3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.033516931636631775,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.74203462600708,
|
|
"epoch": 0.00011172310545543924,
|
|
"grad_norm": 7.40625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.545,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 4264.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.74201259613037,
|
|
"epoch": 0.00022344621091087847,
|
|
"grad_norm": 6.84375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.5105,
|
|
"mean_token_accuracy": 0.0002816901309415698,
|
|
"num_tokens": 8413.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742000961303711,
|
|
"epoch": 0.0003351693163663177,
|
|
"grad_norm": 6.78125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4826,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 12394.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742033767700196,
|
|
"epoch": 0.00044689242182175694,
|
|
"grad_norm": 5.90625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.4451,
|
|
"mean_token_accuracy": 0.0004269361263141036,
|
|
"num_tokens": 16610.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742008590698243,
|
|
"epoch": 0.0005586155272771962,
|
|
"grad_norm": 5.9375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.3813,
|
|
"mean_token_accuracy": 0.0017635513795539737,
|
|
"num_tokens": 21243.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.74174461364746,
|
|
"epoch": 0.0006703386327326354,
|
|
"grad_norm": 5.1875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.2219,
|
|
"mean_token_accuracy": 0.019105370296165346,
|
|
"num_tokens": 25571.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741024494171143,
|
|
"epoch": 0.0007820617381880747,
|
|
"grad_norm": 4.75,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.0233,
|
|
"mean_token_accuracy": 0.0476666621863842,
|
|
"num_tokens": 29412.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.739530563354492,
|
|
"epoch": 0.0008937848436435139,
|
|
"grad_norm": 3.75,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.943,
|
|
"mean_token_accuracy": 0.04776074178516865,
|
|
"num_tokens": 33893.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.736891174316407,
|
|
"epoch": 0.0010055079490989532,
|
|
"grad_norm": 3.09375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.7721,
|
|
"mean_token_accuracy": 0.05200008656829595,
|
|
"num_tokens": 38043.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.733825874328613,
|
|
"epoch": 0.0011172310545543925,
|
|
"grad_norm": 2.75,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7095,
|
|
"mean_token_accuracy": 0.05134744010865688,
|
|
"num_tokens": 42222.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.731979370117188,
|
|
"epoch": 0.0012289541600098315,
|
|
"grad_norm": 2.765625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.6035,
|
|
"mean_token_accuracy": 0.05346306636929512,
|
|
"num_tokens": 46321.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.729666996002198,
|
|
"epoch": 0.0013406772654652708,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5985,
|
|
"mean_token_accuracy": 0.053918526880443095,
|
|
"num_tokens": 50816.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.726247787475586,
|
|
"epoch": 0.0014524003709207101,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.5526,
|
|
"mean_token_accuracy": 0.05777559094130993,
|
|
"num_tokens": 55331.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.722254943847656,
|
|
"epoch": 0.0015641234763761494,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.4754,
|
|
"mean_token_accuracy": 0.060246657207608224,
|
|
"num_tokens": 59379.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.717647266387939,
|
|
"epoch": 0.0016758465818315885,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.3716,
|
|
"mean_token_accuracy": 0.05563458241522312,
|
|
"num_tokens": 63463.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.709985065460206,
|
|
"epoch": 0.0017875696872870278,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.4054,
|
|
"mean_token_accuracy": 0.04782464448362589,
|
|
"num_tokens": 67897.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.698591709136963,
|
|
"epoch": 0.001899292792742467,
|
|
"grad_norm": 2.703125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.2081,
|
|
"mean_token_accuracy": 0.05416969805955887,
|
|
"num_tokens": 71995.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.679736518859864,
|
|
"epoch": 0.0020110158981979064,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.0655,
|
|
"mean_token_accuracy": 0.06419909372925758,
|
|
"num_tokens": 75745.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.639767360687255,
|
|
"epoch": 0.0021227390036533456,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.0156,
|
|
"mean_token_accuracy": 0.06227423809468746,
|
|
"num_tokens": 79641.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.604651832580567,
|
|
"epoch": 0.002234462109108785,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.9827,
|
|
"mean_token_accuracy": 0.07007253468036652,
|
|
"num_tokens": 84368.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.576740074157716,
|
|
"epoch": 0.0023461852145642242,
|
|
"grad_norm": 3.0,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.9633,
|
|
"mean_token_accuracy": 0.06716309636831283,
|
|
"num_tokens": 89177.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.532594108581543,
|
|
"epoch": 0.002457908320019663,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.7625,
|
|
"mean_token_accuracy": 0.06927448399364948,
|
|
"num_tokens": 93266.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.469588470458984,
|
|
"epoch": 0.0025696314254751024,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.6738,
|
|
"mean_token_accuracy": 0.0684173908084631,
|
|
"num_tokens": 97385.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.414745044708251,
|
|
"epoch": 0.0026813545309305417,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.4918,
|
|
"mean_token_accuracy": 0.06782151162624359,
|
|
"num_tokens": 100911.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.343620300292969,
|
|
"epoch": 0.002793077636385981,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.4104,
|
|
"mean_token_accuracy": 0.07619427219033241,
|
|
"num_tokens": 104785.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.282330799102784,
|
|
"epoch": 0.0029048007418414202,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.3421,
|
|
"mean_token_accuracy": 0.0715565849095583,
|
|
"num_tokens": 109434.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.177853965759278,
|
|
"epoch": 0.0030165238472968595,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.3045,
|
|
"mean_token_accuracy": 0.07821577228605747,
|
|
"num_tokens": 113869.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.01875,
|
|
"epoch": 0.003128246952752299,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.2069,
|
|
"mean_token_accuracy": 0.07112646363675594,
|
|
"num_tokens": 118332.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.93814001083374,
|
|
"epoch": 0.003239970058207738,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.0893,
|
|
"mean_token_accuracy": 0.06839369647204877,
|
|
"num_tokens": 122140.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.799712753295898,
|
|
"epoch": 0.003351693163663177,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.9249,
|
|
"mean_token_accuracy": 0.067073168233037,
|
|
"num_tokens": 126515.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.603694915771484,
|
|
"epoch": 0.0034634162691186163,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.897,
|
|
"mean_token_accuracy": 0.07342451699078083,
|
|
"num_tokens": 130397.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.402643203735352,
|
|
"epoch": 0.0035751393745740555,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.815,
|
|
"mean_token_accuracy": 0.07550931498408317,
|
|
"num_tokens": 134428.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.27815113067627,
|
|
"epoch": 0.003686862480029495,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.6921,
|
|
"mean_token_accuracy": 0.08235657699406147,
|
|
"num_tokens": 138294.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.126440715789794,
|
|
"epoch": 0.003798585585484934,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.6466,
|
|
"mean_token_accuracy": 0.0780983179807663,
|
|
"num_tokens": 142750.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.904253101348877,
|
|
"epoch": 0.003910308690940373,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.5833,
|
|
"mean_token_accuracy": 0.06832590401172638,
|
|
"num_tokens": 147220.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.704329586029052,
|
|
"epoch": 0.004022031796395813,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.464,
|
|
"mean_token_accuracy": 0.08249393478035927,
|
|
"num_tokens": 151396.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.544238567352295,
