epoch": 5.9320629660314825,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002350133657232676,
|
|
"loss": 4.7627,
|
|
"mean_token_accuracy": 0.20174376368522645,
|
|
"num_tokens": 13052152.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.286171960830688,
|
|
"epoch": 5.9362054681027345,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002346989368341399,
|
|
"loss": 4.8033,
|
|
"mean_token_accuracy": 0.20869386494159697,
|
|
"num_tokens": 13061704.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.280704736709595,
|
|
"epoch": 5.940347970173985,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023438458923666057,
|
|
"loss": 4.7209,
|
|
"mean_token_accuracy": 0.2152332365512848,
|
|
"num_tokens": 13071055.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.268754768371582,
|
|
"epoch": 5.944490472245236,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002340703235649032,
|
|
"loss": 4.8331,
|
|
"mean_token_accuracy": 0.1926100879907608,
|
|
"num_tokens": 13080023.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.263068819046021,
|
|
"epoch": 5.948632974316487,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023375614045277593,
|
|
"loss": 4.7536,
|
|
"mean_token_accuracy": 0.20691772699356079,
|
|
"num_tokens": 13088690.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.232177972793579,
|
|
"epoch": 5.952775476387738,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023344204053402047,
|
|
"loss": 4.7377,
|
|
"mean_token_accuracy": 0.21229761391878127,
|
|
"num_tokens": 13097116.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.343814706802368,
|
|
"epoch": 5.956917978458989,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023312802444221066,
|
|
"loss": 4.7995,
|
|
"mean_token_accuracy": 0.20769918113946914,
|
|
"num_tokens": 13106463.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.273978662490845,
|
|
"epoch": 5.96106048053024,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002328140928107511,
|
|
"loss": 4.7045,
|
|
"mean_token_accuracy": 0.21413260698318481,
|
|
"num_tokens": 13115769.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.293670988082885,
|
|
"epoch": 5.965202982601491,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0002325002462728766,
|
|
"loss": 4.8058,
|
|
"mean_token_accuracy": 0.20858496129512788,
|
|
"num_tokens": 13124928.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.264481544494629,
|
|
"epoch": 5.9693454846727425,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023218648546164945,
|
|
"loss": 4.7587,
|
|
"mean_token_accuracy": 0.20816877782344817,
|
|
"num_tokens": 13134106.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.401337671279907,
|
|
"epoch": 5.973487986743994,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002318728110099599,
|
|
"loss": 4.9463,
|
|
"mean_token_accuracy": 0.19887659698724747,
|
|
"num_tokens": 13143019.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.267350912094116,
|
|
"epoch": 5.977630488815245,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00023155922355052339,
|
|
"loss": 4.6633,
|
|
"mean_token_accuracy": 0.223431359231472,
|
|
"num_tokens": 13152675.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.256755113601685,
|
|
"epoch": 5.981772990886496,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023124572371588,
|
|
"loss": 4.6968,
|
|
"mean_token_accuracy": 0.2141074687242508,
|
|
"num_tokens": 13161094.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.22665810585022,
|
|
"epoch": 5.985915492957746,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023093231213839344,
|
|
"loss": 4.7801,
|
|
"mean_token_accuracy": 0.2072029694914818,
|
|
"num_tokens": 13170858.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.248539400100708,
|
|
"epoch": 5.990057995028997,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002306189894502488,
|
|
"loss": 4.8402,
|
|
"mean_token_accuracy": 0.1976972222328186,
|
|
"num_tokens": 13179668.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.278863763809204,
|
|
"epoch": 5.994200497100248,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023030575628345246,
|
|
"loss": 4.7021,
|
|
"mean_token_accuracy": 0.2132765233516693,
|
|
"num_tokens": 13188312.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.254744529724121,
|
|
"epoch": 5.998342999171499,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022999261326982985,
|
|
"loss": 4.7474,
|
|
"mean_token_accuracy": 0.20551027208566666,
|
|
"num_tokens": 13196989.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.215828323364258,
|
|
"epoch": 6.0024855012427505,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022967956104102445,
|
|
"loss": 4.6371,
|
|
"mean_token_accuracy": 0.21983437091112137,
|
|
"num_tokens": 13205830.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.325025272369385,
|
|
"epoch": 6.006628003314002,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022936660022849714,
|
|
"loss": 4.5948,
|
|
"mean_token_accuracy": 0.22821232080459594,
|
|
"num_tokens": 13214711.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.262428140640258,
|
|
"epoch": 6.010770505385253,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022905373146352382,
|
|
"loss": 4.6602,
|
|
"mean_token_accuracy": 0.21241429150104524,
|
|
"num_tokens": 13223736.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.257147979736328,
|
|
"epoch": 6.014913007456504,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002287409553771952,
|
|
"loss": 4.556,
|
|
"mean_token_accuracy": 0.22737590819597245,
|
|
"num_tokens": 13232666.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.291457891464233,
|
|
"epoch": 6.019055509527755,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022842827260041455,
|
|
"loss": 4.5848,
|
|
"mean_token_accuracy": 0.21761821508407592,
|
|
"num_tokens": 13241770.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.261249971389771,
|
|
"epoch": 6.023198011599006,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022811568376389745,
|
|
"loss": 4.5326,
|
|
"mean_token_accuracy": 0.22425681203603745,
|
|
"num_tokens": 13250475.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.233007001876831,
|
|
"epoch": 6.027340513670257,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022780318949816963,
|
|
"loss": 4.5183,
|
|
"mean_token_accuracy": 0.22254440933465958,
|
|
"num_tokens": 13259587.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.216541528701782,
|
|
"epoch": 6.031483015741508,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022749079043356623,
|
|
"loss": 4.5404,
|
|
"mean_token_accuracy": 0.23734405487775803,
|
|
"num_tokens": 13269681.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.211581516265869,
|
|
"epoch": 6.0356255178127585,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00022717848720023022,
|
|
"loss": 4.5539,
|
|
"mean_token_accuracy": 0.2277751386165619,
|
|
"num_tokens": 13278186.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.25051851272583,
|
|
"epoch": 6.03976801988401,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022686628042811142,
|
|
"loss": 4.6462,
|
|
"mean_token_accuracy": 0.21404165476560594,
|
|
"num_tokens": 13287194.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.334251356124878,
|
|
"epoch": 6.043910521955261,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000226554170746965,
|
|
"loss": 4.6707,
|
|
"mean_token_accuracy": 0.21294027268886567,
|
|
"num_tokens": 13296774.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.338882112503052,
|
|
"epoch": 6.048053024026512,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022624215878635023,
|
|
"loss": 4.5887,
|
|
"mean_token_accuracy": 0.21931985318660735,
|
|
"num_tokens": 13306368.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.310757160186768,
|
|
"epoch": 6.052195526097763,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022593024517562945,
|
|
"loss": 4.5551,
|
|
"mean_token_accuracy": 0.22974174320697785,
|
|
"num_tokens": 13315393.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.216650152206421,
|
|
"epoch": 6.056338028169014,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022561843054396636,
|
|
"loss": 4.5496,
|
|
"mean_token_accuracy": 0.22554152607917785,
|
|
"num_tokens": 13324384.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.265720653533935,
|
|
"epoch": 6.060480530240265,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00022530671552032521,
|
|
"loss": 4.6095,
|
|
"mean_token_accuracy": 0.2231447860598564,
|
|
"num_tokens": 13333191.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.269006586074829,
|
|
"epoch": 6.064623032311516,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002249951007334693,
|
|
"loss": 4.6016,
|
|
"mean_token_accuracy": 0.21682012677192689,
|
|
"num_tokens": 13342911.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.313802003860474,
|
|
"epoch": 6.068765534382767,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022468358681195966,
|
|
"loss": 4.5911,
|
|
"mean_token_accuracy": 0.2220294862985611,
|
|
"num_tokens": 13352464.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.279155349731445,
|
|
