18dcce3b89
Model: fpadovani/swe-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.4142502071251036,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1382675779584000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|