|
|
"epoch": 0.0041337549018512516,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4503,
|
|
"mean_token_accuracy": 0.07549951747059822,
|
|
"num_tokens": 155336.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.501252269744873,
|
|
"epoch": 0.004245478007306691,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4842,
|
|
"mean_token_accuracy": 0.07688803523778916,
|
|
"num_tokens": 159752.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.432751655578613,
|
|
"epoch": 0.00435720111276213,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.5426,
|
|
"mean_token_accuracy": 0.07533748783171176,
|
|
"num_tokens": 163895.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.257897663116456,
|
|
"epoch": 0.00446892421821757,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.3421,
|
|
"mean_token_accuracy": 0.08197390548884868,
|
|
"num_tokens": 168312.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.259009075164794,
|
|
"epoch": 0.004580647323673009,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3701,
|
|
"mean_token_accuracy": 0.08555959761142731,
|
|
"num_tokens": 172358.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.100054454803466,
|
|
"epoch": 0.0046923704291284484,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.357,
|
|
"mean_token_accuracy": 0.08119111023843288,
|
|
"num_tokens": 177085.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.163977813720702,
|
|
"epoch": 0.004804093534583887,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.4691,
|
|
"mean_token_accuracy": 0.07897469773888588,
|
|
"num_tokens": 181410.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.064214897155761,
|
|
"epoch": 0.004915816640039326,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2085,
|
|
"mean_token_accuracy": 0.08347514495253563,
|
|
"num_tokens": 185649.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.9723944664001465,
|
|
"epoch": 0.005027539745494766,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2571,
|
|
"mean_token_accuracy": 0.08833960294723511,
|
|
"num_tokens": 190055.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.958705711364746,
|
|
"epoch": 0.005139262850950205,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2784,
|
|
"mean_token_accuracy": 0.08142814487218857,
|
|
"num_tokens": 194267.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.980498743057251,
|
|
"epoch": 0.0052509859564056445,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2042,
|
|
"mean_token_accuracy": 0.08119908273220063,
|
|
"num_tokens": 198483.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.903439283370972,
|
|
"epoch": 0.005362709061861083,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1137,
|
|
"mean_token_accuracy": 0.09260561317205429,
|
|
"num_tokens": 202514.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.820244312286377,
|
|
"epoch": 0.005474432167316523,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2591,
|
|
"mean_token_accuracy": 0.08432262316346169,
|
|
"num_tokens": 206748.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.817278146743774,
|
|
"epoch": 0.005586155272771962,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1337,
|
|
"mean_token_accuracy": 0.08795059770345688,
|
|
"num_tokens": 210981.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.902036476135254,
|
|
"epoch": 0.005697878378227402,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.2718,
|
|
"mean_token_accuracy": 0.08792312145233154,
|
|
"num_tokens": 215551.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.720875978469849,
|
|
"epoch": 0.0058096014836828405,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0933,
|
|
"mean_token_accuracy": 0.09327159300446511,
|
|
"num_tokens": 219677.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.8553773880004885,
|
|
"epoch": 0.005921324589138279,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1915,
|
|
"mean_token_accuracy": 0.08988388553261757,
|
|
"num_tokens": 223457.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.8154174327850345,
|
|
"epoch": 0.006033047694593719,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.026,
|
|
"mean_token_accuracy": 0.10610299035906792,
|
|
"num_tokens": 227135.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.805374622344971,
|
|
"epoch": 0.006144770800049158,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2242,
|
|
"mean_token_accuracy": 0.0845766805112362,
|
|
"num_tokens": 231393.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.793313646316529,
|
|
"epoch": 0.006256493905504598,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2039,
|
|
"mean_token_accuracy": 0.09036076590418815,
|
|
"num_tokens": 235959.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.797818899154663,
|
|
"epoch": 0.0063682170109600365,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0391,
|
|
"mean_token_accuracy": 0.10008666068315505,
|
|
"num_tokens": 239875.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.643310832977295,
|
|
"epoch": 0.006479940116415476,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1816,
|
|
"mean_token_accuracy": 0.10088859647512435,
|
|
"num_tokens": 244397.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.749200963973999,
|
|
"epoch": 0.006591663221870915,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1022,
|
|
"mean_token_accuracy": 0.09432346150279045,
|
|
"num_tokens": 248441.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.724449968338012,
|
|
"epoch": 0.006703386327326354,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.1223,
|
|
"mean_token_accuracy": 0.09665866494178772,
|
|
"num_tokens": 252693.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.719059801101684,
|
|
"epoch": 0.006815109432781794,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0852,
|
|
"mean_token_accuracy": 0.0973274439573288,
|
|
"num_tokens": 257135.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.689913558959961,
|
|
"epoch": 0.0069268325382372325,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1788,
|
|
"mean_token_accuracy": 0.09386017993092537,
|
|
"num_tokens": 261919.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.723158597946167,
|
|
"epoch": 0.007038555643692672,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0899,
|
|
"mean_token_accuracy": 0.10055064782500267,
|
|
"num_tokens": 266656.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.58135347366333,
|
|
"epoch": 0.007150278749148111,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0484,
|
|
"mean_token_accuracy": 0.0989685259759426,
|
|
"num_tokens": 270879.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.633032703399659,
|
|
"epoch": 0.007262001854603551,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.0017,
|
|
"mean_token_accuracy": 0.11045164242386818,
|
|
"num_tokens": 274884.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.55345048904419,
|
|
"epoch": 0.00737372496005899,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9434,
|
|
"mean_token_accuracy": 0.10104935392737388,
|
|
"num_tokens": 279096.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.626837635040284,
|
|
"epoch": 0.007485448065514429,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0338,
|
|
"mean_token_accuracy": 0.10009614899754524,
|
|
"num_tokens": 282923.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.528380966186523,
|
|
"epoch": 0.007597171170969868,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.99,
|
|
"mean_token_accuracy": 0.10532780215144158,
|
|
"num_tokens": 287283.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.572710466384888,
|
|
"epoch": 0.007708894276425307,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9504,
|
|
"mean_token_accuracy": 0.10530123338103295,
|
|
"num_tokens": 291409.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.547561979293823,
|
|
"epoch": 0.007820617381880746,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9732,
|
|
"mean_token_accuracy": 0.10650015398859977,
|
|
"num_tokens": 295157.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.669625616073608,
|
|
"epoch": 0.007932340487336187,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0005,
|
|
"mean_token_accuracy": 0.10243007764220238,
|
|
"num_tokens": 299370.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.543743228912353,
|
|
"epoch": 0.008044063592791625,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.0093,
|
|
"mean_token_accuracy": 0.11244696080684662,
|
|
"num_tokens": 303955.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.570756626129151,
|
|
"epoch": 0.008155786698247064,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9899,
|
|
"mean_token_accuracy": 0.10562622845172882,
|
|
"num_tokens": 308047.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.594445466995239,
|
|
"epoch": 0.008267509803702503,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9313,
|
|
"mean_token_accuracy": 0.11213268861174583,