"epoch": 6.072908036454018,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022437217438415393,
|
|
"loss": 4.5871,
|
|
"mean_token_accuracy": 0.22435862571001053,
|
|
"num_tokens": 13361812.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.328032398223877,
|
|
"epoch": 6.0770505385252696,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022406086407820503,
|
|
"loss": 4.6541,
|
|
"mean_token_accuracy": 0.21680645495653153,
|
|
"num_tokens": 13372080.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.283897733688354,
|
|
"epoch": 6.081193040596521,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022374965652205975,
|
|
"loss": 4.6976,
|
|
"mean_token_accuracy": 0.20982298403978347,
|
|
"num_tokens": 13380897.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.338993167877197,
|
|
"epoch": 6.085335542667772,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022343855234345796,
|
|
"loss": 4.6891,
|
|
"mean_token_accuracy": 0.21388355493545533,
|
|
"num_tokens": 13389885.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.288727903366089,
|
|
"epoch": 6.089478044739022,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022312755216993065,
|
|
"loss": 4.6211,
|
|
"mean_token_accuracy": 0.21839348524808883,
|
|
"num_tokens": 13399734.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.224732303619385,
|
|
"epoch": 6.093620546810273,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022281665662879903,
|
|
"loss": 4.5653,
|
|
"mean_token_accuracy": 0.22080505937337874,
|
|
"num_tokens": 13408641.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.2608020305633545,
|
|
"epoch": 6.097763048881524,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022250586634717362,
|
|
"loss": 4.6026,
|
|
"mean_token_accuracy": 0.21388103663921357,
|
|
"num_tokens": 13418585.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.244136095046997,
|
|
"epoch": 6.101905550952775,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022219518195195216,
|
|
"loss": 4.5367,
|
|
"mean_token_accuracy": 0.22432616204023362,
|
|
"num_tokens": 13427897.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.131761312484741,
|
|
"epoch": 6.106048053024026,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022188460406981925,
|
|
"loss": 4.4993,
|
|
"mean_token_accuracy": 0.22558098286390305,
|
|
"num_tokens": 13437311.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.284880685806274,
|
|
"epoch": 6.1101905550952775,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022157413332724436,
|
|
"loss": 4.5608,
|
|
"mean_token_accuracy": 0.22710221409797668,
|
|
"num_tokens": 13446359.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.326997709274292,
|
|
"epoch": 6.114333057166529,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022126377035048072,
|
|
"loss": 4.6056,
|
|
"mean_token_accuracy": 0.22011455297470092,
|
|
"num_tokens": 13455302.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.365768575668335,
|
|
"epoch": 6.11847555923778,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002209535157655646,
|
|
"loss": 4.6937,
|
|
"mean_token_accuracy": 0.21688085198402404,
|
|
"num_tokens": 13466391.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.246037006378174,
|
|
"epoch": 6.122618061309031,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022064337019831322,
|
|
"loss": 4.547,
|
|
"mean_token_accuracy": 0.22545033246278762,
|
|
"num_tokens": 13476130.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.183640670776367,
|
|
"epoch": 6.126760563380282,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00022033333427432422,
|
|
"loss": 4.5326,
|
|
"mean_token_accuracy": 0.22769198417663575,
|
|
"num_tokens": 13485105.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.284799098968506,
|
|
"epoch": 6.130903065451533,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002200234086189738,
|
|
"loss": 4.5988,
|
|
"mean_token_accuracy": 0.21623703986406326,
|
|
"num_tokens": 13494914.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.291166830062866,
|
|
"epoch": 6.135045567522784,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021971359385741608,
|
|
"loss": 4.6312,
|
|
"mean_token_accuracy": 0.21725507229566574,
|
|
"num_tokens": 13504293.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.350364637374878,
|
|
"epoch": 6.139188069594034,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021940389061458115,
|
|
"loss": 4.6935,
|
|
"mean_token_accuracy": 0.22431259900331496,
|
|
"num_tokens": 13513972.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.278038835525512,
|
|
"epoch": 6.1433305716652855,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002190942995151743,
|
|
"loss": 4.6089,
|
|
"mean_token_accuracy": 0.2185722753405571,
|
|
"num_tokens": 13522508.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.296834516525268,
|
|
"epoch": 6.147473073736537,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002187848211836746,
|
|
"loss": 4.53,
|
|
"mean_token_accuracy": 0.22915603816509247,
|
|
"num_tokens": 13531235.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.256722116470337,
|
|
"epoch": 6.151615575807788,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002184754562443337,
|
|
"loss": 4.5707,
|
|
"mean_token_accuracy": 0.22529171109199525,
|
|
"num_tokens": 13541023.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.217341756820678,
|
|
"epoch": 6.155758077879039,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021816620532117448,
|
|
"loss": 4.5852,
|
|
"mean_token_accuracy": 0.2248647004365921,
|
|
"num_tokens": 13550622.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.230591440200806,
|
|
"epoch": 6.15990057995029,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002178570690379899,
|
|
"loss": 4.6031,
|
|
"mean_token_accuracy": 0.21641309708356857,
|
|
"num_tokens": 13559423.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.235513544082641,
|
|
"epoch": 6.164043082021541,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021754804801834144,
|
|
"loss": 4.5917,
|
|
"mean_token_accuracy": 0.21699417978525162,
|
|
"num_tokens": 13569306.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.228752660751343,
|
|
"epoch": 6.168185584092792,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002172391428855585,
|
|
"loss": 4.5366,
|
|
"mean_token_accuracy": 0.22735673934221268,
|
|
"num_tokens": 13578367.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.265543508529663,
|
|
"epoch": 6.172328086164043,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021693035426273633,
|
|
"loss": 4.5986,
|
|
"mean_token_accuracy": 0.21967096477746964,
|
|
"num_tokens": 13586930.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.246473407745361,
|
|
"epoch": 6.176470588235294,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00021662168277273547,
|
|
"loss": 4.5681,
|
|
"mean_token_accuracy": 0.22939085811376572,
|
|
"num_tokens": 13595891.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.319870328903198,
|
|
"epoch": 6.1806130903065455,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002163131290381799,
|
|
"loss": 4.6599,
|
|
"mean_token_accuracy": 0.20967274755239487,
|
|
"num_tokens": 13605786.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.238808870315552,
|
|
"epoch": 6.184755592377797,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021600469368145624,
|
|
"loss": 4.5621,
|
|
"mean_token_accuracy": 0.2280669927597046,
|
|
"num_tokens": 13614561.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.226732301712036,
|
|
"epoch": 6.188898094449047,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002156963773247124,
|
|
"loss": 4.5554,
|
|
"mean_token_accuracy": 0.22617974877357483,
|
|
"num_tokens": 13622853.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.20526967048645,
|
|
"epoch": 6.193040596520298,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002153881805898561,
|
|
"loss": 4.5398,
|
|
"mean_token_accuracy": 0.22647101283073426,
|
|
"num_tokens": 13631709.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.293169021606445,
|
|
"epoch": 6.197183098591549,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002150801040985539,
|
|
"loss": 4.63,
|
|
"mean_token_accuracy": 0.21966254264116286,
|
|
"num_tokens": 13640854.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.218788433074951,
|
|
"epoch": 6.2013256006628,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021477214847222958,
|
|
"loss": 4.5758,
|
|
"mean_token_accuracy": 0.21710532903671265,
|
|
"num_tokens": 13649318.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.315838146209717,
|
|
"epoch": 6.205468102734051,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021446431433206348,
|
|
"loss": 4.6349,
|
|
"mean_token_accuracy": 0.22002605497837066,
|
|
"num_tokens": 13658738.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.187892484664917,
|
|
"epoch": 6.209610604805302,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002141566022989907,
|
|
"loss": 4.5484,
|
|
"mean_token_accuracy": 0.2262681856751442,
|
|
"num_tokens": 13667315.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.282585573196411,
|
|
"epoch": 6.2137531068765535,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002138490129936998,
