llama-30b-math_hard / training_logs.json
joshuaclymer's picture
Upload folder using huggingface_hub
232c380
raw
history blame
25 kB
[
{
"loss": 0.7313,
"learning_rate": 0.0002,
"epoch": 0.02,
"step": 1
},
{
"loss": 0.6794,
"learning_rate": 0.0002,
"epoch": 0.04,
"step": 2
},
{
"loss": 0.6862,
"learning_rate": 0.0002,
"epoch": 0.05,
"step": 3
},
{
"loss": 0.6799,
"learning_rate": 0.0002,
"epoch": 0.07,
"step": 4
},
{
"loss": 0.6825,
"learning_rate": 0.0002,
"epoch": 0.09,
"step": 5
},
{
"loss": 0.6914,
"learning_rate": 0.0002,
"epoch": 0.11,
"step": 6
},
{
"loss": 0.6325,
"learning_rate": 0.0002,
"epoch": 0.12,
"step": 7
},
{
"loss": 0.6633,
"learning_rate": 0.0002,
"epoch": 0.14,
"step": 8
},
{
"loss": 0.6252,
"learning_rate": 0.0002,
"epoch": 0.16,
"step": 9
},
{
"loss": 0.5798,
"learning_rate": 0.0002,
"epoch": 0.18,
"step": 10
},
{
"loss": 0.5536,
"learning_rate": 0.0002,
"epoch": 0.19,
"step": 11
},
{
"loss": 0.5727,
"learning_rate": 0.0002,
"epoch": 0.21,
"step": 12
},
{
"loss": 0.4765,
"learning_rate": 0.0002,
"epoch": 0.23,
"step": 13
},
{
"loss": 0.3769,
"learning_rate": 0.0002,
"epoch": 0.25,
"step": 14
},
{
"loss": 0.2708,
"learning_rate": 0.0002,
"epoch": 0.26,
"step": 15
},
{
"loss": 0.2132,
"learning_rate": 0.0002,
"epoch": 0.28,
"step": 16
},
{
"loss": 0.2204,
"learning_rate": 0.0002,
"epoch": 0.3,
"step": 17
},
{
"loss": 0.3512,
"learning_rate": 0.0002,
"epoch": 0.32,
"step": 18
},
{
"loss": 0.1736,
"learning_rate": 0.0002,
"epoch": 0.33,
"step": 19
},
{
"loss": 0.3119,
"learning_rate": 0.0002,
"epoch": 0.35,
"step": 20
},
{
"loss": 0.1878,
"learning_rate": 0.0002,
"epoch": 0.37,
"step": 21
},
{
"loss": 0.2817,
"learning_rate": 0.0002,
"epoch": 0.39,
"step": 22
},
{
"loss": 0.4016,
"learning_rate": 0.0002,
"epoch": 0.4,
"step": 23
},
{
"loss": 0.156,
"learning_rate": 0.0002,
"epoch": 0.42,
"step": 24
},
{
"loss": 0.1408,
"learning_rate": 0.0002,
"epoch": 0.44,
"step": 25
},
{
"eval_math_hard_loss": 0.2689576745033264,
"eval_math_hard_score": -0.07232501357793808,
"eval_math_hard_brier_score": 0.07232501357793808,
"eval_math_hard_average_probability": 0.8480575084686279,
"eval_math_hard_accuracy": 0.91,
"eval_math_hard_probabilities": [
0.12067427486181259,
0.1707480251789093,
0.22421833872795105,
0.9965168237686157,
0.9968488812446594,
0.8897665739059448,
0.8757681846618652,
0.9160223007202148,
0.4495812952518463,
0.9733548760414124,
0.8542622327804565,
0.8829213976860046,
0.7457072734832764,
0.7373321056365967,
0.7001582980155945,
0.9495631456375122,
0.9757802486419678,
0.9814878702163696,
0.8223016262054443,
0.8257706761360168,
0.7114197611808777,
0.9997543692588806,
0.9997970461845398,
0.9997699856758118,
0.9983037710189819,
0.9911226034164429,
0.9890031218528748,
0.9952235817909241,
0.9955497980117798,
0.9985560774803162,
0.9976275563240051,
0.9978576302528381,
0.9976431727409363,
0.9857828617095947,
0.9856427311897278,
0.9890527129173279,
0.9993459582328796,
0.99762362241745,
0.9985696077346802,
0.9795147180557251,
0.9767017364501953,
0.9040936827659607,
0.989809513092041,
0.9877877831459045,
0.9814877510070801,
0.9997876286506653,
0.9965195655822754,
0.9839633107185364,
0.44732266664505005,
0.5352095365524292,