|
|
"num_tokens": 312050.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.5787379264831545,
|
|
"epoch": 0.008379232909157944,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.9007,
|
|
"mean_token_accuracy": 0.11451447457075119,
|
|
"num_tokens": 315901.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.617064571380615,
|
|
"epoch": 0.008490956014613383,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.9233,
|
|
"mean_token_accuracy": 0.11718286573886871,
|
|
"num_tokens": 320077.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.479556846618652,
|
|
"epoch": 0.008602679120068821,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8783,
|
|
"mean_token_accuracy": 0.11937321722507477,
|
|
"num_tokens": 324365.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.485933542251587,
|
|
"epoch": 0.00871440222552426,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8292,
|
|
"mean_token_accuracy": 0.11517802253365517,
|
|
"num_tokens": 327971.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.489345073699951,
|
|
"epoch": 0.0088261253309797,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8634,
|
|
"mean_token_accuracy": 0.12188869342207909,
|
|
"num_tokens": 331791.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.405192232131958,
|
|
"epoch": 0.00893784843643514,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9151,
|
|
"mean_token_accuracy": 0.11866991743445396,
|
|
"num_tokens": 335616.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.442100572586059,
|
|
"epoch": 0.009049571541890579,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8337,
|
|
"mean_token_accuracy": 0.1216049998998642,
|
|
"num_tokens": 339815.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.466716241836548,
|
|
"epoch": 0.009161294647346017,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8153,
|
|
"mean_token_accuracy": 0.11885289028286934,
|
|
"num_tokens": 344045.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.439525604248047,
|
|
"epoch": 0.009273017752801456,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.9127,
|
|
"mean_token_accuracy": 0.1201499916613102,
|
|
"num_tokens": 348726.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.421639108657837,
|
|
"epoch": 0.009384740858256897,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8973,
|
|
"mean_token_accuracy": 0.11855003610253334,
|
|
"num_tokens": 353228.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.533071041107178,
|
|
"epoch": 0.009496463963712336,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8732,
|
|
"mean_token_accuracy": 0.12777991741895675,
|
|
"num_tokens": 357379.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.404365539550781,
|
|
"epoch": 0.009608187069167775,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8237,
|
|
"mean_token_accuracy": 0.11873817816376686,
|
|
"num_tokens": 361347.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.3829724311828615,
|
|
"epoch": 0.009719910174623213,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8001,
|
|
"mean_token_accuracy": 0.12097886875271797,
|
|
"num_tokens": 365364.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.4662988662719725,
|
|
"epoch": 0.009831633280078652,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8435,
|
|
"mean_token_accuracy": 0.1184915505349636,
|
|
"num_tokens": 369648.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.550151824951172,
|
|
"epoch": 0.009943356385534093,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8211,
|
|
"mean_token_accuracy": 0.1224476382136345,
|
|
"num_tokens": 373734.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.477874803543091,
|
|
"epoch": 0.010055079490989532,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8174,
|
|
"mean_token_accuracy": 0.12541017457842826,
|
|
"num_tokens": 378184.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.454969882965088,
|
|
"epoch": 0.01016680259644497,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.9275,
|
|
"mean_token_accuracy": 0.11591533869504929,
|
|
"num_tokens": 382686.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.344128131866455,
|
|
"epoch": 0.01027852570190041,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8348,
|
|
"mean_token_accuracy": 0.12774229273200036,
|
|
"num_tokens": 387210.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.421209812164307,
|
|
"epoch": 0.01039024880735585,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8216,
|
|
"mean_token_accuracy": 0.12031811997294425,
|
|
"num_tokens": 391247.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.465836143493652,
|
|
"epoch": 0.010501971912811289,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6907,
|
|
"mean_token_accuracy": 0.13249015137553216,
|
|
"num_tokens": 395188.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.280446100234985,
|
|
"epoch": 0.010613695018266728,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7666,
|
|
"mean_token_accuracy": 0.12789778113365174,
|
|
"num_tokens": 399513.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.51575026512146,
|
|
"epoch": 0.010725418123722167,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8438,
|
|
"mean_token_accuracy": 0.12305677309632301,
|
|
"num_tokens": 404052.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.260741233825684,
|
|
"epoch": 0.010837141229177606,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6724,
|
|
"mean_token_accuracy": 0.13464880287647246,
|
|
"num_tokens": 408299.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.3084382057189945,
|
|
"epoch": 0.010948864334633046,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6808,
|
|
"mean_token_accuracy": 0.12868183255195617,
|
|
"num_tokens": 412376.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.318713665008545,
|
|
"epoch": 0.011060587440088485,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7365,
|
|
"mean_token_accuracy": 0.12830623611807823,
|
|
"num_tokens": 417010.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.343731689453125,
|
|
"epoch": 0.011172310545543924,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.615,
|
|
"mean_token_accuracy": 0.1360365241765976,
|
|
"num_tokens": 421167.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.227418041229248,
|
|
"epoch": 0.011284033650999363,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.715,
|
|
"mean_token_accuracy": 0.1310766190290451,
|
|
"num_tokens": 425942.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.354942321777344,
|
|
"epoch": 0.011395756756454803,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7458,
|
|
"mean_token_accuracy": 0.13017034903168678,
|
|
"num_tokens": 430529.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.211854124069214,
|
|
"epoch": 0.011507479861910242,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6012,
|
|
"mean_token_accuracy": 0.12464086860418319,
|
|
"num_tokens": 434464.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.229535865783691,
|
|
"epoch": 0.011619202967365681,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.72,
|
|
"mean_token_accuracy": 0.13534524589776992,
|
|
"num_tokens": 438763.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.300835609436035,
|
|
"epoch": 0.01173092607282112,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.67,
|
|
"mean_token_accuracy": 0.12867324203252792,
|
|
"num_tokens": 442945.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.314332723617554,
|
|
"epoch": 0.011842649178276559,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7556,
|
|
"mean_token_accuracy": 0.12306396216154099,
|
|
"num_tokens": 447301.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.266288375854492,
|
|
"epoch": 0.011954372283732,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6174,
|
|
"mean_token_accuracy": 0.13486616015434266,
|
|
"num_tokens": 451365.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.152557897567749,
|
|
"epoch": 0.012066095389187438,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6467,
|
|
"mean_token_accuracy": 0.13019582405686378,
|
|
"num_tokens": 455469.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.219053745269775,
|
|
"epoch": 0.012177818494642877,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.7099,
|
|
"mean_token_accuracy": 0.13144526258111,
|
|
"num_tokens": 459641.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.226452398300171,
|
|
"epoch": 0.012289541600098316,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6922,
|
|
"mean_token_accuracy": 0.13080282509326935,
|
|
"num_tokens": 463866.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.169753932952881,
|
|
"epoch": 0.012401264705553756,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5534,
|
|
"mean_token_accuracy": 0.14116353392601014,
|
|
"num_tokens": 467692.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.142725467681885,
|
|
"epoch": 0.012512987811009195,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.706,
|