|
|
"loss": 4.6721,
|
|
"mean_token_accuracy": 0.21348128467798233,
|
|
"num_tokens": 13676665.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.2137531068765535,
|
|
"eval_entropy": 5.127327098521088,
|
|
"eval_loss": 5.749592304229736,
|
|
"eval_mean_token_accuracy": 0.16626126474853772,
|
|
"eval_num_tokens": 13676665.0,
|
|
"eval_runtime": 2.7442,
|
|
"eval_samples_per_second": 1365.431,
|
|
"eval_steps_per_second": 170.907,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.220932197570801,
|
|
"epoch": 6.217895608947805,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002135415470366323,
|
|
"loss": 4.6065,
|
|
"mean_token_accuracy": 0.21874047368764876,
|
|
"num_tokens": 13685774.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.261872863769531,
|
|
"epoch": 6.222038111019056,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021323420504798047,
|
|
"loss": 4.6324,
|
|
"mean_token_accuracy": 0.21970156878232955,
|
|
"num_tokens": 13695700.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.301397752761841,
|
|
"epoch": 6.226180613090307,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00021292698764768675,
|
|
"loss": 4.6432,
|
|
"mean_token_accuracy": 0.21851994693279267,
|
|
"num_tokens": 13704675.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.2542356014251705,
|
|
"epoch": 6.230323115161558,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00021261989545544218,
|
|
"loss": 4.6623,
|
|
"mean_token_accuracy": 0.22553315609693528,
|
|
"num_tokens": 13714562.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.318307209014892,
|
|
"epoch": 6.234465617232809,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021231292909068534,
|
|
"loss": 4.6448,
|
|
"mean_token_accuracy": 0.22140212655067443,
|
|
"num_tokens": 13723917.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.254926586151123,
|
|
"epoch": 6.238608119304059,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021200608917260082,
|
|
"loss": 4.5832,
|
|
"mean_token_accuracy": 0.22523251920938492,
|
|
"num_tokens": 13732482.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.272995328903198,
|
|
"epoch": 6.24275062137531,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021169937632011838,
|
|
"loss": 4.6548,
|
|
"mean_token_accuracy": 0.2194363296031952,
|
|
"num_tokens": 13742110.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.238206005096435,
|
|
"epoch": 6.2468931234465614,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021139279115191123,
|
|
"loss": 4.574,
|
|
"mean_token_accuracy": 0.21389349400997162,
|
|
"num_tokens": 13750501.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.236702632904053,
|
|
"epoch": 6.2510356255178126,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021108633428639534,
|
|
"loss": 4.545,
|
|
"mean_token_accuracy": 0.22685909420251846,
|
|
"num_tokens": 13759786.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.279770469665527,
|
|
"epoch": 6.255178127589064,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002107800063417275,
|
|
"loss": 4.6241,
|
|
"mean_token_accuracy": 0.21795655339956282,
|
|
"num_tokens": 13768476.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.29005618095398,
|
|
"epoch": 6.259320629660315,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002104738079358048,
|
|
"loss": 4.6586,
|
|
"mean_token_accuracy": 0.20849181562662125,
|
|
"num_tokens": 13778721.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.281201648712158,
|
|
"epoch": 6.263463131731566,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002101677396862628,
|
|
"loss": 4.6196,
|
|
"mean_token_accuracy": 0.22399465888738632,
|
|
"num_tokens": 13787108.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.1881067752838135,
|
|
"epoch": 6.267605633802817,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020986180221047463,
|
|
"loss": 4.5398,
|
|
"mean_token_accuracy": 0.23230253010988236,
|
|
"num_tokens": 13795021.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.190041732788086,
|
|
"epoch": 6.271748135874068,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020955599612554966,
|
|
"loss": 4.5887,
|
|
"mean_token_accuracy": 0.2193624496459961,
|
|
"num_tokens": 13804750.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.2463373184204105,
|
|
"epoch": 6.275890637945319,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020925032204833206,
|
|
"loss": 4.5784,
|
|
"mean_token_accuracy": 0.22109659761190414,
|
|
"num_tokens": 13813910.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.329188394546509,
|
|
"epoch": 6.28003314001657,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020894478059539995,
|
|
"loss": 4.5427,
|
|
"mean_token_accuracy": 0.2241484045982361,
|
|
"num_tokens": 13822832.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.204854440689087,
|
|
"epoch": 6.284175642087821,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020863937238306375,
|
|
"loss": 4.5541,
|
|
"mean_token_accuracy": 0.2280479699373245,
|
|
"num_tokens": 13831951.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.172011041641236,
|
|
"epoch": 6.2883181441590725,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020833409802736535,
|
|
"loss": 4.524,
|
|
"mean_token_accuracy": 0.23316972702741623,
|
|
"num_tokens": 13840422.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.306167030334473,
|
|
"epoch": 6.292460646230323,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020802895814407636,
|
|
"loss": 4.6365,
|
|
"mean_token_accuracy": 0.22030848115682602,
|
|
"num_tokens": 13850903.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.244341468811035,
|
|
"epoch": 6.296603148301574,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002077239533486972,
|
|
"loss": 4.6811,
|
|
"mean_token_accuracy": 0.20490736216306688,
|
|
"num_tokens": 13859972.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.235301160812378,
|
|
"epoch": 6.300745650372825,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00020741908425645606,
|
|
"loss": 4.658,
|
|
"mean_token_accuracy": 0.21515794545412065,
|
|
"num_tokens": 13868980.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.259301996231079,
|
|
"epoch": 6.304888152444076,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020711435148230707,
|
|
"loss": 4.5874,
|
|
"mean_token_accuracy": 0.21832506507635116,
|
|
"num_tokens": 13877901.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.245576858520508,
|
|
"epoch": 6.309030654515327,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00020680975564092968,
|
|
"loss": 4.5697,
|
|
"mean_token_accuracy": 0.21962837278842925,
|
|
"num_tokens": 13887081.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.283494472503662,
|
|
"epoch": 6.313173156586578,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00020650529734672686,
|
|
"loss": 4.693,
|
|
"mean_token_accuracy": 0.21509333699941635,
|
|
"num_tokens": 13896645.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.340816831588745,
|
|
"epoch": 6.317315658657829,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020620097721382424,
|
|
"loss": 4.6349,
|
|
"mean_token_accuracy": 0.21796356588602067,
|
|
"num_tokens": 13905975.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.3276678085327145,
|
|
"epoch": 6.3214581607290805,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002058967958560689,
|
|
"loss": 4.6067,
|
|
"mean_token_accuracy": 0.22121513932943343,
|
|
"num_tokens": 13914605.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.256007862091065,
|
|
"epoch": 6.325600662800332,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00020559275388702775,
|
|
"loss": 4.5442,
|
|
"mean_token_accuracy": 0.23139581084251404,
|
|
"num_tokens": 13924592.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.187183284759522,
|
|
"epoch": 6.329743164871583,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00020528885191998682,
|
|
"loss": 4.5649,
|
|
"mean_token_accuracy": 0.22332944571971894,
|
|
"num_tokens": 13933513.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.1399694919586185,
|
|
"epoch": 6.333885666942834,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020498509056794952,
|
|
"loss": 4.4652,
|
|
"mean_token_accuracy": 0.23895961940288543,
|
|
"num_tokens": 13942173.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.177120351791382,
|
|
"epoch": 6.338028169014084,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00020468147044363556,
|
|
"loss": 4.4955,
|
|
"mean_token_accuracy": 0.22771447747945786,
|
|
"num_tokens": 13950205.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.22398190498352,
|
|
"epoch": 6.342170671085335,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020437799215948017,
|
|
"loss": 4.5753,
|
|
"mean_token_accuracy": 0.22749822586774826,
|
|
"num_tokens": 13958956.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.209628582000732,
|
|
"epoch": 6.346313173156586,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002040746563276319,
|
|
"loss": 4.5665,
|
|
"mean_token_accuracy": 0.22033885568380357,
|
|
"num_tokens": 13967718.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.235432434082031,
|
|
