fe33fe6f22
Model: fpadovani/swe-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.3140016570008286,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.70417890548706,
|
|
"epoch": 0.004142502071251036,
|
|
"grad_norm": 18.0,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.324,
|
|
"mean_token_accuracy": 9.699321235530078e-05,
|
|
"num_tokens": 9149.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.722008132934571,
|
|
"epoch": 0.008285004142502071,
|
|
"grad_norm": 19.5,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.1689,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 16800.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.784619331359863,
|
|
"epoch": 0.012427506213753107,
|
|
"grad_norm": 23.875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.9756,
|
|
"mean_token_accuracy": 0.00012360939290374519,
|
|
"num_tokens": 26617.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.349014616012573,
|
|
"epoch": 0.016570008285004142,
|
|
"grad_norm": 30.125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.1127,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 35855.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 8.763514041900635,
|
|
"epoch": 0.020712510356255178,
|
|
"grad_norm": 6.3125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.2931,
|
|
"mean_token_accuracy": 0.0002482298179529607,
|
|
"num_tokens": 45644.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.648772144317627,
|
|
"epoch": 0.024855012427506214,
|
|
"grad_norm": 3.203125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.6872,
|
|
"mean_token_accuracy": 0.014276380836963653,
|
|
"num_tokens": 54641.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.69749927520752,
|
|
"epoch": 0.02899751449875725,
|
|
"grad_norm": 3.109375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.4448,
|
|
"mean_token_accuracy": 0.018782001547515394,
|
|
"num_tokens": 63924.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.672551250457763,
|
|
"epoch": 0.033140016570008285,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1617,
|
|
"mean_token_accuracy": 0.02377438023686409,
|
|
"num_tokens": 73452.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.593613147735596,
|
|
"epoch": 0.037282518641259324,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8966,
|
|
"mean_token_accuracy": 0.039151473343372344,
|
|
"num_tokens": 82731.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.570558643341064,
|
|
"epoch": 0.041425020712510356,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7261,
|
|
"mean_token_accuracy": 0.04260772801935673,
|
|
"num_tokens": 92626.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.598753547668457,
|
|
"epoch": 0.045567522783761395,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.669,
|
|
"mean_token_accuracy": 0.04115805197507143,
|
|
"num_tokens": 101380.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.600215435028076,
|
|
"epoch": 0.04971002485501243,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5621,
|
|
"mean_token_accuracy": 0.041873040795326236,
|
|
"num_tokens": 109895.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.531742191314697,
|
|
"epoch": 0.053852526926263466,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.5289,
|
|
"mean_token_accuracy": 0.04000609517097473,
|
|
"num_tokens": 119203.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.568301486968995,
|
|
"epoch": 0.0579950289975145,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.3923,
|
|
"mean_token_accuracy": 0.03999764695763588,
|
|
"num_tokens": 127572.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.564670372009278,
|
|
"epoch": 0.06213753106876554,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.3471,
|
|
"mean_token_accuracy": 0.042604190111160276,
|
|
"num_tokens": 137161.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.54567642211914,
|
|
"epoch": 0.06628003314001657,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.2877,
|
|
"mean_token_accuracy": 0.04093840681016445,
|
|
"num_tokens": 145991.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.513956260681152,
|
|
"epoch": 0.07042253521126761,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1852,
|
|
"mean_token_accuracy": 0.041304165683686736,
|
|
"num_tokens": 154927.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.496724605560303,
|
|
"epoch": 0.07456503728251865,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.0189,
|
|
"mean_token_accuracy": 0.046985261142253876,
|
|
"num_tokens": 163234.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.47362985610962,
|
|
"epoch": 0.07870753935376967,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.9566,
|
|
"mean_token_accuracy": 0.05403314344584942,
|
|
"num_tokens": 171554.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.420413589477539,
|
|
"epoch": 0.08285004142502071,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8624,
|
|
"mean_token_accuracy": 0.0495030127465725,
|
|
"num_tokens": 180171.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.291831016540527,
|
|
"epoch": 0.08699254349627175,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.7136,
|
|
"mean_token_accuracy": 0.05688973888754845,
|
|
"num_tokens": 189216.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.206434440612792,
|
|
"epoch": 0.09113504556752279,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.735,
|
|
"mean_token_accuracy": 0.05266922526061535,
|
|
"num_tokens": 200022.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.191181755065918,
|
|
"epoch": 0.09527754763877382,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.5465,
|
|
"mean_token_accuracy": 0.05359771586954594,
|
|
"num_tokens": 208686.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.06379051208496,
|
|
"epoch": 0.09942004971002485,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.4093,
|
|
"mean_token_accuracy": 0.054267140850424764,
|
|
"num_tokens": 218266.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.88507251739502,
|
|
"epoch": 0.1035625517812759,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.2985,
|
|
"mean_token_accuracy": 0.04877174161374569,
|
|
"num_tokens": 227658.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.759512615203857,
|
|
"epoch": 0.10770505385252693,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.0999,
|
|
"mean_token_accuracy": 0.06026250720024109,
|
|
"num_tokens": 236301.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.516020011901855,
|
|
"epoch": 0.11184755592377796,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.1214,
|
|
"mean_token_accuracy": 0.04890455044806004,
|
|
"num_tokens": 246355.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.335341739654542,
|
|
"epoch": 0.115990057995029,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.9629,
|
|
"mean_token_accuracy": 0.05138283483684063,
|
|
"num_tokens": 256023.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.04491310119629,
|
|
"epoch": 0.12013256006628004,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.9599,
|
|
"mean_token_accuracy": 0.046007866784930226,
|
|
"num_tokens": 266893.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.849730587005615,
|
|
"epoch": 0.12427506213753108,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.7312,
|
|
"mean_token_accuracy": 0.0518551766872406,
|
|
"num_tokens": 276075.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.576015663146972,
|
|
"epoch": 0.12841756420878211,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.7017,
|
|
"mean_token_accuracy": 0.056488437578082085,
|
|
"num_tokens": 285529.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.416762542724609,
|
|
"epoch": 0.13256006628003314,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.686,
|
|
"mean_token_accuracy": 0.05418187938630581,
|
|
"num_tokens": 294403.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.284829902648926,
|
|
"epoch": 0.13670256835128416,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.712,
|
|
"mean_token_accuracy": 0.05654969438910484,
|
|
"num_tokens": 303507.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.158583068847657,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.5667,
|
|
"mean_token_accuracy": 0.0604190930724144,
|
|
"num_tokens": 312393.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.053034448623658,
|
|
"epoch": 0.14498757249378624,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.544,
|
|
"mean_token_accuracy": 0.06070729456841946,
|
|
"num_tokens": 321309.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.106309843063354,
|
|