0.5794621109962463,
0.9967998266220093,
0.9975603818893433,
0.9964653253555298,
0.9998288154602051,
0.9998891353607178,
0.9999051094055176,
0.2663218379020691,
0.4094083905220032,
0.7260100245475769,
0.9965078234672546,
0.9969045519828796,
0.998040497303009,
0.9707159399986267,
0.9864614605903625,
0.9651244878768921,
0.9996368885040283,
0.9995032548904419,
0.9982764720916748,
0.6076152920722961,
0.585483729839325,
0.5877071619033813,
0.9988491535186768,
0.9986512064933777,
0.9980589747428894,
0.9256494045257568,
0.9805311560630798,
0.9392337799072266,
0.7655142545700073,
0.6185799241065979,
0.541168212890625,
0.782398521900177,
0.8448206782341003,
0.7403358817100525,
0.8463240265846252,
0.8631186485290527,
0.8114821314811707,
0.9599024057388306,
0.7725688815116882,
0.3576231896877289,
0.027788693085312843,
0.8065083622932434,
0.8443471789360046,
0.8439624905586243,
0.727486789226532,
0.8130761384963989,
0.9100518822669983,
0.9845901727676392,
0.9753082990646362,
0.9655715823173523
],
"eval_math_hard_runtime": 86.0301,
"eval_math_hard_samples_per_second": 1.162,
"eval_math_hard_steps_per_second": 0.046,
"epoch": 0.44,
"step": 25
},
{
"loss": 0.1166,
"learning_rate": 0.0002,
"epoch": 0.46,
"step": 26
},
{
"loss": 0.1934,
"learning_rate": 0.0002,
"epoch": 0.47,
"step": 27
},
{
"loss": 0.1695,
"learning_rate": 0.0002,
"epoch": 0.49,
"step": 28
},
{
"loss": 0.1415,
"learning_rate": 0.0002,
"epoch": 0.51,
"step": 29
},
{
"loss": 0.2946,
"learning_rate": 0.0002,
"epoch": 0.53,
"step": 30
},
{
"loss": 0.1945,
"learning_rate": 0.0002,
"epoch": 0.54,
"step": 31
},
{
"loss": 0.1672,
"learning_rate": 0.0002,
"epoch": 0.56,
"step": 32
},
{
"loss": 0.1961,
"learning_rate": 0.0002,
"epoch": 0.58,
"step": 33
},
{
"loss": 0.3305,
"learning_rate": 0.0002,
"epoch": 0.6,
"step": 34
},
{
"loss": 0.0823,
"learning_rate": 0.0002,
"epoch": 0.61,
"step": 35
},
{
"loss": 0.1063,
"learning_rate": 0.0002,
"epoch": 0.63,
"step": 36
},
{
"loss": 0.1566,
"learning_rate": 0.0002,
"epoch": 0.65,
"step": 37
},
{
"loss": 0.1957,
"learning_rate": 0.0002,
"epoch": 0.67,
"step": 38
},
{
"loss": 0.0823,
"learning_rate": 0.0002,
"epoch": 0.68,
"step": 39
},
{
"loss": 0.0602,
"learning_rate": 0.0002,
"epoch": 0.7,
"step": 40
},
{
"loss": 0.1042,
"learning_rate": 0.0002,
"epoch": 0.72,
"step": 41
},
{
"loss": 0.0696,
"learning_rate": 0.0002,
"epoch": 0.74,
"step": 42
},
{
"loss": 0.0618,
"learning_rate": 0.0002,
"epoch": 0.75,
"step": 43
},
{
"loss": 0.1516,
"learning_rate": 0.0002,
"epoch": 0.77,
"step": 44
},
{
"loss": 0.1877,
"learning_rate": 0.0002,
"epoch": 0.79,
"step": 45
},
{
"loss": 0.2814,
"learning_rate": 0.0002,
"epoch": 0.81,
"step": 46
},
{
"loss": 0.3869,
"learning_rate": 0.0002,
"epoch": 0.82,
"step": 47
},
{
"loss": 0.0865,
"learning_rate": 0.0002,
"epoch": 0.84,
"step": 48
},
{
"loss": 0.5889,
"learning_rate": 0.0002,
"epoch": 0.86,
"step": 49
},
{
"loss": 0.108,
"learning_rate": 0.0002,
"epoch": 0.88,
"step": 50
},
{
"eval_math_hard_loss": 0.4195532202720642,
"eval_math_hard_score": -0.08378518372774124,
"eval_math_hard_brier_score": 0.08378518372774124,
"eval_math_hard_average_probability": 0.8607721924781799,
"eval_math_hard_accuracy": 0.89,