"epoch": 6.350455675227837,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020377146355995257,
|
|
"loss": 4.6617,
|
|
"mean_token_accuracy": 0.21480744183063508,
|
|
"num_tokens": 13976841.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.287872982025147,
|
|
"epoch": 6.3545981772990885,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00020346841446801495,
|
|
"loss": 4.5404,
|
|
"mean_token_accuracy": 0.22639259397983552,
|
|
"num_tokens": 13984946.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.257662773132324,
|
|
"epoch": 6.35874067937034,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020316550966310222,
|
|
"loss": 4.6067,
|
|
"mean_token_accuracy": 0.21685410737991334,
|
|
"num_tokens": 13993521.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.183994293212891,
|
|
"epoch": 6.362883181441591,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002028627497562065,
|
|
"loss": 4.5771,
|
|
"mean_token_accuracy": 0.2181665539741516,
|
|
"num_tokens": 14002777.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.246474266052246,
|
|
"epoch": 6.367025683512842,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020256013535802748,
|
|
"loss": 4.6261,
|
|
"mean_token_accuracy": 0.21209287494421006,
|
|
"num_tokens": 14012708.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.28727445602417,
|
|
"epoch": 6.371168185584093,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002022576670789717,
|
|
"loss": 4.6394,
|
|
"mean_token_accuracy": 0.2155505269765854,
|
|
"num_tokens": 14021088.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.280440044403076,
|
|
"epoch": 6.375310687655344,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002019553455291505,
|
|
"loss": 4.6378,
|
|
"mean_token_accuracy": 0.21615593284368514,
|
|
"num_tokens": 14031650.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.240458631515503,
|
|
"epoch": 6.379453189726595,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002016531713183797,
|
|
"loss": 4.6676,
|
|
"mean_token_accuracy": 0.21988083124160768,
|
|
"num_tokens": 14040741.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.242792797088623,
|
|
"epoch": 6.383595691797846,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020135114505617752,
|
|
"loss": 4.6373,
|
|
"mean_token_accuracy": 0.2093031659722328,
|
|
"num_tokens": 14049913.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.277075576782226,
|
|
"epoch": 6.387738193869097,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020104926735176393,
|
|
"loss": 4.5918,
|
|
"mean_token_accuracy": 0.22735825777053834,
|
|
"num_tokens": 14059583.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.327569150924683,
|
|
"epoch": 6.391880695940348,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020074753881405949,
|
|
"loss": 4.5526,
|
|
"mean_token_accuracy": 0.22584011256694794,
|
|
"num_tokens": 14067376.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.247530269622803,
|
|
"epoch": 6.396023198011599,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002004459600516833,
|
|
"loss": 4.6272,
|
|
"mean_token_accuracy": 0.22265007048845292,
|
|
"num_tokens": 14076415.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.252097845077515,
|
|
"epoch": 6.40016570008285,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020014453167295298,
|
|
"loss": 4.6191,
|
|
"mean_token_accuracy": 0.21771289706230162,
|
|
"num_tokens": 14085418.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.246077394485473,
|
|
"epoch": 6.404308202154101,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019984325428588229,
|
|
"loss": 4.6022,
|
|
"mean_token_accuracy": 0.22459561228752137,
|
|
"num_tokens": 14094454.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.269437408447265,
|
|
"epoch": 6.408450704225352,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019954212849818077,
|
|
"loss": 4.5397,
|
|
"mean_token_accuracy": 0.23432922810316087,
|
|
"num_tokens": 14103611.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.17622013092041,
|
|
"epoch": 6.412593206296603,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019924115491725197,
|
|
"loss": 4.4661,
|
|
"mean_token_accuracy": 0.23387826085090638,
|
|
"num_tokens": 14112145.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.237372732162475,
|
|
"epoch": 6.416735708367854,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019894033415019253,
|
|
"loss": 4.6056,
|
|
"mean_token_accuracy": 0.2221802607178688,
|
|
"num_tokens": 14121215.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.171102142333984,
|
|
"epoch": 6.420878210439105,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019863966680379086,
|
|
"loss": 4.4967,
|
|
"mean_token_accuracy": 0.22152459472417832,
|
|
"num_tokens": 14130106.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.18139157295227,
|
|
"epoch": 6.4250207125103564,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001983391534845258,
|
|
"loss": 4.5424,
|
|
"mean_token_accuracy": 0.2322377920150757,
|
|
"num_tokens": 14139071.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.224069595336914,
|
|
"epoch": 6.4291632145816076,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019803879479856547,
|
|
"loss": 4.5974,
|
|
"mean_token_accuracy": 0.21825987547636033,
|
|
"num_tokens": 14148018.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.2009124755859375,
|
|
"epoch": 6.433305716652859,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019773859135176626,
|
|
"loss": 4.5285,
|
|
"mean_token_accuracy": 0.21998629420995713,
|
|
"num_tokens": 14156369.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.215542888641357,
|
|
"epoch": 6.43744821872411,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001974385437496713,
|
|
"loss": 4.6106,
|
|
"mean_token_accuracy": 0.21984836757183074,
|
|
"num_tokens": 14165930.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.2183332443237305,
|
|
"epoch": 6.44159072079536,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00019713865259750945,
|
|
"loss": 4.5777,
|
|
"mean_token_accuracy": 0.228591625392437,
|
|
"num_tokens": 14175557.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.288492584228516,
|
|
"epoch": 6.445733222866611,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019683891850019391,
|
|
"loss": 4.65,
|
|
"mean_token_accuracy": 0.2170352444052696,
|
|
"num_tokens": 14185532.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.372134733200073,
|
|
"epoch": 6.449875724937862,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019653934206232098,
|
|
"loss": 4.7328,
|
|
"mean_token_accuracy": 0.21203465312719344,
|
|
"num_tokens": 14194307.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.317110586166382,
|
|
"epoch": 6.454018227009113,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019623992388816935,
|
|
"loss": 4.6817,
|
|
"mean_token_accuracy": 0.22098743617534639,
|
|
"num_tokens": 14203900.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.264908695220948,
|
|
"epoch": 6.458160729080364,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001959406645816979,
|
|
"loss": 4.604,
|
|
"mean_token_accuracy": 0.22145358771085738,
|
|
"num_tokens": 14213232.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.2155252456665036,
|
|
"epoch": 6.4623032311516155,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019564156474654554,
|
|
"loss": 4.545,
|
|
"mean_token_accuracy": 0.21945291459560395,
|
|
"num_tokens": 14221365.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.183708381652832,
|
|
"epoch": 6.466445733222867,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019534262498602928,
|
|
"loss": 4.5722,
|
|
"mean_token_accuracy": 0.21912614554166793,
|
|
"num_tokens": 14229878.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.247611427307129,
|
|
"epoch": 6.470588235294118,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00019504384590314318,
|
|
"loss": 4.6163,
|
|
"mean_token_accuracy": 0.21504329293966293,
|
|
"num_tokens": 14239132.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.27608289718628,
|
|
"epoch": 6.474730737365369,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00019474522810055745,
|
|
"loss": 4.6315,
|
|
"mean_token_accuracy": 0.22245015799999238,
|
|
"num_tokens": 14247116.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.237215900421143,
|
|
"epoch": 6.47887323943662,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00019444677218061673,
|
|
"loss": 4.5876,
|
|
"mean_token_accuracy": 0.22378402203321457,
|
|
"num_tokens": 14255423.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.22880802154541,
|
|
"epoch": 6.483015741507871,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001941484787453393,
|
|
"loss": 4.5576,
|
|
"mean_token_accuracy": 0.22750293612480163,
|
|
"num_tokens": 14263836.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.1975545406341555,
|
|
"epoch": 6.487158243579122,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019385034839641558,
|
|
"loss": 4.6818,
|
|
"mean_token_accuracy": 0.21816913187503814,
|
|
"num_tokens": 14273460.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.288527727127075,
|
|
"epoch": 6.491300745650372,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001935523817352071,