"epoch": 0.1491300745650373,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.6075,
|
|
"mean_token_accuracy": 0.05758780501782894,
|
|
"num_tokens": 330217.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.025533390045165,
|
|
"epoch": 0.15327257663628832,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5973,
|
|
"mean_token_accuracy": 0.06240675076842308,
|
|
"num_tokens": 339587.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.989250421524048,
|
|
"epoch": 0.15741507870753935,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.5142,
|
|
"mean_token_accuracy": 0.06207436956465244,
|
|
"num_tokens": 349367.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.934307050704956,
|
|
"epoch": 0.1615575807787904,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4192,
|
|
"mean_token_accuracy": 0.06271038055419922,
|
|
"num_tokens": 358272.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.99296875,
|
|
"epoch": 0.16570008285004142,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.6202,
|
|
"mean_token_accuracy": 0.06270747631788254,
|
|
"num_tokens": 367709.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.9447160243988035,
|
|
"epoch": 0.16984258492129245,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4595,
|
|
"mean_token_accuracy": 0.06487247794866562,
|
|
"num_tokens": 376180.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.915022087097168,
|
|
"epoch": 0.1739850869925435,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.574,
|
|
"mean_token_accuracy": 0.06149500273168087,
|
|
"num_tokens": 385636.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.981282186508179,
|
|
"epoch": 0.17812758906379453,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.4416,
|
|
"mean_token_accuracy": 0.06958327889442444,
|
|
"num_tokens": 394486.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.828207683563233,
|
|
"epoch": 0.18227009113504558,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.4059,
|
|
"mean_token_accuracy": 0.05610249787569046,
|
|
"num_tokens": 403765.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.8190673828125,
|
|
"epoch": 0.1864125932062966,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3815,
|
|
"mean_token_accuracy": 0.06853485442698001,
|
|
"num_tokens": 412988.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.780495929718017,
|
|
"epoch": 0.19055509527754763,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.3492,
|
|
"mean_token_accuracy": 0.06311094798147679,
|
|
"num_tokens": 422096.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.860241508483886,
|
|
"epoch": 0.19469759734879868,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.4279,
|
|
"mean_token_accuracy": 0.0612465362995863,
|
|
"num_tokens": 431333.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.842223691940307,
|
|
"epoch": 0.1988400994200497,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.4979,
|
|
"mean_token_accuracy": 0.06898829713463783,
|
|
"num_tokens": 440382.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.790421056747436,
|
|
"epoch": 0.20298260149130073,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3766,
|
|
"mean_token_accuracy": 0.07021775655448437,
|
|
"num_tokens": 448469.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.793370962142944,
|
|
"epoch": 0.2071251035625518,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.4374,
|
|
"mean_token_accuracy": 0.0661789108067751,
|
|
"num_tokens": 457356.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.799811267852784,
|
|
"epoch": 0.2112676056338028,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.4418,
|
|
"mean_token_accuracy": 0.0664214115589857,
|
|
"num_tokens": 465928.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.788989782333374,
|
|
"epoch": 0.21541010770505387,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3311,
|
|
"mean_token_accuracy": 0.06980733200907707,
|
|
"num_tokens": 475269.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.780997276306152,
|
|
"epoch": 0.2195526097763049,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.295,
|
|
"mean_token_accuracy": 0.06686929501593113,
|
|
"num_tokens": 485450.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.6190821647644045,
|
|
"epoch": 0.22369511184755592,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3345,
|
|
"mean_token_accuracy": 0.07399698086082936,
|
|
"num_tokens": 494029.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.713169765472412,
|
|
"epoch": 0.22783761391880697,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.294,
|
|
"mean_token_accuracy": 0.07526539191603661,
|
|
"num_tokens": 503253.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.728218841552734,
|
|
"epoch": 0.231980115990058,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2987,
|
|
"mean_token_accuracy": 0.07004385031759738,
|
|
"num_tokens": 512857.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.657222604751587,
|
|
"epoch": 0.23612261806130902,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.3416,
|
|
"mean_token_accuracy": 0.07092855796217919,
|
|
"num_tokens": 522319.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.699072313308716,
|
|
"epoch": 0.24026512013256007,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.282,
|
|
"mean_token_accuracy": 0.07054338790476322,
|
|
"num_tokens": 531469.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.673645305633545,
|
|
"epoch": 0.2444076222038111,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.3049,
|
|
"mean_token_accuracy": 0.07169978618621826,
|
|
"num_tokens": 540470.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.67633023262024,
|
|
"epoch": 0.24855012427506215,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2885,
|
|
"mean_token_accuracy": 0.07255204916000366,
|
|
"num_tokens": 550317.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.703630876541138,
|
|
"epoch": 0.2526926263463132,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2609,
|
|
"mean_token_accuracy": 0.07455097585916519,
|
|
"num_tokens": 559409.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.531597137451172,
|
|
"epoch": 0.25683512841756423,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2618,
|
|
"mean_token_accuracy": 0.0835557732731104,
|
|
"num_tokens": 568726.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.69538688659668,
|
|
"epoch": 0.2609776304888152,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.272,
|
|
"mean_token_accuracy": 0.070323546230793,
|
|
"num_tokens": 577926.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.653127574920655,
|
|
"epoch": 0.2651201325600663,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2668,
|
|
"mean_token_accuracy": 0.0749148778617382,
|
|
"num_tokens": 586695.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.685172080993652,
|
|
"epoch": 0.26926263463131733,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2778,
|
|
"mean_token_accuracy": 0.072557582706213,
|
|
"num_tokens": 596250.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.5769373893737795,
|
|
"epoch": 0.27340513670256833,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.297,
|
|
"mean_token_accuracy": 0.07127823494374752,
|
|
"num_tokens": 605761.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.630343723297119,
|
|
"epoch": 0.2775476387738194,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.2279,
|
|
"mean_token_accuracy": 0.07685448080301285,
|
|
"num_tokens": 614313.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.6338982582092285,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2426,
|
|
"mean_token_accuracy": 0.07610973864793777,
|
|
"num_tokens": 624470.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.5630439758300785,
|
|
"epoch": 0.28583264291632143,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1876,
|
|
"mean_token_accuracy": 0.07631983868777752,
|
|
"num_tokens": 633458.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.555823183059692,
|
|
"epoch": 0.2899751449875725,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1541,
|
|
"mean_token_accuracy": 0.07695739194750786,
|
|
"num_tokens": 643251.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.54128098487854,
|
|
"epoch": 0.29411764705882354,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.2095,
|
|
"mean_token_accuracy": 0.07986946441233159,
|
|
"num_tokens": 652005.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.541937828063965,
|
|
"epoch": 0.2982601491300746,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.1099,
|
|
"mean_token_accuracy": 0.08143376260995865,