"eval_math_hard_probabilities": [
0.001347638200968504,
0.0014310380211099982,
0.0292666032910347,
0.9999281167984009,
0.9999291896820068,
0.9515820741653442,
0.9991993308067322,
0.9985817670822144,
0.4427696764469147,
0.9999922513961792,
0.5870345830917358,
0.9338187575340271,
0.9995662569999695,
0.999679446220398,
0.9990984201431274,
0.8128757476806641,
0.8317935466766357,
0.8578028082847595,
0.767571747303009,
0.7276169061660767,
0.16569289565086365,
0.9999608993530273,
0.9999557733535767,
0.9999984502792358,
0.9999668598175049,
0.9999934434890747,
0.9996861219406128,
0.999996542930603,
0.9999967813491821,
0.9999997615814209,
0.999997615814209,
0.9999953508377075,
0.9999957084655762,
0.9973000884056091,
0.9996415376663208,
0.9997121691703796,
0.9999990463256836,
0.9999995231628418,
0.9999998807907104,
0.9920561909675598,
0.9743956327438354,
0.518930196762085,
0.9998225569725037,
0.9935445785522461,
0.9992994070053101,
0.9982439279556274,
0.9694525599479675,
0.8038814067840576,
0.5726821422576904,
0.4458751678466797,
0.741835355758667,
0.9999998807907104,
1.0,
0.9999977350234985,
0.9999512434005737,
1.0,
1.0,
0.1753261685371399,
0.2425578236579895,
0.8445780873298645,
0.9961126446723938,
0.9984958171844482,
0.9912378191947937,
0.9592325687408447,
0.9976963400840759,
0.866991400718689,
0.9999998807907104,
0.9999804496765137,
0.9998711347579956,
0.26644328236579895,
0.829546332359314,
0.8019995093345642,
0.9999985694885254,
0.9999955892562866,
0.9999771118164062,
0.9985451698303223,
0.9999727010726929,
0.9993960857391357,
0.9808396100997925,
0.8794113993644714,
0.8073936700820923,
0.963763952255249,
0.9844830632209778,
0.958784818649292,
0.939505934715271,
0.9660537838935852,
0.817690372467041,
0.9990904331207275,
0.7585221529006958,
0.4643564522266388,
0.11916442960500717,
0.9887877106666565,
0.9978222846984863,
0.8583523631095886,
0.8125313520431519,
0.8561255931854248,
0.8449611067771912,
0.9997431635856628,
0.9992627501487732,
0.9998737573623657
],
"eval_math_hard_runtime": 86.0735,
"eval_math_hard_samples_per_second": 1.162,
"eval_math_hard_steps_per_second": 0.046,
"epoch": 0.88,
"step": 50
},
{
"loss": 0.0879,
"learning_rate": 0.0002,
"epoch": 0.89,
"step": 51
},
{
"loss": 0.2006,
"learning_rate": 0.0002,
"epoch": 0.91,
"step": 52
},
{
"loss": 0.1101,
"learning_rate": 0.0002,
"epoch": 0.93,
"step": 53
},
{
"loss": 0.1182,
"learning_rate": 0.0002,
"epoch": 0.95,
"step": 54
},
{
"loss": 0.1047,
"learning_rate": 0.0002,
"epoch": 0.96,
"step": 55
},
{
"loss": 0.1168,
"learning_rate": 0.0002,
"epoch": 0.98,
"step": 56
},
{
"loss": 0.1458,
"learning_rate": 0.0002,
"epoch": 1.0,
"step": 57
},
{
"loss": 0.0369,
"learning_rate": 0.0002,
"epoch": 1.02,
"step": 58
},
{
"loss": 0.1703,
"learning_rate": 0.0002,
"epoch": 1.04,
"step": 59
},
{
"loss": 0.0705,
"learning_rate": 0.0002,
"epoch": 1.05,
"step": 60
},
{
"loss": 0.0306,
"learning_rate": 0.0002,
"epoch": 1.07,
"step": 61
},
{
"loss": 0.0199,
"learning_rate": 0.0002,
"epoch": 1.09,
"step": 62
},
{
"loss": 0.0059,
"learning_rate": 0.0002,
"epoch": 1.11,
"step": 63
},
{
"loss": 0.0019,
"learning_rate": 0.0002,
"epoch": 1.12,
"step": 64
},
{
"loss": 0.05,
"learning_rate": 0.0002,