|
"mean_token_accuracy": 0.13806632682681083,
|
|
"num_tokens": 471719.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.179370450973511,
|
|
"epoch": 0.012624710916464634,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6323,
|
|
"mean_token_accuracy": 0.12942282631993293,
|
|
"num_tokens": 475978.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.241915130615235,
|
|
"epoch": 0.012736434021920073,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7049,
|
|
"mean_token_accuracy": 0.13150753006339072,
|
|
"num_tokens": 480538.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.216012620925904,
|
|
"epoch": 0.012848157127375512,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6041,
|
|
"mean_token_accuracy": 0.13532499819993973,
|
|
"num_tokens": 484604.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.280426120758056,
|
|
"epoch": 0.012959880232830952,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.6293,
|
|
"mean_token_accuracy": 0.13353263586759567,
|
|
"num_tokens": 489117.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.161381196975708,
|
|
"epoch": 0.013071603338286391,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5337,
|
|
"mean_token_accuracy": 0.14060464650392532,
|
|
"num_tokens": 493020.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.130830383300781,
|
|
"epoch": 0.01318332644374183,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.688,
|
|
"mean_token_accuracy": 0.1310403361916542,
|
|
"num_tokens": 497224.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.2293980598449705,
|
|
"epoch": 0.013295049549197269,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6199,
|
|
"mean_token_accuracy": 0.13552179485559462,
|
|
"num_tokens": 501323.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.092097234725952,
|
|
"epoch": 0.013406772654652708,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6871,
|
|
"mean_token_accuracy": 0.1318632557988167,
|
|
"num_tokens": 505544.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.2185986042022705,
|
|
"epoch": 0.013518495760108148,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.624,
|
|
"mean_token_accuracy": 0.14037465676665306,
|
|
"num_tokens": 509980.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.0673870086669925,
|
|
"epoch": 0.013630218865563587,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.4906,
|
|
"mean_token_accuracy": 0.1455707758665085,
|
|
"num_tokens": 514264.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.061304330825806,
|
|
"epoch": 0.013741941971019026,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.573,
|
|
"mean_token_accuracy": 0.14321403577923775,
|
|
"num_tokens": 518482.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.22766146659851,
|
|
"epoch": 0.013853665076474465,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6066,
|
|
"mean_token_accuracy": 0.1336400181055069,
|
|
"num_tokens": 522765.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.029418087005615,
|
|
"epoch": 0.013965388181929906,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4457,
|
|
"mean_token_accuracy": 0.14515347853302957,
|
|
"num_tokens": 526794.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.065269136428833,
|
|
"epoch": 0.014077111287385344,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.5766,
|
|
"mean_token_accuracy": 0.14270118325948716,
|
|
"num_tokens": 531070.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.208616876602173,
|
|
"epoch": 0.014188834392840783,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6389,
|
|
"mean_token_accuracy": 0.13457323089241982,
|
|
"num_tokens": 535655.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.065006256103516,
|
|
"epoch": 0.014300557498296222,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.666,
|
|
"mean_token_accuracy": 0.13108374997973443,
|
|
"num_tokens": 540231.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.094880771636963,
|
|
"epoch": 0.014412280603751661,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5078,
|
|
"mean_token_accuracy": 0.14435381516814233,
|
|
"num_tokens": 544180.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.076756286621094,
|
|
"epoch": 0.014524003709207102,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4786,
|
|
"mean_token_accuracy": 0.13342859596014023,
|
|
"num_tokens": 547905.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.96412615776062,
|
|
"epoch": 0.01463572681466254,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5565,
|
|
"mean_token_accuracy": 0.14217662066221237,
|
|
"num_tokens": 552248.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.046826982498169,
|
|
"epoch": 0.01474744992011798,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3918,
|
|
"mean_token_accuracy": 0.14909711182117463,
|
|
"num_tokens": 556430.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.026239967346191,
|
|
"epoch": 0.014859173025573418,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4678,
|
|
"mean_token_accuracy": 0.15065777227282523,
|
|
"num_tokens": 560114.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.857525396347046,
|
|
"epoch": 0.014970896131028859,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4711,
|
|
"mean_token_accuracy": 0.14087381735444068,
|
|
"num_tokens": 564624.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.029787588119507,
|
|
"epoch": 0.015082619236484298,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4073,
|
|
"mean_token_accuracy": 0.14824407249689103,
|
|
"num_tokens": 568918.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.077581787109375,
|
|
"epoch": 0.015194342341939737,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4725,
|
|
"mean_token_accuracy": 0.15365976840257645,
|
|
"num_tokens": 572972.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.980326414108276,
|
|
"epoch": 0.015306065447395175,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4873,
|
|
"mean_token_accuracy": 0.15378629714250563,
|
|
"num_tokens": 577250.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.116094398498535,
|
|
"epoch": 0.015417788552850614,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4245,
|
|
"mean_token_accuracy": 0.1488552987575531,
|
|
"num_tokens": 581418.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.932662296295166,
|
|
"epoch": 0.015529511658306055,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5042,
|
|
"mean_token_accuracy": 0.15049396306276322,
|
|
"num_tokens": 585675.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.966062307357788,
|
|
"epoch": 0.015641234763761492,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.392,
|
|
"mean_token_accuracy": 0.1462944969534874,
|
|
"num_tokens": 590032.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.915396404266358,
|
|
"epoch": 0.015752957869216933,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3949,
|
|
"mean_token_accuracy": 0.1510719522833824,
|
|
"num_tokens": 594185.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.896506881713867,
|
|
"epoch": 0.015864680974672373,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3488,
|
|
"mean_token_accuracy": 0.15710561126470565,
|
|
"num_tokens": 598246.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.905340528488159,
|
|
"epoch": 0.01597640408012781,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4826,
|
|
"mean_token_accuracy": 0.14725056737661363,
|
|
"num_tokens": 602676.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.877582645416259,
|
|
"epoch": 0.01608812718558325,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.3352,
|
|
"mean_token_accuracy": 0.16034641861915588,
|
|
"num_tokens": 606693.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.880430507659912,
|
|
"epoch": 0.01619985029103869,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3591,
|
|
"mean_token_accuracy": 0.14268538355827332,
|
|
"num_tokens": 610780.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.902221918106079,
|
|
"epoch": 0.01631157339649413,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4226,
|
|
"mean_token_accuracy": 0.1481914848089218,
|
|
"num_tokens": 614746.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.847220468521118,
|
|
"epoch": 0.01642329650194957,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.3719,
|
|
"mean_token_accuracy": 0.15163339525461197,
|
|
"num_tokens": 619222.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.968106889724732,
|
|
"epoch": 0.016535019607405006,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5165,
|
|
"mean_token_accuracy": 0.14626363143324853,
|
|
"num_tokens": 623746.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.9071704864501955,
|
|
"epoch": 0.016646742712860447,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.366,
|
|
"mean_token_accuracy": 0.15544869527220725,
|
|
"num_tokens": 628392.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.845142030715943,
|
|