|
|
"num_tokens": 661003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.468397426605224,
|
|
"epoch": 0.3024026512013256,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0991,
|
|
"mean_token_accuracy": 0.07771601304411888,
|
|
"num_tokens": 670221.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.572120714187622,
|
|
"epoch": 0.30654515327257664,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.3705,
|
|
"mean_token_accuracy": 0.06953227184712887,
|
|
"num_tokens": 680756.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.47379961013794,
|
|
"epoch": 0.3106876553438277,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0678,
|
|
"mean_token_accuracy": 0.08287248685956002,
|
|
"num_tokens": 689641.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.441448068618774,
|
|
"epoch": 0.3148301574150787,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0139,
|
|
"mean_token_accuracy": 0.0792611964046955,
|
|
"num_tokens": 698209.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.345022249221802,
|
|
"epoch": 0.31897265948632975,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.0205,
|
|
"mean_token_accuracy": 0.08480528593063355,
|
|
"num_tokens": 706712.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.502147245407104,
|
|
"epoch": 0.3231151615575808,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.2079,
|
|
"mean_token_accuracy": 0.07785556949675083,
|
|
"num_tokens": 716087.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.44799690246582,
|
|
"epoch": 0.3272576636288318,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0887,
|
|
"mean_token_accuracy": 0.07472220808267593,
|
|
"num_tokens": 725913.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.450920248031617,
|
|
"epoch": 0.33140016570008285,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1256,
|
|
"mean_token_accuracy": 0.08386809080839157,
|
|
"num_tokens": 735020.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.34827938079834,
|
|
"epoch": 0.3355426677713339,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0323,
|
|
"mean_token_accuracy": 0.07652226313948632,
|
|
"num_tokens": 744299.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.45903434753418,
|
|
"epoch": 0.3396851698425849,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.0115,
|
|
"mean_token_accuracy": 0.08094536513090134,
|
|
"num_tokens": 753231.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.382762813568116,
|
|
"epoch": 0.34382767191383595,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.069,
|
|
"mean_token_accuracy": 0.0813817098736763,
|
|
"num_tokens": 763043.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.398455286026001,
|
|
"epoch": 0.347970173985087,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.1014,
|
|
"mean_token_accuracy": 0.08094010204076767,
|
|
"num_tokens": 771977.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.3298886775970455,
|
|
"epoch": 0.352112676056338,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.9887,
|
|
"mean_token_accuracy": 0.0867611140012741,
|
|
"num_tokens": 780285.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.321137142181397,
|
|
"epoch": 0.35625517812758906,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0564,
|
|
"mean_token_accuracy": 0.08066153451800347,
|
|
"num_tokens": 789980.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.415691232681274,
|
|
"epoch": 0.3603976801988401,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.0081,
|
|
"mean_token_accuracy": 0.0885832205414772,
|
|
"num_tokens": 798700.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.261818742752075,
|
|
"epoch": 0.36454018227009116,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.0726,
|
|
"mean_token_accuracy": 0.07862451300024986,
|
|
"num_tokens": 808758.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.4208532810211185,
|
|
"epoch": 0.36868268434134216,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0513,
|
|
"mean_token_accuracy": 0.08192357867956161,
|
|
"num_tokens": 817692.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.318561601638794,
|
|
"epoch": 0.3728251864125932,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.0764,
|
|
"mean_token_accuracy": 0.09159427806735039,
|
|
"num_tokens": 827052.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.326152086257935,
|
|
"epoch": 0.37696768848384427,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.0672,
|
|
"mean_token_accuracy": 0.08785221576690674,
|
|
"num_tokens": 836442.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.290367841720581,
|
|
"epoch": 0.38111019055509526,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9496,
|
|
"mean_token_accuracy": 0.08379893153905868,
|
|
"num_tokens": 846441.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.274770307540893,
|
|
"epoch": 0.3852526926263463,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0006,
|
|
"mean_token_accuracy": 0.08496225997805595,
|
|
"num_tokens": 856108.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.303304958343506,
|
|
"epoch": 0.38939519469759737,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.9345,
|
|
"mean_token_accuracy": 0.0826059028506279,
|
|
"num_tokens": 865788.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.17899341583252,
|
|
"epoch": 0.39353769676884837,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9851,
|
|
"mean_token_accuracy": 0.07972754761576653,
|
|
"num_tokens": 874833.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.176127529144287,
|
|
"epoch": 0.3976801988400994,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9418,
|
|
"mean_token_accuracy": 0.08143137246370316,
|
|
"num_tokens": 883685.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.265657234191894,
|
|
"epoch": 0.40182270091135047,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.9269,
|
|
"mean_token_accuracy": 0.08215037062764168,
|
|
"num_tokens": 893908.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.271769285202026,
|
|
"epoch": 0.40596520298260147,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.0375,
|
|
"mean_token_accuracy": 0.08106703236699105,
|
|
"num_tokens": 903081.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.133934879302979,
|
|
"epoch": 0.4101077050538525,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8953,
|
|
"mean_token_accuracy": 0.08535027876496315,
|
|
"num_tokens": 912515.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.248036670684814,
|
|
"epoch": 0.4142502071251036,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.9103,
|
|
"mean_token_accuracy": 0.08826312869787216,
|
|
"num_tokens": 921056.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.4142502071251036,
|
|
"eval_entropy": 7.152716042898866,
|
|
"eval_loss": 7.043540954589844,
|
|
"eval_mean_token_accuracy": 0.08588246503141897,
|
|
"eval_num_tokens": 921056.0,
|
|
"eval_runtime": 2.8013,
|
|
"eval_samples_per_second": 1337.598,
|
|
"eval_steps_per_second": 167.423,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.137938928604126,
|
|
"epoch": 0.4183927091963546,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8537,
|
|
"mean_token_accuracy": 0.0894715815782547,
|
|
"num_tokens": 931112.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.109851598739624,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.9643,
|
|
"mean_token_accuracy": 0.08381934091448784,
|
|
"num_tokens": 941120.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.161206293106079,
|
|
"epoch": 0.4266777133388567,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9286,
|
|
"mean_token_accuracy": 0.07928110212087632,
|
|
"num_tokens": 950041.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.187362718582153,
|
|
"epoch": 0.43082021541010773,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.9,
|
|
"mean_token_accuracy": 0.08709124103188515,
|
|
"num_tokens": 958608.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.231058073043823,
|
|
"epoch": 0.43496271748135873,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.0099,
|
|
"mean_token_accuracy": 0.08367124274373054,
|
|
"num_tokens": 966901.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.111437368392944,
|
|
"epoch": 0.4391052195526098,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8679,
|
|
"mean_token_accuracy": 0.0846832774579525,
|
|
"num_tokens": 976198.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.2298952579498295,
|
|
"epoch": 0.44324772162386084,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.9271,
|
|
"mean_token_accuracy": 0.08150140345096588,
|
|