|
|
"loss": 4.6991,
|
|
"mean_token_accuracy": 0.21707435250282286,
|
|
"num_tokens": 14282925.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.283282852172851,
|
|
"epoch": 6.4954432477216235,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019325457936274516,
|
|
"loss": 4.5817,
|
|
"mean_token_accuracy": 0.22622769176959992,
|
|
"num_tokens": 14292059.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.274545192718506,
|
|
"epoch": 6.499585749792875,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00019295694187972955,
|
|
"loss": 4.6371,
|
|
"mean_token_accuracy": 0.2125675469636917,
|
|
"num_tokens": 14301276.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.247177505493164,
|
|
"epoch": 6.503728251864126,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019265946988652783,
|
|
"loss": 4.6386,
|
|
"mean_token_accuracy": 0.21370143294334412,
|
|
"num_tokens": 14310259.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.227504730224609,
|
|
"epoch": 6.507870753935377,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001923621639831733,
|
|
"loss": 4.5105,
|
|
"mean_token_accuracy": 0.22504902780056,
|
|
"num_tokens": 14318782.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.173471689224243,
|
|
"epoch": 6.512013256006628,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019206502476936467,
|
|
"loss": 4.4963,
|
|
"mean_token_accuracy": 0.23206298500299455,
|
|
"num_tokens": 14327571.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.135555934906006,
|
|
"epoch": 6.516155758077879,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019176805284446414,
|
|
"loss": 4.5319,
|
|
"mean_token_accuracy": 0.23313156962394715,
|
|
"num_tokens": 14336974.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.291125202178955,
|
|
"epoch": 6.52029826014913,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019147124880749633,
|
|
"loss": 4.6926,
|
|
"mean_token_accuracy": 0.218401962518692,
|
|
"num_tokens": 14346367.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.287627267837524,
|
|
"epoch": 6.524440762220381,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019117461325714768,
|
|
"loss": 4.6304,
|
|
"mean_token_accuracy": 0.22132983505725862,
|
|
"num_tokens": 14355641.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.246808195114136,
|
|
"epoch": 6.528583264291632,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019087814679176436,
|
|
"loss": 4.5621,
|
|
"mean_token_accuracy": 0.22451686710119248,
|
|
"num_tokens": 14364927.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.234672212600708,
|
|
"epoch": 6.5327257663628835,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019058185000935175,
|
|
"loss": 4.5399,
|
|
"mean_token_accuracy": 0.22587109208106995,
|
|
"num_tokens": 14373370.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.254017162322998,
|
|
"epoch": 6.536868268434135,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019028572350757278,
|
|
"loss": 4.6124,
|
|
"mean_token_accuracy": 0.21654532849788666,
|
|
"num_tokens": 14381820.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.2422840118408205,
|
|
"epoch": 6.541010770505386,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018998976788374682,
|
|
"loss": 4.5875,
|
|
"mean_token_accuracy": 0.21863880306482314,
|
|
"num_tokens": 14391158.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.201485252380371,
|
|
"epoch": 6.545153272576636,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00018969398373484898,
|
|
"loss": 4.6299,
|
|
"mean_token_accuracy": 0.21829961985349655,
|
|
"num_tokens": 14399945.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.200627136230469,
|
|
"epoch": 6.549295774647887,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018939837165750796,
|
|
"loss": 4.5447,
|
|
"mean_token_accuracy": 0.22442623525857924,
|
|
"num_tokens": 14410275.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.29403018951416,
|
|
"epoch": 6.553438276719138,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00018910293224800584,
|
|
"loss": 4.6482,
|
|
"mean_token_accuracy": 0.2173393338918686,
|
|
"num_tokens": 14420369.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.291232109069824,
|
|
"epoch": 6.557580778790389,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018880766610227608,
|
|
"loss": 4.6234,
|
|
"mean_token_accuracy": 0.22409190386533737,
|
|
"num_tokens": 14429078.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.234523487091065,
|
|
"epoch": 6.56172328086164,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018851257381590264,
|
|
"loss": 4.5824,
|
|
"mean_token_accuracy": 0.22088503539562226,
|
|
"num_tokens": 14438920.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.292790508270263,
|
|
"epoch": 6.5658657829328915,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018821765598411917,
|
|
"loss": 4.7699,
|
|
"mean_token_accuracy": 0.20645501464605331,
|
|
"num_tokens": 14449164.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.241003036499023,
|
|
"epoch": 6.570008285004143,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00018792291320180693,
|
|
"loss": 4.6977,
|
|
"mean_token_accuracy": 0.2070661887526512,
|
|
"num_tokens": 14459418.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.294680643081665,
|
|
"epoch": 6.574150787075394,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018762834606349455,
|
|
"loss": 4.5906,
|
|
"mean_token_accuracy": 0.22685577124357223,
|
|
"num_tokens": 14468691.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.322669315338135,
|
|
"epoch": 6.578293289146645,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00018733395516335588,
|
|
"loss": 4.6731,
|
|
"mean_token_accuracy": 0.21569546610116958,
|
|
"num_tokens": 14477936.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.285602855682373,
|
|
"epoch": 6.582435791217896,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018703974109520976,
|
|
"loss": 4.6442,
|
|
"mean_token_accuracy": 0.22226325422525406,
|
|
"num_tokens": 14487191.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.2095410346984865,
|
|
"epoch": 6.586578293289147,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018674570445251798,
|
|
"loss": 4.566,
|
|
"mean_token_accuracy": 0.2255881741642952,
|
|
"num_tokens": 14495731.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.200072622299194,
|
|
"epoch": 6.590720795360397,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018645184582838465,
|
|
"loss": 4.5668,
|
|
"mean_token_accuracy": 0.22047847658395767,
|
|
"num_tokens": 14505009.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.211784410476684,
|
|
"epoch": 6.594863297431648,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018615816581555466,
|
|
"loss": 4.614,
|
|
"mean_token_accuracy": 0.21996732503175737,
|
|
"num_tokens": 14513843.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.188229322433472,
|
|
"epoch": 6.5990057995028994,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001858646650064128,
|
|
"loss": 4.6074,
|
|
"mean_token_accuracy": 0.22044946551322936,
|
|
"num_tokens": 14523405.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.254159069061279,
|
|
"epoch": 6.603148301574151,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018557134399298215,
|
|
"loss": 4.6781,
|
|
"mean_token_accuracy": 0.21682086437940598,
|
|
"num_tokens": 14532615.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.2646036624908445,
|
|
"epoch": 6.607290803645402,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0001852782033669233,
|
|
"loss": 4.6101,
|
|
"mean_token_accuracy": 0.22458707094192504,
|
|
"num_tokens": 14541297.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.216501903533936,
|
|
"epoch": 6.611433305716653,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018498524371953295,
|
|
"loss": 4.556,
|
|
"mean_token_accuracy": 0.22257563024759291,
|
|
"num_tokens": 14549867.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.22313437461853,
|
|
"epoch": 6.615575807787904,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018469246564174275,
|
|
"loss": 4.6508,
|
|
"mean_token_accuracy": 0.20906404107809068,
|
|
"num_tokens": 14558708.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.215989065170288,
|
|
"epoch": 6.619718309859155,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000184399869724118,
|
|
"loss": 4.648,
|
|
"mean_token_accuracy": 0.21317926198244094,
|
|
"num_tokens": 14567917.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.194528484344483,
|
|
"epoch": 6.623860811930406,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00018410745655685662,
|
|
"loss": 4.5161,
|
|
"mean_token_accuracy": 0.23143290728330612,
|
|
"num_tokens": 14576417.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.19383282661438,
|
|
"epoch": 6.628003314001657,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00018381522672978806,
|
|
"loss": 4.5967,
|
|
"mean_token_accuracy": 0.2205110549926758,
|
|
"num_tokens": 14585172.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.628003314001657,
|
|
"eval_entropy": 5.144060653409978,
|
|
"eval_loss": 5.7370734214782715,
|
|
"eval_mean_token_accuracy": 0.1679431346020719,