"epoch": 0.016758465818315887,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3775,
|
|
"mean_token_accuracy": 0.14660717099905013,
|
|
"num_tokens": 632439.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.813414907455444,
|
|
"epoch": 0.016870188923771325,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4088,
|
|
"mean_token_accuracy": 0.1431469090282917,
|
|
"num_tokens": 636801.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.88551287651062,
|
|
"epoch": 0.016981912029226765,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.2173,
|
|
"mean_token_accuracy": 0.16097238063812255,
|
|
"num_tokens": 640818.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.7791321754455565,
|
|
"epoch": 0.017093635134682202,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3784,
|
|
"mean_token_accuracy": 0.14307227358222008,
|
|
"num_tokens": 644914.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.967279005050659,
|
|
"epoch": 0.017205358240137643,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4658,
|
|
"mean_token_accuracy": 0.14935086220502852,
|
|
"num_tokens": 649368.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.788582992553711,
|
|
"epoch": 0.017317081345593083,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.2984,
|
|
"mean_token_accuracy": 0.15846630781888962,
|
|
"num_tokens": 653537.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.87021164894104,
|
|
"epoch": 0.01742880445104852,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4177,
|
|
"mean_token_accuracy": 0.13871796652674676,
|
|
"num_tokens": 658232.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.85303897857666,
|
|
"epoch": 0.01754052755650396,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3841,
|
|
"mean_token_accuracy": 0.14893919229507446,
|
|
"num_tokens": 662512.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.839845895767212,
|
|
"epoch": 0.0176522506619594,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.372,
|
|
"mean_token_accuracy": 0.15011741518974303,
|
|
"num_tokens": 666759.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.646692132949829,
|
|
"epoch": 0.01776397376741484,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.2896,
|
|
"mean_token_accuracy": 0.1571863681077957,
|
|
"num_tokens": 670870.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.806870698928833,
|
|
"epoch": 0.01787569687287028,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.2513,
|
|
"mean_token_accuracy": 0.158031065762043,
|
|
"num_tokens": 675150.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.714898729324341,
|
|
"epoch": 0.017987419978325717,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.1745,
|
|
"mean_token_accuracy": 0.1610417976975441,
|
|
"num_tokens": 679083.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.70656795501709,
|
|
"epoch": 0.018099143083781157,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.371,
|
|
"mean_token_accuracy": 0.15746122524142264,
|
|
"num_tokens": 683477.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.743958902359009,
|
|
"epoch": 0.018210866189236594,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.1656,
|
|
"mean_token_accuracy": 0.16271859258413315,
|
|
"num_tokens": 687819.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.8285057067871096,
|
|
"epoch": 0.018322589294692035,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3924,
|
|
"mean_token_accuracy": 0.15080199763178825,
|
|
"num_tokens": 692017.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.8905205726623535,
|
|
"epoch": 0.018434312400147475,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3853,
|
|
"mean_token_accuracy": 0.14643741324543952,
|
|
"num_tokens": 696853.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.839770412445068,
|
|
"epoch": 0.018546035505602913,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4221,
|
|
"mean_token_accuracy": 0.1416957326233387,
|
|
"num_tokens": 701131.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.750663423538208,
|
|
"epoch": 0.018657758611058353,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3757,
|
|
"mean_token_accuracy": 0.1462937667965889,
|
|
"num_tokens": 705428.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.672054672241211,
|
|
"epoch": 0.018769481716513794,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.2377,
|
|
"mean_token_accuracy": 0.16088497787714004,
|
|
"num_tokens": 709760.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.700072860717773,
|
|
"epoch": 0.01888120482196923,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2073,
|
|
"mean_token_accuracy": 0.1617395505309105,
|
|
"num_tokens": 713964.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.665010023117065,
|
|
"epoch": 0.01899292792742467,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3575,
|
|
"mean_token_accuracy": 0.15287077873945237,
|
|
"num_tokens": 718305.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.806714677810669,
|
|
"epoch": 0.01910465103288011,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2661,
|
|
"mean_token_accuracy": 0.15892615169286728,
|
|
"num_tokens": 722705.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.528658866882324,
|
|
"epoch": 0.01921637413833555,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.1765,
|
|
"mean_token_accuracy": 0.16424758732318878,
|
|
"num_tokens": 726529.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.659326124191284,
|
|
"epoch": 0.01932809724379099,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.2522,
|
|
"mean_token_accuracy": 0.15339272618293762,
|
|
"num_tokens": 730535.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.74210753440857,
|
|
"epoch": 0.019439820349246427,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2117,
|
|
"mean_token_accuracy": 0.15879523605108262,
|
|
"num_tokens": 734836.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.65736346244812,
|
|
"epoch": 0.019551543454701868,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2891,
|
|
"mean_token_accuracy": 0.15371360629796982,
|
|
"num_tokens": 739176.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.574539995193481,
|
|
"epoch": 0.019663266560157305,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2674,
|
|
"mean_token_accuracy": 0.15556667894124984,
|
|
"num_tokens": 743497.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.720397853851319,
|
|
"epoch": 0.019774989665612745,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2818,
|
|
"mean_token_accuracy": 0.16183461248874664,
|
|
"num_tokens": 747185.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.739978790283203,
|
|
"epoch": 0.019886712771068186,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4179,
|
|
"mean_token_accuracy": 0.1448575735092163,
|
|
"num_tokens": 751345.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.7309082508087155,
|
|
"epoch": 0.019998435876523623,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2944,
|
|
"mean_token_accuracy": 0.15704837888479234,
|
|
"num_tokens": 755618.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.653637266159057,
|
|
"epoch": 0.020110158981979064,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2788,
|
|
"mean_token_accuracy": 0.15478281527757645,
|
|
"num_tokens": 759583.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.644137716293335,
|
|
"epoch": 0.0202218820874345,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.1758,
|
|
"mean_token_accuracy": 0.1591601625084877,
|
|
"num_tokens": 763861.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.53189525604248,
|
|
"epoch": 0.02033360519288994,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.1717,
|
|
"mean_token_accuracy": 0.16117519289255142,
|
|
"num_tokens": 768046.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.686648035049439,
|
|
"epoch": 0.020445328298345382,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.1927,
|
|
"mean_token_accuracy": 0.15583422034978867,
|
|
"num_tokens": 772570.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.651849317550659,
|
|
"epoch": 0.02055705140380082,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3181,
|
|
"mean_token_accuracy": 0.15458481013774872,
|
|
"num_tokens": 777324.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.669686937332154,
|
|
"epoch": 0.02066877450925626,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3017,
|
|
"mean_token_accuracy": 0.14780564680695535,
|
|
"num_tokens": 781829.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.655124092102051,
|
|
"epoch": 0.0207804976147117,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1719,
|
|
"mean_token_accuracy": 0.15766340345144272,
|
|
"num_tokens": 786089.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.433820819854736,
|
|
"epoch": 0.020892220720167137,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1607,
|
|
"mean_token_accuracy": 0.16852562725543976,
|
|
"num_tokens": 790273.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.675999402999878,
|
|