"epoch": 1.14,
"step": 65
},
{
"loss": 0.0192,
"learning_rate": 0.0002,
"epoch": 1.16,
"step": 66
},
{
"loss": 0.2126,
"learning_rate": 0.0002,
"epoch": 1.18,
"step": 67
},
{
"loss": 0.0073,
"learning_rate": 0.0002,
"epoch": 1.19,
"step": 68
},
{
"loss": 0.0883,
"learning_rate": 0.0002,
"epoch": 1.21,
"step": 69
},
{
"loss": 0.0251,
"learning_rate": 0.0002,
"epoch": 1.23,
"step": 70
},
{
"loss": 0.0673,
"learning_rate": 0.0002,
"epoch": 1.25,
"step": 71
},
{
"loss": 0.0042,
"learning_rate": 0.0002,
"epoch": 1.26,
"step": 72
},
{
"loss": 0.0769,
"learning_rate": 0.0002,
"epoch": 1.28,
"step": 73
},
{
"loss": 0.1407,
"learning_rate": 0.0002,
"epoch": 1.3,
"step": 74
},
{
"loss": 0.0613,
"learning_rate": 0.0002,
"epoch": 1.32,
"step": 75
},
{
"eval_math_hard_loss": 0.40808790922164917,
"eval_math_hard_score": -0.07436995208263397,
"eval_math_hard_brier_score": 0.07436995208263397,
"eval_math_hard_average_probability": 0.9035561084747314,
"eval_math_hard_accuracy": 0.92,
"eval_math_hard_probabilities": [
0.01734217070043087,
0.025694051757454872,
0.1909039467573166,
1.0,
1.0,
0.9912930130958557,
0.9999858140945435,
0.9999847412109375,
0.9637160897254944,
1.0,
0.9999996423721313,
0.9999971389770508,
0.9771614074707031,
0.9833316802978516,
0.966616153717041,
0.9999768733978271,
0.9999898672103882,
0.9999890327453613,
0.9881917834281921,
0.9601098895072937,
0.08695478737354279,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9999998807907104,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9999986886978149,
0.9999992847442627,
0.6397180557250977,
0.9994465708732605,
0.9973899722099304,
0.9993732571601868,
1.0,
0.9999997615814209,
0.99986732006073,
0.9971725940704346,
0.8597153425216675,
0.9993921518325806,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.0010394210694357753,
0.04478495568037033,
0.9842627644538879,
0.9999969005584717,
0.9999995231628418,
0.709723174571991,
0.9994927644729614,
0.9999569654464722,
0.9997255206108093,
1.0,
1.0,
1.0,
0.98405921459198,
0.9982985854148865,
0.9936298727989197,
1.0,
0.9999997615814209,
0.9999988079071045,
0.9999998807907104,
1.0,
0.9999998807907104,
0.9991475343704224,
0.8117725253105164,
0.20917673408985138,
0.9854962229728699,
0.998820960521698,
0.9700117111206055,
0.7556695938110352,
0.8359659314155579,
0.9189376831054688,
1.0,
0.8505204916000366,
0.6733924150466919,
3.9892434870125726e-05,
0.9999310970306396,
0.9999845027923584,
0.9993377327919006,
0.9925682544708252,
0.9994389414787292,
0.9971237778663635,
1.0,
1.0,
0.9999998807907104
],
"eval_math_hard_runtime": 86.0452,
"eval_math_hard_samples_per_second": 1.162,
"eval_math_hard_steps_per_second": 0.046,
"epoch": 1.32,
"step": 75
},
{
"loss": 0.0338,
"learning_rate": 0.0002,
"epoch": 1.33,
"step": 76
},
{
"loss": 0.0115,
"learning_rate": 0.0002,
"epoch": 1.35,
"step": 77
},
{
"loss": 0.0876,
"learning_rate": 0.0002,
"epoch": 1.37,
"step": 78
},
{
"loss": 0.0071,
"learning_rate": 0.0002,
"epoch": 1.39,
"step": 79
},
{
"loss": 0.0739,
"learning_rate": 0.0002,
"epoch": 1.4,
"step": 80
},
{
"loss": 0.0016,
"learning_rate": 0.0002,
"epoch": 1.42,
"step": 81
},
{
"loss": 0.1257,
"learning_rate": 0.0002,
"epoch": 1.44,