"num_tokens": 985248.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.10087308883667,
|
|
"epoch": 0.44739022369511183,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8356,
|
|
"mean_token_accuracy": 0.08679510727524757,
|
|
"num_tokens": 993289.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.159850454330444,
|
|
"epoch": 0.4515327257663629,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8915,
|
|
"mean_token_accuracy": 0.08651292622089386,
|
|
"num_tokens": 1001784.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.133292531967163,
|
|
"epoch": 0.45567522783761394,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.8731,
|
|
"mean_token_accuracy": 0.09236607924103737,
|
|
"num_tokens": 1010445.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.113561677932739,
|
|
"epoch": 0.45981772990886494,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.9026,
|
|
"mean_token_accuracy": 0.08968292474746704,
|
|
"num_tokens": 1019084.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.082435131072998,
|
|
"epoch": 0.463960231980116,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8235,
|
|
"mean_token_accuracy": 0.08769916668534279,
|
|
"num_tokens": 1027263.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.080044603347778,
|
|
"epoch": 0.46810273405136704,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.9333,
|
|
"mean_token_accuracy": 0.0789696842432022,
|
|
"num_tokens": 1036275.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.151401376724243,
|
|
"epoch": 0.47224523612261804,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.9681,
|
|
"mean_token_accuracy": 0.08598844856023788,
|
|
"num_tokens": 1045208.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.106773471832275,
|
|
"epoch": 0.4763877381938691,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8902,
|
|
"mean_token_accuracy": 0.08505614772439003,
|
|
"num_tokens": 1054703.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.964013481140137,
|
|
"epoch": 0.48053024026512015,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7593,
|
|
"mean_token_accuracy": 0.08789120018482208,
|
|
"num_tokens": 1063652.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.994414854049682,
|
|
"epoch": 0.48467274233637114,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.853,
|
|
"mean_token_accuracy": 0.08468487113714218,
|
|
"num_tokens": 1072070.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.174046993255615,
|
|
"epoch": 0.4888152444076222,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8959,
|
|
"mean_token_accuracy": 0.0872904784977436,
|
|
"num_tokens": 1079992.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.0769284725189205,
|
|
"epoch": 0.49295774647887325,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.8845,
|
|
"mean_token_accuracy": 0.08729023486375809,
|
|
"num_tokens": 1088862.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.002944040298462,
|
|
"epoch": 0.4971002485501243,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7078,
|
|
"mean_token_accuracy": 0.09125389754772187,
|
|
"num_tokens": 1097614.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.979242372512817,
|
|
"epoch": 0.5012427506213754,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.8893,
|
|
"mean_token_accuracy": 0.09072866439819335,
|
|
"num_tokens": 1107347.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.11456413269043,
|
|
"epoch": 0.5053852526926264,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.8977,
|
|
"mean_token_accuracy": 0.0837775819003582,
|
|
"num_tokens": 1116707.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.103943634033203,
|
|
"epoch": 0.5095277547638773,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.889,
|
|
"mean_token_accuracy": 0.08799269050359726,
|
|
"num_tokens": 1126356.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.997752904891968,
|
|
"epoch": 0.5136702568351285,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.8317,
|
|
"mean_token_accuracy": 0.08332241624593735,
|
|
"num_tokens": 1136281.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.969416952133178,
|
|
"epoch": 0.5178127589063795,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7914,
|
|
"mean_token_accuracy": 0.08854400739073753,
|
|
"num_tokens": 1145466.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.913108205795288,
|
|
"epoch": 0.5219552609776305,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7835,
|
|
"mean_token_accuracy": 0.09256977885961533,
|
|
"num_tokens": 1154225.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.981909942626953,
|
|
"epoch": 0.5260977630488816,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.8447,
|
|
"mean_token_accuracy": 0.0854923166334629,
|
|
"num_tokens": 1163872.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.036574649810791,
|
|
"epoch": 0.5302402651201326,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7584,
|
|
"mean_token_accuracy": 0.09340214282274246,
|
|
"num_tokens": 1172695.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.875000143051148,
|
|
"epoch": 0.5343827671913836,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6344,
|
|
"mean_token_accuracy": 0.09556335210800171,
|
|
"num_tokens": 1181291.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.9969175338745115,
|
|
"epoch": 0.5385252692626347,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.8093,
|
|
"mean_token_accuracy": 0.08966382555663585,
|
|
"num_tokens": 1189927.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.955785942077637,
|
|
"epoch": 0.5426677713338857,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.7423,
|
|
"mean_token_accuracy": 0.09562051445245742,
|
|
"num_tokens": 1197913.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.968462419509888,
|
|
"epoch": 0.5468102734051367,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.6775,
|
|
"mean_token_accuracy": 0.09469703435897828,
|
|
"num_tokens": 1207032.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.832558679580688,
|
|
"epoch": 0.5509527754763878,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7769,
|
|
"mean_token_accuracy": 0.09034623727202415,
|
|
"num_tokens": 1215710.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.937397909164429,
|
|
"epoch": 0.5550952775476388,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.7588,
|
|
"mean_token_accuracy": 0.08644548580050468,
|
|
"num_tokens": 1225619.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.87630820274353,
|
|
"epoch": 0.5592377796188898,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6684,
|
|
"mean_token_accuracy": 0.09475696682929993,
|
|
"num_tokens": 1235000.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.888253688812256,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.6846,
|
|
"mean_token_accuracy": 0.09072612524032593,
|
|
"num_tokens": 1243594.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.984683322906494,
|
|
"epoch": 0.5675227837613919,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.8658,
|
|
"mean_token_accuracy": 0.0922737754881382,
|
|
"num_tokens": 1252279.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.964951038360596,
|
|
"epoch": 0.5716652858326429,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7816,
|
|
"mean_token_accuracy": 0.09199510142207146,
|
|
"num_tokens": 1261314.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.807141304016113,
|
|
"epoch": 0.575807787903894,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6534,
|
|
"mean_token_accuracy": 0.09733643308281899,
|
|
"num_tokens": 1270854.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.822793102264404,
|
|
"epoch": 0.579950289975145,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6378,
|
|
"mean_token_accuracy": 0.09607320502400399,
|
|
"num_tokens": 1279183.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.798097324371338,
|
|
"epoch": 0.584092792046396,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.7234,
|
|
"mean_token_accuracy": 0.09325428456068038,
|
|
"num_tokens": 1288347.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.899397277832032,
|
|
"epoch": 0.5882352941176471,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.7217,
|
|
"mean_token_accuracy": 0.09395038411021232,
|
|
"num_tokens": 1297449.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.90986647605896,
|
|
"epoch": 0.5923777961888981,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.6996,
|
|
"mean_token_accuracy": 0.09421416223049164,
|
|
"num_tokens": 1306232.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.846466302871704,
|
|