"epoch": 0.021003943825622578,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1583,
|
|
"mean_token_accuracy": 0.17233449518680571,
|
|
"num_tokens": 794600.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.556878328323364,
|
|
"epoch": 0.021115666931078015,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.149,
|
|
"mean_token_accuracy": 0.16213736683130264,
|
|
"num_tokens": 798756.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.6472193717956545,
|
|
"epoch": 0.021227390036533456,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1599,
|
|
"mean_token_accuracy": 0.16358468383550645,
|
|
"num_tokens": 803122.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.386259412765503,
|
|
"epoch": 0.021339113141988896,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1891,
|
|
"mean_token_accuracy": 0.15860249921679498,
|
|
"num_tokens": 807393.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.601777124404907,
|
|
"epoch": 0.021450836247444333,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.16311845034360886,
|
|
"num_tokens": 811735.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.646067714691162,
|
|
"epoch": 0.021562559352899774,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3186,
|
|
"mean_token_accuracy": 0.15336783230304718,
|
|
"num_tokens": 816327.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.546494245529175,
|
|
"epoch": 0.02167428245835521,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1048,
|
|
"mean_token_accuracy": 0.1641931340098381,
|
|
"num_tokens": 820476.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.502486848831177,
|
|
"epoch": 0.02178600556381065,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.16595971435308457,
|
|
"num_tokens": 824563.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.51844539642334,
|
|
"epoch": 0.021897728669266092,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2249,
|
|
"mean_token_accuracy": 0.15903169065713882,
|
|
"num_tokens": 829307.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.48329792022705,
|
|
"epoch": 0.02200945177472153,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2234,
|
|
"mean_token_accuracy": 0.16081805527210236,
|
|
"num_tokens": 833876.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.577699089050293,
|
|
"epoch": 0.02212117488017697,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3108,
|
|
"mean_token_accuracy": 0.15140475034713746,
|
|
"num_tokens": 838484.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.628869199752808,
|
|
"epoch": 0.022232897985632407,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1561,
|
|
"mean_token_accuracy": 0.1592697098851204,
|
|
"num_tokens": 842813.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.393803596496582,
|
|
"epoch": 0.022344621091087848,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0899,
|
|
"mean_token_accuracy": 0.16069612354040147,
|
|
"num_tokens": 846638.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.4989090919494625,
|
|
"epoch": 0.022456344196543288,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.0626,
|
|
"mean_token_accuracy": 0.17749472558498383,
|
|
"num_tokens": 850577.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.439338541030883,
|
|
"epoch": 0.022568067301998725,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.0906,
|
|
"mean_token_accuracy": 0.16202376335859298,
|
|
"num_tokens": 854904.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.5132981300354,
|
|
"epoch": 0.022679790407454166,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.1076,
|
|
"mean_token_accuracy": 0.17384408712387084,
|
|
"num_tokens": 858859.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.43322901725769,
|
|
"epoch": 0.022791513512909607,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.055,
|
|
"mean_token_accuracy": 0.16654047518968582,
|
|
"num_tokens": 863121.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.506139421463013,
|
|
"epoch": 0.022903236618365044,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.1709,
|
|
"mean_token_accuracy": 0.15935743749141693,
|
|
"num_tokens": 867805.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.488983964920044,
|
|
"epoch": 0.023014959723820484,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.1244,
|
|
"mean_token_accuracy": 0.1593026451766491,
|
|
"num_tokens": 871988.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.497362184524536,
|
|
"epoch": 0.02312668282927592,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.0959,
|
|
"mean_token_accuracy": 0.16356074959039688,
|
|
"num_tokens": 876181.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.301354026794433,
|
|
"epoch": 0.023238405934731362,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.18099125623703002,
|
|
"num_tokens": 880521.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.322625303268433,
|
|
"epoch": 0.023350129040186803,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.1019,
|
|
"mean_token_accuracy": 0.16592444032430648,
|
|
"num_tokens": 885006.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.541764831542968,
|
|
"epoch": 0.02346185214564224,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.1038,
|
|
"mean_token_accuracy": 0.15975722372531892,
|
|
"num_tokens": 889184.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.410001420974732,
|
|
"epoch": 0.02357357525109768,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.0039,
|
|
"mean_token_accuracy": 0.17406690418720244,
|
|
"num_tokens": 893806.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.391436147689819,
|
|
"epoch": 0.023685298356553117,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.0131,
|
|
"mean_token_accuracy": 0.17006947249174117,
|
|
"num_tokens": 897578.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.47973518371582,
|
|
"epoch": 0.023797021462008558,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.2734,
|
|
"mean_token_accuracy": 0.15798939615488053,
|
|
"num_tokens": 902225.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.551846933364868,
|
|
"epoch": 0.023908744567464,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.1246,
|
|
"mean_token_accuracy": 0.1635009005665779,
|
|
"num_tokens": 906477.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.386241388320923,
|
|
"epoch": 0.024020467672919436,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.1123,
|
|
"mean_token_accuracy": 0.15879142582416533,
|
|
"num_tokens": 910948.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.503281879425049,
|
|
"epoch": 0.024132190778374876,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.2186,
|
|
"mean_token_accuracy": 0.14976079761981964,
|
|
"num_tokens": 915157.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.450709199905395,
|
|
"epoch": 0.024243913883830313,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.0396,
|
|
"mean_token_accuracy": 0.16553105041384697,
|
|
"num_tokens": 920251.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.222010898590088,
|
|
"epoch": 0.024355636989285754,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.1224,
|
|
"mean_token_accuracy": 0.16098514199256897,
|
|
"num_tokens": 924510.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.347459030151367,
|
|
"epoch": 0.024467360094741195,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.0245,
|
|
"mean_token_accuracy": 0.1654239609837532,
|
|
"num_tokens": 928925.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.316383314132691,
|
|
"epoch": 0.02457908320019663,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.0384,
|
|
"mean_token_accuracy": 0.17247851192951202,
|
|
"num_tokens": 933211.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.317654371261597,
|
|
"epoch": 0.024690806305652072,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.1915,
|
|
"mean_token_accuracy": 0.15772637948393822,
|
|
"num_tokens": 937621.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.42638087272644,
|
|
"epoch": 0.024802529411107513,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.0898,
|
|
"mean_token_accuracy": 0.1647167555987835,
|
|
"num_tokens": 941629.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.330166149139404,
|
|
"epoch": 0.02491425251656295,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.0709,
|
|
"mean_token_accuracy": 0.16174031496047975,
|
|
"num_tokens": 945853.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.225450944900513,
|
|
"epoch": 0.02502597562201839,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 5.9055,
|
|
"mean_token_accuracy": 0.17898752093315123,
|
|
"num_tokens": 949758.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.182316493988037,
|
|
"epoch": 0.025137698727473828,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.0268,
|
|
"mean_token_accuracy": 0.17987219095230103,
|
|
"num_tokens": 954028.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.319781017303467,
|
|