"step": 82
},
{
"loss": 0.0048,
"learning_rate": 0.0002,
"epoch": 1.46,
"step": 83
},
{
"loss": 0.007,
"learning_rate": 0.0002,
"epoch": 1.47,
"step": 84
},
{
"loss": 0.0162,
"learning_rate": 0.0002,
"epoch": 1.49,
"step": 85
},
{
"loss": 0.042,
"learning_rate": 0.0002,
"epoch": 1.51,
"step": 86
},
{
"loss": 0.0231,
"learning_rate": 0.0002,
"epoch": 1.53,
"step": 87
},
{
"loss": 0.0103,
"learning_rate": 0.0002,
"epoch": 1.54,
"step": 88
},
{
"loss": 0.0077,
"learning_rate": 0.0002,
"epoch": 1.56,
"step": 89
},
{
"loss": 0.0014,
"learning_rate": 0.0002,
"epoch": 1.58,
"step": 90
},
{
"loss": 0.014,
"learning_rate": 0.0002,
"epoch": 1.6,
"step": 91
},
{
"loss": 0.0054,
"learning_rate": 0.0002,
"epoch": 1.61,
"step": 92
},
{
"loss": 0.0349,
"learning_rate": 0.0002,
"epoch": 1.63,
"step": 93
},
{
"loss": 0.0961,
"learning_rate": 0.0002,
"epoch": 1.65,
"step": 94
},
{
"loss": 0.0058,
"learning_rate": 0.0002,
"epoch": 1.67,
"step": 95
},
{
"loss": 0.0311,
"learning_rate": 0.0002,
"epoch": 1.68,
"step": 96
},
{
"loss": 0.0143,
"learning_rate": 0.0002,
"epoch": 1.7,
"step": 97
},
{
"loss": 0.0007,
"learning_rate": 0.0002,
"epoch": 1.72,
"step": 98
},
{
"loss": 0.1104,
"learning_rate": 0.0002,
"epoch": 1.74,
"step": 99
},
{
"loss": 0.0337,
"learning_rate": 0.0002,
"epoch": 1.75,
"step": 100
},
{
"eval_math_hard_loss": 0.7644428014755249,
"eval_math_hard_score": -0.1077132597565651,
"eval_math_hard_brier_score": 0.1077132597565651,
"eval_math_hard_average_probability": 0.8629195690155029,
"eval_math_hard_accuracy": 0.88,
"eval_math_hard_probabilities": [
0.01672576367855072,
0.0001243828737642616,
0.09437736123800278,
1.0,
1.0,
0.9999958276748657,
1.0,
1.0,
0.9949105381965637,
0.9999998807907104,
0.9999997615814209,
0.9999969005584717,
0.9774806499481201,
0.9809224009513855,
0.06269947439432144,
0.9999384880065918,
1.0,
1.0,
0.7058337330818176,
0.6948679089546204,
0.33711233735084534,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.13868287205696106,
0.9946752786636353,
0.7222106456756592,
0.864886999130249,
1.0,
1.0,
0.9999991655349731,
0.9912755489349365,
0.07769276201725006,
0.996044397354126,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
5.921328556723893e-05,
0.054774124175310135,
0.8164430856704712,
1.0,
1.0,
0.538241446018219,
0.9993982315063477,
0.9931572079658508,
0.999657154083252,
1.0,
1.0,
1.0,
0.7599838376045227,
0.9545750617980957,
0.9728327989578247,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9995208978652954,
0.7313249111175537,
0.005399945192039013,
0.5453264713287354,
0.9951422214508057,
0.9464260935783386,
0.43800926208496094,
0.897794246673584,
0.9954219460487366,
1.0,
0.9996411800384521,
0.9987420439720154,
1.534920940349238e-10,
0.9999021291732788,
0.9999223947525024,
0.9999954700469971,
0.9998113512992859,
0.9999969005584717,
0.9999964237213135,
1.0,
1.0,
1.0
],
"eval_math_hard_runtime": 86.022,
"eval_math_hard_samples_per_second": 1.162,
"eval_math_hard_steps_per_second": 0.046,
"epoch": 1.75,
"step": 100
},
{
"train_runtime": 8937.8559,
"train_samples_per_second": 0.358,
"train_steps_per_second": 0.011,
"total_flos": 0.0,
"train_loss": 0.18627140538301318,
"epoch": 1.75,
"step": 100
}
]