"epoch": 0.5965202982601492,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6972,
|
|
"mean_token_accuracy": 0.09734119176864624,
|
|
"num_tokens": 1315073.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.890099143981933,
|
|
"epoch": 0.6006628003314002,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.7379,
|
|
"mean_token_accuracy": 0.08920296430587768,
|
|
"num_tokens": 1324629.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.895062398910523,
|
|
"epoch": 0.6048053024026512,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.8505,
|
|
"mean_token_accuracy": 0.08717696964740754,
|
|
"num_tokens": 1334152.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.952538633346558,
|
|
"epoch": 0.6089478044739023,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.827,
|
|
"mean_token_accuracy": 0.08865289464592933,
|
|
"num_tokens": 1343063.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.843909358978271,
|
|
"epoch": 0.6130903065451533,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.7434,
|
|
"mean_token_accuracy": 0.08947005942463875,
|
|
"num_tokens": 1352632.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.815534782409668,
|
|
"epoch": 0.6172328086164043,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.6749,
|
|
"mean_token_accuracy": 0.09639113694429398,
|
|
"num_tokens": 1361103.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.879958820343018,
|
|
"epoch": 0.6213753106876554,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.6594,
|
|
"mean_token_accuracy": 0.09863266348838806,
|
|
"num_tokens": 1370658.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.780733728408814,
|
|
"epoch": 0.6255178127589064,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6808,
|
|
"mean_token_accuracy": 0.09909494146704674,
|
|
"num_tokens": 1379817.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.838653945922852,
|
|
"epoch": 0.6296603148301574,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6471,
|
|
"mean_token_accuracy": 0.09778247401118279,
|
|
"num_tokens": 1389329.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.66096248626709,
|
|
"epoch": 0.6338028169014085,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6749,
|
|
"mean_token_accuracy": 0.09297099262475968,
|
|
"num_tokens": 1399279.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.80253233909607,
|
|
"epoch": 0.6379453189726595,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.603,
|
|
"mean_token_accuracy": 0.09690207317471504,
|
|
"num_tokens": 1408765.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.849671459197998,
|
|
"epoch": 0.6420878210439105,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.7456,
|
|
"mean_token_accuracy": 0.09820655733346939,
|
|
"num_tokens": 1418615.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.82501745223999,
|
|
"epoch": 0.6462303231151616,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6678,
|
|
"mean_token_accuracy": 0.0880895622074604,
|
|
"num_tokens": 1427359.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.867260122299195,
|
|
"epoch": 0.6503728251864126,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.8025,
|
|
"mean_token_accuracy": 0.09724401906132699,
|
|
"num_tokens": 1435516.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.804732322692871,
|
|
"epoch": 0.6545153272576636,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.7582,
|
|
"mean_token_accuracy": 0.09707285091280937,
|
|
"num_tokens": 1444438.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.864415264129638,
|
|
"epoch": 0.6586578293289147,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.6723,
|
|
"mean_token_accuracy": 0.10008604601025581,
|
|
"num_tokens": 1453740.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.7923424243927,
|
|
"epoch": 0.6628003314001657,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.6596,
|
|
"mean_token_accuracy": 0.0992733120918274,
|
|
"num_tokens": 1462083.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.66512622833252,
|
|
"epoch": 0.6669428334714167,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5895,
|
|
"mean_token_accuracy": 0.10381095260381698,
|
|
"num_tokens": 1471459.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.769164514541626,
|
|
"epoch": 0.6710853355426678,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5678,
|
|
"mean_token_accuracy": 0.10211108550429344,
|
|
"num_tokens": 1480516.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.621196842193603,
|
|
"epoch": 0.6752278376139188,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4912,
|
|
"mean_token_accuracy": 0.09601941630244255,
|
|
"num_tokens": 1489268.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.848507595062256,
|
|
"epoch": 0.6793703396851698,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.7021,
|
|
"mean_token_accuracy": 0.09602859765291213,
|
|
"num_tokens": 1498819.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.716065549850464,
|
|
"epoch": 0.6835128417564209,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6693,
|
|
"mean_token_accuracy": 0.10195070579648018,
|
|
"num_tokens": 1507852.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.764216947555542,
|
|
"epoch": 0.6876553438276719,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.6844,
|
|
"mean_token_accuracy": 0.1013753168284893,
|
|
"num_tokens": 1517538.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.660551071166992,
|
|
"epoch": 0.6917978458989229,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.567,
|
|
"mean_token_accuracy": 0.10303043052554131,
|
|
"num_tokens": 1525389.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.7510017395019535,
|
|
"epoch": 0.695940347970174,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5375,
|
|
"mean_token_accuracy": 0.100385482609272,
|
|
"num_tokens": 1534117.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.690871620178223,
|
|
"epoch": 0.700082850041425,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.5814,
|
|
"mean_token_accuracy": 0.10090366378426552,
|
|
"num_tokens": 1542610.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.683345317840576,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5864,
|
|
"mean_token_accuracy": 0.10127236992120743,
|
|
"num_tokens": 1551771.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.748124027252198,
|
|
"epoch": 0.7083678541839271,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.7407,
|
|
"mean_token_accuracy": 0.09586205929517747,
|
|
"num_tokens": 1560231.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.7252332210540775,
|
|
"epoch": 0.7125103562551781,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.6918,
|
|
"mean_token_accuracy": 0.0975484773516655,
|
|
"num_tokens": 1569578.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.756966400146484,
|
|
"epoch": 0.7166528583264291,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.5691,
|
|
"mean_token_accuracy": 0.10365280956029892,
|
|
"num_tokens": 1579308.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.701776933670044,
|
|
"epoch": 0.7207953603976802,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.7528,
|
|
"mean_token_accuracy": 0.09729094281792641,
|
|
"num_tokens": 1589506.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.6439666748046875,
|
|
"epoch": 0.7249378624689312,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.543,
|
|
"mean_token_accuracy": 0.10177418887615204,
|
|
"num_tokens": 1598127.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.685148239135742,
|
|
"epoch": 0.7290803645401823,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5684,
|
|
"mean_token_accuracy": 0.10523401126265526,
|
|
"num_tokens": 1608073.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.695839357376099,
|
|
"epoch": 0.7332228666114333,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.587,
|
|
"mean_token_accuracy": 0.10461484640836716,
|
|
"num_tokens": 1616272.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.641492891311645,
|
|
"epoch": 0.7373653686826843,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.6058,
|
|
"mean_token_accuracy": 0.09599140211939812,
|
|
"num_tokens": 1625577.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.810349798202514,
|
|
"epoch": 0.7415078707539354,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.6845,
|
|
"mean_token_accuracy": 0.09567000940442086,
|
|
"num_tokens": 1634902.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.6667379379272464,
|
|
"epoch": 0.7456503728251864,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5977,
|
|
"mean_token_accuracy": 0.1015617847442627,