"epoch": 0.02524942183292927,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.16309117376804352,
|
|
"num_tokens": 958006.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.320125341415405,
|
|
"epoch": 0.02536114493838471,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 5.9716,
|
|
"mean_token_accuracy": 0.17731243520975112,
|
|
"num_tokens": 962637.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.350974798202515,
|
|
"epoch": 0.025472868043840146,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.17742093056440353,
|
|
"num_tokens": 966770.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.1780962467193605,
|
|
"epoch": 0.025584591149295587,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 5.9758,
|
|
"mean_token_accuracy": 0.1761695474386215,
|
|
"num_tokens": 970976.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.276739263534546,
|
|
"epoch": 0.025696314254751024,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.0095,
|
|
"mean_token_accuracy": 0.16108598709106445,
|
|
"num_tokens": 975253.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.226910066604614,
|
|
"epoch": 0.025808037360206464,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 5.998,
|
|
"mean_token_accuracy": 0.1661012887954712,
|
|
"num_tokens": 979388.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.162249755859375,
|
|
"epoch": 0.025919760465661905,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 5.9613,
|
|
"mean_token_accuracy": 0.1830698937177658,
|
|
"num_tokens": 983216.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.258333396911621,
|
|
"epoch": 0.026031483571117342,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.061,
|
|
"mean_token_accuracy": 0.16143736988306046,
|
|
"num_tokens": 987850.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.322597026824951,
|
|
"epoch": 0.026143206676572783,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 5.9894,
|
|
"mean_token_accuracy": 0.17105733901262282,
|
|
"num_tokens": 992232.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.21760311126709,
|
|
"epoch": 0.02625492978202822,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.0848,
|
|
"mean_token_accuracy": 0.1664736032485962,
|
|
"num_tokens": 996428.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.334726476669312,
|
|
"epoch": 0.02636665288748366,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 5.9449,
|
|
"mean_token_accuracy": 0.17343248575925826,
|
|
"num_tokens": 1000336.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.3395665168762205,
|
|
"epoch": 0.0264783759929391,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 5.9253,
|
|
"mean_token_accuracy": 0.17060509622097014,
|
|
"num_tokens": 1004312.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.365295124053955,
|
|
"epoch": 0.026590099098394538,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.1563,
|
|
"mean_token_accuracy": 0.16267625764012336,
|
|
"num_tokens": 1008728.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.270880270004272,
|
|
"epoch": 0.02670182220384998,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.0663,
|
|
"mean_token_accuracy": 0.16970421522855758,
|
|
"num_tokens": 1013385.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.310678911209107,
|
|
"epoch": 0.026813545309305416,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.0449,
|
|
"mean_token_accuracy": 0.16335367113351823,
|
|
"num_tokens": 1017537.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.093309450149536,
|
|
"epoch": 0.026925268414760856,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 5.9362,
|
|
"mean_token_accuracy": 0.17884828001260758,
|
|
"num_tokens": 1021753.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.228685140609741,
|
|
"epoch": 0.027036991520216297,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.0468,
|
|
"mean_token_accuracy": 0.16449340134859086,
|
|
"num_tokens": 1026248.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.192647790908813,
|
|
"epoch": 0.027148714625671734,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 5.838,
|
|
"mean_token_accuracy": 0.18312328010797502,
|
|
"num_tokens": 1030284.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.250751209259033,
|
|
"epoch": 0.027260437731127175,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.1342,
|
|
"mean_token_accuracy": 0.16936966106295587,
|
|
"num_tokens": 1034444.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.304910802841187,
|
|
"epoch": 0.027372160836582615,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 5.9104,
|
|
"mean_token_accuracy": 0.18955854922533036,
|
|
"num_tokens": 1038819.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.332531595230103,
|
|
"epoch": 0.027483883942038052,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.1242,
|
|
"mean_token_accuracy": 0.16566465348005294,
|
|
"num_tokens": 1043303.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.162945175170899,
|
|
"epoch": 0.027595607047493493,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 5.9639,
|
|
"mean_token_accuracy": 0.1748690575361252,
|
|
"num_tokens": 1047451.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.292677593231201,
|
|
"epoch": 0.02770733015294893,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.0031,
|
|
"mean_token_accuracy": 0.17595543414354325,
|
|
"num_tokens": 1052370.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.406116533279419,
|
|
"epoch": 0.02781905325840437,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.1119,
|
|
"mean_token_accuracy": 0.1625329688191414,
|
|
"num_tokens": 1056755.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.257381248474121,
|
|
"epoch": 0.02793077636385981,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 5.9983,
|
|
"mean_token_accuracy": 0.16609450280666352,
|
|
"num_tokens": 1061380.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.04865870475769,
|
|
"epoch": 0.02804249946931525,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 5.8919,
|
|
"mean_token_accuracy": 0.18033342361450194,
|
|
"num_tokens": 1065633.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.278530168533325,
|
|
"epoch": 0.02815422257477069,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.0283,
|
|
"mean_token_accuracy": 0.16999331265687942,
|
|
"num_tokens": 1070131.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.240628862380982,
|
|
"epoch": 0.028265945680226126,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 5.9501,
|
|
"mean_token_accuracy": 0.16984693259000777,
|
|
"num_tokens": 1074465.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.062196207046509,
|
|
"epoch": 0.028377668785681567,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 5.828,
|
|
"mean_token_accuracy": 0.17757159918546678,
|
|
"num_tokens": 1078738.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.12146635055542,
|
|
"epoch": 0.028489391891137007,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 5.9453,
|
|
"mean_token_accuracy": 0.17620262503623962,
|
|
"num_tokens": 1082529.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.328038311004638,
|
|
"epoch": 0.028601114996592444,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.0216,
|
|
"mean_token_accuracy": 0.16488975435495376,
|
|
"num_tokens": 1087043.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.050336456298828,
|
|
"epoch": 0.028712838102047885,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 5.9187,
|
|
"mean_token_accuracy": 0.18301421254873276,
|
|
"num_tokens": 1091045.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.1567244052886965,
|
|
"epoch": 0.028824561207503322,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 5.8778,
|
|
"mean_token_accuracy": 0.18505997806787491,
|
|
"num_tokens": 1094949.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.193764543533325,
|
|
"epoch": 0.028936284312958763,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 5.942,
|
|
"mean_token_accuracy": 0.17170550525188447,
|
|
"num_tokens": 1099131.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.024008512496948,
|
|
"epoch": 0.029048007418414203,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 5.8863,
|
|
"mean_token_accuracy": 0.18198106437921524,
|
|
"num_tokens": 1102967.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.191930198669434,
|
|
"epoch": 0.02915973052386964,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 5.8698,
|
|
"mean_token_accuracy": 0.18373733162879943,
|
|
"num_tokens": 1106777.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.126851320266724,
|
|
"epoch": 0.02927145362932508,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 5.9212,
|
|
"mean_token_accuracy": 0.17913941144943238,
|
|
"num_tokens": 1111184.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.192364168167114,
|
|
"epoch": 0.02938317673478052,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 5.9298,
|
|
"mean_token_accuracy": 0.17791166305541992,
|
|
"num_tokens": 1115651.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.26938943862915,