|
|
"eval_num_tokens": 14585172.0,
|
|
"eval_runtime": 2.746,
|
|
"eval_samples_per_second": 1364.549,
|
|
"eval_steps_per_second": 170.796,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.2945750713348385,
|
|
"epoch": 6.632145816072908,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018352318083237167,
|
|
"loss": 4.6209,
|
|
"mean_token_accuracy": 0.21357694566249846,
|
|
"num_tokens": 14593758.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.269504880905151,
|
|
"epoch": 6.636288318144159,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018323131945369603,
|
|
"loss": 4.5957,
|
|
"mean_token_accuracy": 0.2212613955140114,
|
|
"num_tokens": 14602558.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.292761898040771,
|
|
"epoch": 6.6404308202154105,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018293964318247735,
|
|
"loss": 4.5895,
|
|
"mean_token_accuracy": 0.22187672108411788,
|
|
"num_tokens": 14611463.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.268460607528686,
|
|
"epoch": 6.644573322286661,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00018264815260705863,
|
|
"loss": 4.6303,
|
|
"mean_token_accuracy": 0.22261828631162645,
|
|
"num_tokens": 14619951.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.312491035461425,
|
|
"epoch": 6.648715824357912,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018235684831540826,
|
|
"loss": 4.6601,
|
|
"mean_token_accuracy": 0.2227204129099846,
|
|
"num_tokens": 14629332.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.215784406661987,
|
|
"epoch": 6.652858326429163,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018206573089511862,
|
|
"loss": 4.6116,
|
|
"mean_token_accuracy": 0.22559183984994888,
|
|
"num_tokens": 14639010.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.217174816131592,
|
|
"epoch": 6.657000828500414,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00018177480093340564,
|
|
"loss": 4.644,
|
|
"mean_token_accuracy": 0.2131648376584053,
|
|
"num_tokens": 14648507.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.26403660774231,
|
|
"epoch": 6.661143330571665,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001814840590171065,
|
|
"loss": 4.6395,
|
|
"mean_token_accuracy": 0.216891747713089,
|
|
"num_tokens": 14657892.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.276500272750854,
|
|
"epoch": 6.665285832642916,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018119350573267979,
|
|
"loss": 4.5962,
|
|
"mean_token_accuracy": 0.22690920978784562,
|
|
"num_tokens": 14666460.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.25149884223938,
|
|
"epoch": 6.669428334714167,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018090314166620302,
|
|
"loss": 4.5902,
|
|
"mean_token_accuracy": 0.2189852774143219,
|
|
"num_tokens": 14675046.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.219921970367432,
|
|
"epoch": 6.6735708367854185,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018061296740337225,
|
|
"loss": 4.6258,
|
|
"mean_token_accuracy": 0.2165866807103157,
|
|
"num_tokens": 14685055.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.248022747039795,
|
|
"epoch": 6.67771333885667,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018032298352950077,
|
|
"loss": 4.5972,
|
|
"mean_token_accuracy": 0.2188133031129837,
|
|
"num_tokens": 14695371.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.225955820083618,
|
|
"epoch": 6.681855840927921,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018003319062951756,
|
|
"loss": 4.569,
|
|
"mean_token_accuracy": 0.2223624899983406,
|
|
"num_tokens": 14703671.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.195835924148559,
|
|
"epoch": 6.685998342999172,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017974358928796682,
|
|
"loss": 4.5951,
|
|
"mean_token_accuracy": 0.22712722718715667,
|
|
"num_tokens": 14712527.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.231344604492188,
|
|
"epoch": 6.690140845070422,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017945418008900594,
|
|
"loss": 4.6311,
|
|
"mean_token_accuracy": 0.22212356179952622,
|
|
"num_tokens": 14720711.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.25670371055603,
|
|
"epoch": 6.694283347141674,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001791649636164049,
|
|
"loss": 4.6664,
|
|
"mean_token_accuracy": 0.21566284000873565,
|
|
"num_tokens": 14729667.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.32373685836792,
|
|
"epoch": 6.698425849212924,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000178875940453545,
|
|
"loss": 4.612,
|
|
"mean_token_accuracy": 0.2260928839445114,
|
|
"num_tokens": 14738614.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.285801553726197,
|
|
"epoch": 6.702568351284175,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017858711118341742,
|
|
"loss": 4.611,
|
|
"mean_token_accuracy": 0.22119891494512559,
|
|
"num_tokens": 14747724.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.189185285568238,
|
|
"epoch": 6.7067108533554265,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017829847638862255,
|
|
"loss": 4.5783,
|
|
"mean_token_accuracy": 0.22308655232191085,
|
|
"num_tokens": 14757436.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.163452911376953,
|
|
"epoch": 6.710853355426678,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017801003665136823,
|
|
"loss": 4.5621,
|
|
"mean_token_accuracy": 0.22859434187412261,
|
|
"num_tokens": 14765948.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.2175110340118405,
|
|
"epoch": 6.714995857497929,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017772179255346883,
|
|
"loss": 4.5924,
|
|
"mean_token_accuracy": 0.21859526336193086,
|
|
"num_tokens": 14775397.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.2649805545806885,
|
|
"epoch": 6.71913835956918,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017743374467634433,
|
|
"loss": 4.5547,
|
|
"mean_token_accuracy": 0.22480292171239852,
|
|
"num_tokens": 14783605.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.262083387374878,
|
|
"epoch": 6.723280861640431,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017714589360101862,
|
|
"loss": 4.6719,
|
|
"mean_token_accuracy": 0.2111577272415161,
|
|
"num_tokens": 14792210.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.165537548065186,
|
|
"epoch": 6.727423363711682,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000176858239908119,
|
|
"loss": 4.5663,
|
|
"mean_token_accuracy": 0.21831904202699662,
|
|
"num_tokens": 14801186.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.171479225158691,
|
|
"epoch": 6.731565865782933,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017657078417787414,
|
|
"loss": 4.5126,
|
|
"mean_token_accuracy": 0.2303812935948372,
|
|
"num_tokens": 14810100.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.1626139163970945,
|
|
"epoch": 6.735708367854184,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017628352699011396,
|
|
"loss": 4.4932,
|
|
"mean_token_accuracy": 0.22342629581689835,
|
|
"num_tokens": 14819335.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.184713125228882,
|
|
"epoch": 6.739850869925435,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001759964689242674,
|
|
"loss": 4.6115,
|
|
"mean_token_accuracy": 0.22678097039461137,
|
|
"num_tokens": 14828686.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.233674144744873,
|
|
"epoch": 6.743993371996686,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001757096105593619,
|
|
"loss": 4.6202,
|
|
"mean_token_accuracy": 0.21709106266498565,
|
|
"num_tokens": 14838312.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.149096870422364,
|
|
"epoch": 6.748135874067937,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017542295247402234,
|
|
"loss": 4.5595,
|
|
"mean_token_accuracy": 0.22249870449304582,
|
|
"num_tokens": 14846838.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.202677154541016,
|
|
"epoch": 6.752278376139188,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00017513649524646919,
|
|
"loss": 4.6182,
|
|
"mean_token_accuracy": 0.21352985352277756,
|
|
"num_tokens": 14857427.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.231053304672241,
|
|
"epoch": 6.756420878210439,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001748502394545181,
|
|
"loss": 4.6142,
|
|
"mean_token_accuracy": 0.2143697887659073,
|
|
"num_tokens": 14866244.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.2318687915802,
|
|
"epoch": 6.76056338028169,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001745641856755783,
|
|
"loss": 4.6272,
|
|
"mean_token_accuracy": 0.2168770670890808,
|
|
"num_tokens": 14875697.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.213436222076416,
|
|
"epoch": 6.764705882352941,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017427833448665136,
|
|
"loss": 4.6431,
|
|
"mean_token_accuracy": 0.22060779482126236,
|
|
"num_tokens": 14884650.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.221727609634399,
|
|
"epoch": 6.768848384424192,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017399268646433065,
|
|