|
|
"num_tokens": 1644131.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.560829305648804,
|
|
"epoch": 0.7497928748964374,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5118,
|
|
"mean_token_accuracy": 0.11086556538939477,
|
|
"num_tokens": 1652272.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.772023248672485,
|
|
"epoch": 0.7539353769676885,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.6107,
|
|
"mean_token_accuracy": 0.10556575432419776,
|
|
"num_tokens": 1660962.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.561828851699829,
|
|
"epoch": 0.7580778790389395,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.539,
|
|
"mean_token_accuracy": 0.10557058155536651,
|
|
"num_tokens": 1670570.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.672806739807129,
|
|
"epoch": 0.7622203811101905,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5859,
|
|
"mean_token_accuracy": 0.10312820002436637,
|
|
"num_tokens": 1679674.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.678398847579956,
|
|
"epoch": 0.7663628831814416,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.565,
|
|
"mean_token_accuracy": 0.10954021513462067,
|
|
"num_tokens": 1688849.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.5227796077728275,
|
|
"epoch": 0.7705053852526926,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4534,
|
|
"mean_token_accuracy": 0.10681168809533119,
|
|
"num_tokens": 1698565.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.671809673309326,
|
|
"epoch": 0.7746478873239436,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.5485,
|
|
"mean_token_accuracy": 0.11151607260107994,
|
|
"num_tokens": 1708009.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.625020933151245,
|
|
"epoch": 0.7787903893951947,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.5537,
|
|
"mean_token_accuracy": 0.10801653489470482,
|
|
"num_tokens": 1716497.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.526713752746582,
|
|
"epoch": 0.7829328914664457,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.5447,
|
|
"mean_token_accuracy": 0.11139797195792198,
|
|
"num_tokens": 1725609.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.60435209274292,
|
|
"epoch": 0.7870753935376967,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4829,
|
|
"mean_token_accuracy": 0.10714786946773529,
|
|
"num_tokens": 1734165.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.554976940155029,
|
|
"epoch": 0.7912178956089478,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5828,
|
|
"mean_token_accuracy": 0.10263153910636902,
|
|
"num_tokens": 1744856.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.635064172744751,
|
|
"epoch": 0.7953603976801988,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5938,
|
|
"mean_token_accuracy": 0.1016570694744587,
|
|
"num_tokens": 1753386.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.552286529541016,
|
|
"epoch": 0.7995028997514498,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4448,
|
|
"mean_token_accuracy": 0.1083413727581501,
|
|
"num_tokens": 1762521.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.489339590072632,
|
|
"epoch": 0.8036454018227009,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.5099,
|
|
"mean_token_accuracy": 0.10716646909713745,
|
|
"num_tokens": 1772403.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.546666860580444,
|
|
"epoch": 0.8077879038939519,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3975,
|
|
"mean_token_accuracy": 0.1070093922317028,
|
|
"num_tokens": 1780924.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.623412990570069,
|
|
"epoch": 0.8119304059652029,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5443,
|
|
"mean_token_accuracy": 0.10209290608763695,
|
|
"num_tokens": 1789992.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.60557599067688,
|
|
"epoch": 0.816072908036454,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.5226,
|
|
"mean_token_accuracy": 0.11116698980331421,
|
|
"num_tokens": 1799513.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.434801244735718,
|
|
"epoch": 0.820215410107705,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3764,
|
|
"mean_token_accuracy": 0.10629670321941376,
|
|
"num_tokens": 1807393.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.577297115325928,
|
|
"epoch": 0.824357912178956,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.6109,
|
|
"mean_token_accuracy": 0.10033297091722489,
|
|
"num_tokens": 1817284.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.540279865264893,
|
|
"epoch": 0.8285004142502072,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.4127,
|
|
"mean_token_accuracy": 0.1117069512605667,
|
|
"num_tokens": 1825591.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8285004142502072,
|
|
"eval_entropy": 6.394233569407514,
|
|
"eval_loss": 6.601587295532227,
|
|
"eval_mean_token_accuracy": 0.1064636675454279,
|
|
"eval_num_tokens": 1825591.0,
|
|
"eval_runtime": 2.7507,
|
|
"eval_samples_per_second": 1362.188,
|
|
"eval_steps_per_second": 170.501,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.598631191253662,
|
|
"epoch": 0.8326429163214581,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999998547681232,
|
|
"loss": 6.5313,
|
|
"mean_token_accuracy": 0.10966545790433883,
|
|
"num_tokens": 1834655.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.366255187988282,
|
|
"epoch": 0.8367854183927091,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999992647639447,
|
|
"loss": 6.3637,
|
|
"mean_token_accuracy": 0.10939211472868919,
|
|
"num_tokens": 1844187.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.511251926422119,
|
|
"epoch": 0.8409279204639603,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999982209116618,
|
|
"loss": 6.445,
|
|
"mean_token_accuracy": 0.10647044554352761,
|
|
"num_tokens": 1852768.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.5381862163543705,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999967232133796,
|
|
"loss": 6.4772,
|
|
"mean_token_accuracy": 0.11856938228011131,
|
|
"num_tokens": 1862095.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.492068910598755,
|
|
"epoch": 0.8492129246064622,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999947716721196,
|
|
"loss": 6.524,
|
|
"mean_token_accuracy": 0.104226516187191,
|
|
"num_tokens": 1871239.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.561237716674805,
|
|
"epoch": 0.8533554266777134,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999923662918179,
|
|
"loss": 6.5388,
|
|
"mean_token_accuracy": 0.10700825974345207,
|
|
"num_tokens": 1880065.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.622053861618042,
|
|
"epoch": 0.8574979287489644,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999895070773266,
|
|
"loss": 6.5681,
|
|
"mean_token_accuracy": 0.10293134152889252,
|
|
"num_tokens": 1889065.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.498682117462158,
|
|
"epoch": 0.8616404308202155,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999861940344129,
|
|
"loss": 6.5818,
|
|
"mean_token_accuracy": 0.10599627569317818,
|
|
"num_tokens": 1898238.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.620818948745727,
|
|
"epoch": 0.8657829328914665,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999824271697597,
|
|
"loss": 6.5041,
|
|
"mean_token_accuracy": 0.11954014375805855,
|
|
"num_tokens": 1906780.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.587788820266724,
|
|
"epoch": 0.8699254349627175,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499978206490965,
|
|
"loss": 6.5855,
|
|
"mean_token_accuracy": 0.10859053954482079,
|
|
"num_tokens": 1916183.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.472416591644287,
|
|
"epoch": 0.8740679370339686,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999735320065427,
|
|
"loss": 6.4335,
|
|
"mean_token_accuracy": 0.11546618118882179,
|
|
"num_tokens": 1925382.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.469833660125732,
|
|
"epoch": 0.8782104391052196,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999684037259216,
|
|
"loss": 6.2955,
|
|
"mean_token_accuracy": 0.12006475627422333,
|
|
"num_tokens": 1933639.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.452498483657837,
|
|
"epoch": 0.8823529411764706,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999628216594456,
|
|
"loss": 6.5491,
|
|
"mean_token_accuracy": 0.10926123335957527,
|
|
"num_tokens": 1943044.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.504137325286865,