|
|
"epoch": 0.02949489984023596,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 5.955,
|
|
"mean_token_accuracy": 0.1659138545393944,
|
|
"num_tokens": 1120012.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.076159715652466,
|
|
"epoch": 0.0296066229456914,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.8588,
|
|
"mean_token_accuracy": 0.17886159271001817,
|
|
"num_tokens": 1124549.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.182869958877563,
|
|
"epoch": 0.029718346051146836,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 5.9345,
|
|
"mean_token_accuracy": 0.17838005125522613,
|
|
"num_tokens": 1128968.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.133463668823242,
|
|
"epoch": 0.029830069156602277,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 5.9933,
|
|
"mean_token_accuracy": 0.17328195571899413,
|
|
"num_tokens": 1133396.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.253204965591431,
|
|
"epoch": 0.029941792262057718,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 5.9025,
|
|
"mean_token_accuracy": 0.170744127035141,
|
|
"num_tokens": 1137577.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.061755275726318,
|
|
"epoch": 0.030053515367513155,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 5.7801,
|
|
"mean_token_accuracy": 0.18417277485132216,
|
|
"num_tokens": 1141551.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.116015100479126,
|
|
"epoch": 0.030165238472968595,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 5.8398,
|
|
"mean_token_accuracy": 0.17423712015151976,
|
|
"num_tokens": 1145295.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.067533349990844,
|
|
"epoch": 0.030276961578424032,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.0567,
|
|
"mean_token_accuracy": 0.16484877467155457,
|
|
"num_tokens": 1149946.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.1512785911560055,
|
|
"epoch": 0.030388684683879473,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 5.9399,
|
|
"mean_token_accuracy": 0.18146816939115523,
|
|
"num_tokens": 1153955.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.143903684616089,
|
|
"epoch": 0.030500407789334914,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 5.849,
|
|
"mean_token_accuracy": 0.18261384814977646,
|
|
"num_tokens": 1158568.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.127876806259155,
|
|
"epoch": 0.03061213089479035,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 5.7759,
|
|
"mean_token_accuracy": 0.18599571883678437,
|
|
"num_tokens": 1162217.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.076024913787842,
|
|
"epoch": 0.03072385400024579,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 5.8861,
|
|
"mean_token_accuracy": 0.17389037311077118,
|
|
"num_tokens": 1165905.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.11526198387146,
|
|
"epoch": 0.03083557710570123,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 5.8432,
|
|
"mean_token_accuracy": 0.17682357281446456,
|
|
"num_tokens": 1170352.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.046456527709961,
|
|
"epoch": 0.03094730021115667,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.9274,
|
|
"mean_token_accuracy": 0.17616797983646393,
|
|
"num_tokens": 1174560.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.11228313446045,
|
|
"epoch": 0.03105902331661211,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 5.8708,
|
|
"mean_token_accuracy": 0.17557800114154815,
|
|
"num_tokens": 1178544.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.975264167785644,
|
|
"epoch": 0.031170746422067547,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 5.6949,
|
|
"mean_token_accuracy": 0.19889509081840515,
|
|
"num_tokens": 1182541.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.97807993888855,
|
|
"epoch": 0.031282469527522984,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 5.7847,
|
|
"mean_token_accuracy": 0.1825215622782707,
|
|
"num_tokens": 1186684.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.036240196228027,
|
|
"epoch": 0.03139419263297843,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 5.7286,
|
|
"mean_token_accuracy": 0.1909311071038246,
|
|
"num_tokens": 1190785.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.977891159057617,
|
|
"epoch": 0.031505915738433865,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 5.7739,
|
|
"mean_token_accuracy": 0.17418278902769088,
|
|
"num_tokens": 1194940.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.920260810852051,
|
|
"epoch": 0.0316176388438893,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 5.7381,
|
|
"mean_token_accuracy": 0.18921381384134292,
|
|
"num_tokens": 1198765.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.071567487716675,
|
|
"epoch": 0.031729361949344746,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 5.9463,
|
|
"mean_token_accuracy": 0.17053111791610717,
|
|
"num_tokens": 1203266.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.066260862350464,
|
|
"epoch": 0.03184108505480018,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 5.7346,
|
|
"mean_token_accuracy": 0.19455005526542662,
|
|
"num_tokens": 1207414.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.978516721725464,
|
|
"epoch": 0.03195280816025562,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 5.7596,
|
|
"mean_token_accuracy": 0.1821496695280075,
|
|
"num_tokens": 1211325.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.088147068023682,
|
|
"epoch": 0.032064531265711065,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 5.7876,
|
|
"mean_token_accuracy": 0.17757320106029512,
|
|
"num_tokens": 1215866.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.105999946594238,
|
|
"epoch": 0.0321762543711665,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 5.8637,
|
|
"mean_token_accuracy": 0.16661322563886644,
|
|
"num_tokens": 1220241.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.943767118453979,
|
|
"epoch": 0.03228797747662194,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.8393,
|
|
"mean_token_accuracy": 0.17783186137676238,
|
|
"num_tokens": 1224330.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.21411657333374,
|
|
"epoch": 0.03239970058207738,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 5.9352,
|
|
"mean_token_accuracy": 0.16817386895418168,
|
|
"num_tokens": 1228518.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.985192489624024,
|
|
"epoch": 0.03251142368753282,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 5.8087,
|
|
"mean_token_accuracy": 0.1855456992983818,
|
|
"num_tokens": 1232556.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.902926826477051,
|
|
"epoch": 0.03262314679298826,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 5.8417,
|
|
"mean_token_accuracy": 0.17538347840309143,
|
|
"num_tokens": 1236464.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.1002099990844725,
|
|
"epoch": 0.032734869898443694,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 5.8552,
|
|
"mean_token_accuracy": 0.1877426788210869,
|
|
"num_tokens": 1240478.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.054091215133667,
|
|
"epoch": 0.03284659300389914,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 5.8806,
|
|
"mean_token_accuracy": 0.1829858422279358,
|
|
"num_tokens": 1244149.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.208575963973999,
|
|
"epoch": 0.032958316109354575,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 5.9965,
|
|
"mean_token_accuracy": 0.15834348127245904,
|
|
"num_tokens": 1248617.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.053374433517456,
|
|
"epoch": 0.03307003921481001,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 5.8381,
|
|
"mean_token_accuracy": 0.18422103077173232,
|
|
"num_tokens": 1253044.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.099059247970581,
|
|
"epoch": 0.03318176232026546,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 5.8787,
|
|
"mean_token_accuracy": 0.1814146339893341,
|
|
"num_tokens": 1257651.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.0209493160247805,
|
|
"epoch": 0.033293485425720894,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 5.8113,
|
|
"mean_token_accuracy": 0.19222917556762695,
|
|
"num_tokens": 1261805.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.859741830825806,
|
|
"epoch": 0.03340520853117633,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 5.6211,
|
|
"mean_token_accuracy": 0.20293725579977034,
|
|
"num_tokens": 1265625.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.956333827972412,
|
|
"epoch": 0.033516931636631775,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 5.7941,
|
|
"mean_token_accuracy": 0.18911935687065123,
|
|
"num_tokens": 1269876.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 272346935132160.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|