"loss": 4.5726,
|
|
"mean_token_accuracy": 0.22035471200942994,
|
|
"num_tokens": 14894058.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.291379785537719,
|
|
"epoch": 6.772990886495443,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017370724218479912,
|
|
"loss": 4.6928,
|
|
"mean_token_accuracy": 0.21716551035642623,
|
|
"num_tokens": 14903219.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.280435705184937,
|
|
"epoch": 6.7771333885666944,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00017342200222382934,
|
|
"loss": 4.6104,
|
|
"mean_token_accuracy": 0.21606069803237915,
|
|
"num_tokens": 14913051.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.284905195236206,
|
|
"epoch": 6.781275890637946,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017313696715678134,
|
|
"loss": 4.6263,
|
|
"mean_token_accuracy": 0.2211444079875946,
|
|
"num_tokens": 14921777.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.336038494110108,
|
|
"epoch": 6.785418392709197,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000172852137558602,
|
|
"loss": 4.7612,
|
|
"mean_token_accuracy": 0.20792590975761413,
|
|
"num_tokens": 14931558.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.2495917797088625,
|
|
"epoch": 6.789560894780448,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017256751400382384,
|
|
"loss": 4.6749,
|
|
"mean_token_accuracy": 0.2186842069029808,
|
|
"num_tokens": 14940584.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.260048294067383,
|
|
"epoch": 6.793703396851699,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017228309706656354,
|
|
"loss": 4.6003,
|
|
"mean_token_accuracy": 0.2234319806098938,
|
|
"num_tokens": 14950830.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.309694099426269,
|
|
"epoch": 6.797845898922949,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001719988873205213,
|
|
"loss": 4.6046,
|
|
"mean_token_accuracy": 0.2206121563911438,
|
|
"num_tokens": 14958990.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.258207225799561,
|
|
"epoch": 6.8019884009942,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017171488533897916,
|
|
"loss": 4.6389,
|
|
"mean_token_accuracy": 0.216607928276062,
|
|
"num_tokens": 14967451.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.220820093154908,
|
|
"epoch": 6.806130903065451,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017143109169480005,
|
|
"loss": 4.5703,
|
|
"mean_token_accuracy": 0.2230973109602928,
|
|
"num_tokens": 14976157.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.277642917633057,
|
|
"epoch": 6.810273405136702,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017114750696042695,
|
|
"loss": 4.6916,
|
|
"mean_token_accuracy": 0.211100971698761,
|
|
"num_tokens": 14985591.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.277077102661133,
|
|
"epoch": 6.8144159072079535,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017086413170788107,
|
|
"loss": 4.5868,
|
|
"mean_token_accuracy": 0.22545621395111085,
|
|
"num_tokens": 14994513.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.249056625366211,
|
|
"epoch": 6.818558409279205,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017058096650876143,
|
|
"loss": 4.6555,
|
|
"mean_token_accuracy": 0.21752652078866958,
|
|
"num_tokens": 15003252.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.268651342391967,
|
|
"epoch": 6.822700911350456,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017029801193424313,
|
|
"loss": 4.5678,
|
|
"mean_token_accuracy": 0.21832710206508638,
|
|
"num_tokens": 15012342.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.17280683517456,
|
|
"epoch": 6.826843413421707,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017001526855507632,
|
|
"loss": 4.5248,
|
|
"mean_token_accuracy": 0.22667909115552903,
|
|
"num_tokens": 15021336.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.232009410858154,
|
|
"epoch": 6.830985915492958,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016973273694158555,
|
|
"loss": 4.5752,
|
|
"mean_token_accuracy": 0.21583533883094788,
|
|
"num_tokens": 15031016.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.281693077087402,
|
|
"epoch": 6.835128417564209,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016945041766366773,
|
|
"loss": 4.5846,
|
|
"mean_token_accuracy": 0.22162794917821885,
|
|
"num_tokens": 15039181.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.255547618865966,
|
|
"epoch": 6.83927091963546,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016916831129079187,
|
|
"loss": 4.647,
|
|
"mean_token_accuracy": 0.2128545254468918,
|
|
"num_tokens": 15048563.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.248698711395264,
|
|
"epoch": 6.84341342170671,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001688864183919973,
|
|
"loss": 4.5936,
|
|
"mean_token_accuracy": 0.22442127764225006,
|
|
"num_tokens": 15056960.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.307019329071045,
|
|
"epoch": 6.8475559237779615,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016860473953589266,
|
|
"loss": 4.6347,
|
|
"mean_token_accuracy": 0.21388418227434158,
|
|
"num_tokens": 15065645.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.223530387878418,
|
|
"epoch": 6.851698425849213,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00016832327529065522,
|
|
"loss": 4.5822,
|
|
"mean_token_accuracy": 0.22318441718816756,
|
|
"num_tokens": 15074544.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.267363739013672,
|
|
"epoch": 6.855840927920464,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001680420262240289,
|
|
"loss": 4.6343,
|
|
"mean_token_accuracy": 0.21753862798213958,
|
|
"num_tokens": 15083543.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.204693603515625,
|
|
"epoch": 6.859983429991715,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00016776099290332397,
|
|
"loss": 4.6145,
|
|
"mean_token_accuracy": 0.2235179975628853,
|
|
"num_tokens": 15092291.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.3212086200714115,
|
|
"epoch": 6.864125932062966,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016748017589541525,
|
|
"loss": 4.6697,
|
|
"mean_token_accuracy": 0.2193456619977951,
|
|
"num_tokens": 15100875.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.275527429580689,
|
|
"epoch": 6.868268434134217,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001671995757667414,
|
|
"loss": 4.6246,
|
|
"mean_token_accuracy": 0.22207115739583969,
|
|
"num_tokens": 15111385.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.325754880905151,
|
|
"epoch": 6.872410936205468,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016691919308330343,
|
|
"loss": 4.6412,
|
|
"mean_token_accuracy": 0.2226441338658333,
|
|
"num_tokens": 15121153.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.278840351104736,
|
|
"epoch": 6.876553438276719,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016663902841066404,
|
|
"loss": 4.6258,
|
|
"mean_token_accuracy": 0.21953531950712205,
|
|
"num_tokens": 15130132.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.229125213623047,
|
|
"epoch": 6.88069594034797,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016635908231394587,
|
|
"loss": 4.6111,
|
|
"mean_token_accuracy": 0.2153362438082695,
|
|
"num_tokens": 15138878.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.198756742477417,
|
|
"epoch": 6.8848384424192215,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001660793553578307,
|
|
"loss": 4.6868,
|
|
"mean_token_accuracy": 0.20835813581943513,
|
|
"num_tokens": 15148727.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.257393503189087,
|
|
"epoch": 6.888980944490473,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016579984810655856,
|
|
"loss": 4.6404,
|
|
"mean_token_accuracy": 0.21736776530742646,
|
|
"num_tokens": 15158107.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.275184488296508,
|
|
"epoch": 6.893123446561724,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016552056112392594,
|
|
"loss": 4.6178,
|
|
"mean_token_accuracy": 0.21456918865442276,
|
|
"num_tokens": 15168101.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.249782991409302,
|
|
"epoch": 6.897265948632974,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001652414949732854,
|
|
"loss": 4.5573,
|
|
"mean_token_accuracy": 0.22505220770835876,
|
|
"num_tokens": 15176376.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.2404684066772464,
|
|
"epoch": 6.901408450704225,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016496265021754365,
|
|
"loss": 4.5925,
|
|
"mean_token_accuracy": 0.2170558288693428,
|
|
"num_tokens": 15184936.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.271342658996582,
|
|
"epoch": 6.905550952775476,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016468402741916124,
|
|
"loss": 4.6586,
|
|
"mean_token_accuracy": 0.22344124019145967,
|
|
"num_tokens": 15194846.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.263687086105347,
|
|
"epoch": 6.909693454846727,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001644056271401506,
|
|
"loss": 4.6187,
|
|
"mean_token_accuracy": 0.22206373363733292,