|
|
"epoch": 0.8864954432477217,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000499956785818375,
|
|
"loss": 6.4177,
|
|
"mean_token_accuracy": 0.10931780487298966,
|
|
"num_tokens": 1951777.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.467433595657349,
|
|
"epoch": 0.8906379453189727,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999502962148843,
|
|
"loss": 6.4799,
|
|
"mean_token_accuracy": 0.10977004319429398,
|
|
"num_tokens": 1961356.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.609413576126099,
|
|
"epoch": 0.8947804473902237,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499943352862064,
|
|
"loss": 6.529,
|
|
"mean_token_accuracy": 0.10762318223714828,
|
|
"num_tokens": 1970341.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.496694040298462,
|
|
"epoch": 0.8989229494614748,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999359557739194,
|
|
"loss": 6.383,
|
|
"mean_token_accuracy": 0.11510946005582809,
|
|
"num_tokens": 1978576.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.428346872329712,
|
|
"epoch": 0.9030654515327258,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999281049653713,
|
|
"loss": 6.4282,
|
|
"mean_token_accuracy": 0.11499780714511872,
|
|
"num_tokens": 1987229.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.449847650527954,
|
|
"epoch": 0.9072079536039768,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999198004522557,
|
|
"loss": 6.3618,
|
|
"mean_token_accuracy": 0.11787119433283806,
|
|
"num_tokens": 1996473.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.411348676681518,
|
|
"epoch": 0.9113504556752279,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999110422513237,
|
|
"loss": 6.3786,
|
|
"mean_token_accuracy": 0.11625739708542823,
|
|
"num_tokens": 2005551.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.447707653045654,
|
|
"epoch": 0.9154929577464789,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999018303802414,
|
|
"loss": 6.4902,
|
|
"mean_token_accuracy": 0.11025101318955421,
|
|
"num_tokens": 2015012.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.500209903717041,
|
|
"epoch": 0.9196354598177299,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998921648575904,
|
|
"loss": 6.4993,
|
|
"mean_token_accuracy": 0.1120341457426548,
|
|
"num_tokens": 2024435.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.45713210105896,
|
|
"epoch": 0.923777961888981,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998820457028669,
|
|
"loss": 6.3952,
|
|
"mean_token_accuracy": 0.11182878091931343,
|
|
"num_tokens": 2033991.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.520167350769043,
|
|
"epoch": 0.927920463960232,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998714729364824,
|
|
"loss": 6.4059,
|
|
"mean_token_accuracy": 0.1159439042210579,
|
|
"num_tokens": 2043117.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.4226549625396725,
|
|
"epoch": 0.932062966031483,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998604465797634,
|
|
"loss": 6.4284,
|
|
"mean_token_accuracy": 0.11458204984664917,
|
|
"num_tokens": 2052304.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.308629035949707,
|
|
"epoch": 0.9362054681027341,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998489666549512,
|
|
"loss": 6.354,
|
|
"mean_token_accuracy": 0.1193151444196701,
|
|
"num_tokens": 2061217.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.363573312759399,
|
|
"epoch": 0.9403479701739851,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998370331852021,
|
|
"loss": 6.3565,
|
|
"mean_token_accuracy": 0.11765335276722907,
|
|
"num_tokens": 2070796.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.606226682662964,
|
|
"epoch": 0.9444904722452361,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998246461945872,
|
|
"loss": 6.4943,
|
|
"mean_token_accuracy": 0.11730922684073448,
|
|
"num_tokens": 2080202.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.445854616165161,
|
|
"epoch": 0.9486329743164872,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998118057080923,
|
|
"loss": 6.4025,
|
|
"mean_token_accuracy": 0.11324022263288498,
|
|
"num_tokens": 2089788.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.401117420196533,
|
|
"epoch": 0.9527754763877382,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004997985117516183,
|
|
"loss": 6.4367,
|
|
"mean_token_accuracy": 0.11083348914980888,
|
|
"num_tokens": 2098727.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.464313554763794,
|
|
"epoch": 0.9569179784589892,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997847643519803,
|
|
"loss": 6.322,
|
|
"mean_token_accuracy": 0.11506827175617218,
|
|
"num_tokens": 2107645.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.274763011932373,
|
|
"epoch": 0.9610604805302403,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997705635369085,
|
|
"loss": 6.3093,
|
|
"mean_token_accuracy": 0.11011608317494392,
|
|
"num_tokens": 2116613.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.368350839614868,
|
|
"epoch": 0.9652029826014913,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997559093350474,
|
|
"loss": 6.3013,
|
|
"mean_token_accuracy": 0.11394222304224969,
|
|
"num_tokens": 2125585.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.498920965194702,
|
|
"epoch": 0.9693454846727423,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997408017759562,
|
|
"loss": 6.4282,
|
|
"mean_token_accuracy": 0.11357887014746666,
|
|
"num_tokens": 2134795.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.396416473388672,
|
|
"epoch": 0.9734879867439934,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997252408901084,
|
|
"loss": 6.4224,
|
|
"mean_token_accuracy": 0.11246276721358299,
|
|
"num_tokens": 2143699.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.5009948253631595,
|
|
"epoch": 0.9776304888152444,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997092267088922,
|
|
"loss": 6.3638,
|
|
"mean_token_accuracy": 0.1147896520793438,
|
|
"num_tokens": 2152187.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.340474557876587,
|
|
"epoch": 0.9817729908864954,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996927592646097,
|
|
"loss": 6.4264,
|
|
"mean_token_accuracy": 0.11860019415616989,
|
|
"num_tokens": 2160941.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.451793146133423,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996758385904776,
|
|
"loss": 6.4028,
|
|
"mean_token_accuracy": 0.11521425396203995,
|
|
"num_tokens": 2168946.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.310120868682861,
|
|
"epoch": 0.9900579950289975,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996584647206269,
|
|
"loss": 6.315,
|
|
"mean_token_accuracy": 0.11753729954361916,
|
|
"num_tokens": 2177662.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.437413454055786,
|
|
"epoch": 0.9942004971002486,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996406376901026,
|
|
"loss": 6.4181,
|
|
"mean_token_accuracy": 0.1255168005824089,
|
|
"num_tokens": 2187461.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.393413114547729,
|
|
"epoch": 0.9983429991714996,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996223575348635,
|
|
"loss": 6.3,
|
|
"mean_token_accuracy": 0.12105931863188743,
|
|
"num_tokens": 2196878.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.42917308807373,
|
|
"epoch": 1.0024855012427507,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996036242917828,
|
|
"loss": 6.2979,
|
|
"mean_token_accuracy": 0.11524101421236992,
|
|
"num_tokens": 2206178.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.32910008430481,
|
|
"epoch": 1.0066280033140016,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995844379986476,
|
|
"loss": 6.1628,
|
|
"mean_token_accuracy": 0.1222687229514122,
|
|
"num_tokens": 2215330.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.457520198822022,
|
|
"epoch": 1.0107705053852527,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995647986941587,
|
|
"loss": 6.1389,
|
|
"mean_token_accuracy": 0.12433998733758926,
|
|
"num_tokens": 2224146.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.348219490051269,
|
|
"epoch": 1.0149130074565038,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995447064179308,
|
|
"loss": 6.1297,
|
|
"mean_token_accuracy": 0.12844619452953338,
|
|
"num_tokens": 2233646.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.102799320220948,
|
|