|
|
"num_tokens": 15204543.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.221840810775757,
|
|
"epoch": 6.913835956917978,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00016412744994207573,
|
|
"loss": 4.5897,
|
|
"mean_token_accuracy": 0.215163554251194,
|
|
"num_tokens": 15213459.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.209072399139404,
|
|
"epoch": 6.9179784589892295,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016384949638605025,
|
|
"loss": 4.6864,
|
|
"mean_token_accuracy": 0.20790564268827438,
|
|
"num_tokens": 15223841.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.225527238845825,
|
|
"epoch": 6.922120961060481,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00016357176703273696,
|
|
"loss": 4.6187,
|
|
"mean_token_accuracy": 0.22101123183965682,
|
|
"num_tokens": 15233594.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.251284885406494,
|
|
"epoch": 6.926263463131732,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001632942624423463,
|
|
"loss": 4.5643,
|
|
"mean_token_accuracy": 0.22933652848005295,
|
|
"num_tokens": 15242876.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.297107553482055,
|
|
"epoch": 6.930405965202983,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016301698317463525,
|
|
"loss": 4.6565,
|
|
"mean_token_accuracy": 0.2228442072868347,
|
|
"num_tokens": 15251581.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.252490663528443,
|
|
"epoch": 6.934548467274234,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016273992978890658,
|
|
"loss": 4.5084,
|
|
"mean_token_accuracy": 0.22431502342224122,
|
|
"num_tokens": 15260601.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.232750654220581,
|
|
"epoch": 6.938690969345485,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001624631028440071,
|
|
"loss": 4.5322,
|
|
"mean_token_accuracy": 0.22127106189727783,
|
|
"num_tokens": 15270123.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.301280069351196,
|
|
"epoch": 6.942833471416735,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001621865028983271,
|
|
"loss": 4.6664,
|
|
"mean_token_accuracy": 0.2162109360098839,
|
|
"num_tokens": 15278623.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.187017822265625,
|
|
"epoch": 6.946975973487987,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001619101305097988,
|
|
"loss": 4.569,
|
|
"mean_token_accuracy": 0.22274162173271178,
|
|
"num_tokens": 15287789.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.105927991867065,
|
|
"epoch": 6.9511184755592375,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00016163398623589563,
|
|
"loss": 4.5276,
|
|
"mean_token_accuracy": 0.23134084939956664,
|
|
"num_tokens": 15296266.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.268940114974976,
|
|
"epoch": 6.955260977630489,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00016135807063363074,
|
|
"loss": 4.6269,
|
|
"mean_token_accuracy": 0.21198562532663345,
|
|
"num_tokens": 15304785.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.171991634368896,
|
|
"epoch": 6.95940347970174,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016108238425955584,
|
|
"loss": 4.5327,
|
|
"mean_token_accuracy": 0.22577450573444366,
|
|
"num_tokens": 15313727.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.254316186904907,
|
|
"epoch": 6.963545981772991,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001608069276697608,
|
|
"loss": 4.7158,
|
|
"mean_token_accuracy": 0.21332326233386995,
|
|
"num_tokens": 15322911.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.300360822677613,
|
|
"epoch": 6.967688483844242,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001605317014198714,
|
|
"loss": 4.6163,
|
|
"mean_token_accuracy": 0.22070427536964415,
|
|
"num_tokens": 15330882.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.221141958236695,
|
|
"epoch": 6.971830985915493,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016025670606504926,
|
|
"loss": 4.523,
|
|
"mean_token_accuracy": 0.22398052662611007,
|
|
"num_tokens": 15339915.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.213225173950195,
|
|
"epoch": 6.975973487986744,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001599819421599899,
|
|
"loss": 4.5878,
|
|
"mean_token_accuracy": 0.21864742040634155,
|
|
"num_tokens": 15348181.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.188560724258423,
|
|
"epoch": 6.980115990057995,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015970741025892233,
|
|
"loss": 4.5452,
|
|
"mean_token_accuracy": 0.2271549642086029,
|
|
"num_tokens": 15357128.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.209831953048706,
|
|
"epoch": 6.984258492129246,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00015943311091560721,
|
|
"loss": 4.6538,
|
|
"mean_token_accuracy": 0.22213561683893204,
|
|
"num_tokens": 15366878.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.3097316265106205,
|
|
"epoch": 6.988400994200497,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00015915904468333635,
|
|
"loss": 4.5939,
|
|
"mean_token_accuracy": 0.22847030758857728,
|
|
"num_tokens": 15376101.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.18035774230957,
|
|
"epoch": 6.9925434962717485,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015888521211493135,
|
|
"loss": 4.5421,
|
|
"mean_token_accuracy": 0.21641024351119995,
|
|
"num_tokens": 15384942.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.193719911575317,
|
|
"epoch": 6.996685998342999,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001586116137627423,
|
|
"loss": 4.5648,
|
|
"mean_token_accuracy": 0.22552343606948852,
|
|
"num_tokens": 15394380.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.28035192489624,
|
|
"epoch": 7.00082850041425,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015833825017864711,
|
|
"loss": 4.4966,
|
|
"mean_token_accuracy": 0.23881841003894805,
|
|
"num_tokens": 15402180.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.25820837020874,
|
|
"epoch": 7.004971002485501,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015806512191405,
|
|
"loss": 4.5534,
|
|
"mean_token_accuracy": 0.22912415564060212,
|
|
"num_tokens": 15411434.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.271618843078613,
|
|
"epoch": 7.009113504556752,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001577922295198803,
|
|
"loss": 4.4939,
|
|
"mean_token_accuracy": 0.2291717991232872,
|
|
"num_tokens": 15419905.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.211508417129517,
|
|
"epoch": 7.013256006628003,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015751957354659206,
|
|
"loss": 4.5173,
|
|
"mean_token_accuracy": 0.22768738716840745,
|
|
"num_tokens": 15429200.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.230689239501953,
|
|
"epoch": 7.017398508699254,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00015724715454416206,
|
|
"loss": 4.5014,
|
|
"mean_token_accuracy": 0.23492475599050522,
|
|
"num_tokens": 15439328.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.239945983886718,
|
|
"epoch": 7.021541010770505,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015697497306208926,
|
|
"loss": 4.449,
|
|
"mean_token_accuracy": 0.23373402953147887,
|
|
"num_tokens": 15447739.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.2335375308990475,
|
|
"epoch": 7.0256835128417565,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001567030296493934,
|
|
"loss": 4.4117,
|
|
"mean_token_accuracy": 0.2391123041510582,
|
|
"num_tokens": 15456598.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.264567565917969,
|
|
"epoch": 7.029826014913008,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015643132485461395,
|
|
"loss": 4.5407,
|
|
"mean_token_accuracy": 0.22584614008665085,
|
|
"num_tokens": 15466633.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.269935941696167,
|
|
"epoch": 7.033968516984259,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001561598592258094,
|
|
"loss": 4.4839,
|
|
"mean_token_accuracy": 0.23932840377092363,
|
|
"num_tokens": 15476809.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.200196313858032,
|
|
"epoch": 7.03811101905551,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015588863331055541,
|
|
"loss": 4.4345,
|
|
"mean_token_accuracy": 0.22763314247131347,
|
|
"num_tokens": 15484964.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.208463573455811,
|
|
"epoch": 7.042253521126761,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001556176476559444,
|
|
"loss": 4.4668,
|
|
"mean_token_accuracy": 0.2335176944732666,
|
|
"num_tokens": 15493600.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.042253521126761,
|
|
"eval_entropy": 5.129485646798921,
|
|
"eval_loss": 5.728693962097168,
|
|
"eval_mean_token_accuracy": 0.16830202490726767,
|
|
"eval_num_tokens": 15493600.0,
|
|
"eval_runtime": 2.7433,
|
|
"eval_samples_per_second": 1365.87,
|
|
"eval_steps_per_second": 170.962,
|
|
"step": 8500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 12060,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2.3124410665344e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|