"epoch": 1.0190555095277547,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000499524161210492,
|
|
"loss": 6.0188,
|
|
"mean_token_accuracy": 0.12751995995640755,
|
|
"num_tokens": 2243663.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.376331901550293,
|
|
"epoch": 1.0231980115990058,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995031631132844,
|
|
"loss": 6.1946,
|
|
"mean_token_accuracy": 0.13004939407110214,
|
|
"num_tokens": 2252564.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.368073844909668,
|
|
"epoch": 1.027340513670257,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004994817121686633,
|
|
"loss": 6.1988,
|
|
"mean_token_accuracy": 0.11914310157299042,
|
|
"num_tokens": 2262504.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.29468092918396,
|
|
"epoch": 1.0314830157415078,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004994598084198979,
|
|
"loss": 6.0995,
|
|
"mean_token_accuracy": 0.13067468032240867,
|
|
"num_tokens": 2270553.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.198537969589234,
|
|
"epoch": 1.035625517812759,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004994374519111701,
|
|
"loss": 6.0946,
|
|
"mean_token_accuracy": 0.12411084324121475,
|
|
"num_tokens": 2279319.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.336030292510986,
|
|
"epoch": 1.03976801988401,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994146426875757,
|
|
"loss": 6.2448,
|
|
"mean_token_accuracy": 0.12142780497670173,
|
|
"num_tokens": 2288749.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.316008758544922,
|
|
"epoch": 1.043910521955261,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004993913807951232,
|
|
"loss": 6.1098,
|
|
"mean_token_accuracy": 0.12393713593482972,
|
|
"num_tokens": 2298272.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.254574918746949,
|
|
"epoch": 1.048053024026512,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004993676662807345,
|
|
"loss": 6.1518,
|
|
"mean_token_accuracy": 0.12555598765611647,
|
|
"num_tokens": 2307230.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.343469524383545,
|
|
"epoch": 1.0521955260977631,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004993434991922443,
|
|
"loss": 6.2458,
|
|
"mean_token_accuracy": 0.11990855634212494,
|
|
"num_tokens": 2316306.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.305857372283936,
|
|
"epoch": 1.056338028169014,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004993188795784002,
|
|
"loss": 6.1404,
|
|
"mean_token_accuracy": 0.12261386215686798,
|
|
"num_tokens": 2325672.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.25411434173584,
|
|
"epoch": 1.0604805302402651,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004992938074888628,
|
|
"loss": 6.1622,
|
|
"mean_token_accuracy": 0.12227201014757157,
|
|
"num_tokens": 2335251.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.313401746749878,
|
|
"epoch": 1.0646230323115162,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004992682829742052,
|
|
"loss": 6.1248,
|
|
"mean_token_accuracy": 0.1251654766499996,
|
|
"num_tokens": 2344054.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.137462472915649,
|
|
"epoch": 1.0687655343827671,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004992423060859132,
|
|
"loss": 6.1157,
|
|
"mean_token_accuracy": 0.127670818567276,
|
|
"num_tokens": 2353493.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.367279481887818,
|
|
"epoch": 1.0729080364540182,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992158768763849,
|
|
"loss": 6.2362,
|
|
"mean_token_accuracy": 0.12816474214196205,
|
|
"num_tokens": 2362687.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.272933483123779,
|
|
"epoch": 1.0770505385252693,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004991889953989308,
|
|
"loss": 6.208,
|
|
"mean_token_accuracy": 0.12367280796170235,
|
|
"num_tokens": 2372325.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.2988440990448,
|
|
"epoch": 1.0811930405965202,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991616617077741,
|
|
"loss": 6.2379,
|
|
"mean_token_accuracy": 0.12832726687192916,
|
|
"num_tokens": 2381987.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.4650335788726805,
|
|
"epoch": 1.0853355426677713,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004991338758580496,
|
|
"loss": 6.125,
|
|
"mean_token_accuracy": 0.12739104330539702,
|
|
"num_tokens": 2390901.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.090523910522461,
|
|
"epoch": 1.0894780447390224,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004991056379058045,
|
|
"loss": 6.1601,
|
|
"mean_token_accuracy": 0.12110325694084167,
|
|
"num_tokens": 2400756.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.28847484588623,
|
|
"epoch": 1.0936205468102733,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004990769479079976,
|
|
"loss": 6.004,
|
|
"mean_token_accuracy": 0.12852884232997894,
|
|
"num_tokens": 2409526.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.171352672576904,
|
|
"epoch": 1.0977630488815244,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004990478059225003,
|
|
"loss": 6.194,
|
|
"mean_token_accuracy": 0.12339239865541458,
|
|
"num_tokens": 2419583.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.393887710571289,
|
|
"epoch": 1.1019055509527755,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004990182120080945,
|
|
"loss": 6.1337,
|
|
"mean_token_accuracy": 0.12826344668865203,
|
|
"num_tokens": 2429439.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.1219724178314205,
|
|
"epoch": 1.1060480530240264,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989881662244748,
|
|
"loss": 6.1239,
|
|
"mean_token_accuracy": 0.12123734056949616,
|
|
"num_tokens": 2438467.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.308976459503174,
|
|
"epoch": 1.1101905550952775,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004989576686322468,
|
|
"loss": 6.0713,
|
|
"mean_token_accuracy": 0.13452951610088348,
|
|
"num_tokens": 2446619.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.185806179046631,
|
|
"epoch": 1.1143330571665286,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004989267192929272,
|
|
"loss": 6.1432,
|
|
"mean_token_accuracy": 0.12741851955652236,
|
|
"num_tokens": 2454998.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.214543724060059,
|
|
"epoch": 1.1184755592377795,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004988953182689445,
|
|
"loss": 6.117,
|
|
"mean_token_accuracy": 0.12976484671235083,
|
|
"num_tokens": 2463497.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.218276834487915,
|
|
"epoch": 1.1226180613090306,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004988634656236378,
|
|
"loss": 6.0278,
|
|
"mean_token_accuracy": 0.1300082579255104,
|
|
"num_tokens": 2471763.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.184001922607422,
|
|
"epoch": 1.1267605633802817,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004988311614212576,
|
|
"loss": 6.0879,
|
|
"mean_token_accuracy": 0.13324105516076087,
|
|
"num_tokens": 2479949.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.281342220306397,
|
|
"epoch": 1.1309030654515326,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987984057269648,
|
|
"loss": 6.1428,
|
|
"mean_token_accuracy": 0.1314686894416809,
|
|
"num_tokens": 2489097.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.040010833740235,
|
|
"epoch": 1.1350455675227837,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004987651986068311,
|
|
"loss": 5.9729,
|
|
"mean_token_accuracy": 0.13274277672171592,
|
|
"num_tokens": 2497625.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.246344804763794,
|
|
"epoch": 1.1391880695940348,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004987315401278394,
|
|
"loss": 6.216,
|
|
"mean_token_accuracy": 0.12787129282951354,
|
|
"num_tokens": 2507107.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.39310736656189,
|
|
"epoch": 1.143330571665286,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004986974303578819,
|
|
"loss": 6.2224,
|
|
"mean_token_accuracy": 0.12255537360906602,
|
|
"num_tokens": 2516629.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.220417118072509,
|
|
"epoch": 1.1474730737365368,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986628693657621,
|
|
"loss": 6.0929,
|
|
"mean_token_accuracy": 0.13176359012722969,
|
|
"num_tokens": 2525924.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.315376138687133,
|
|
"epoch": 1.151615575807788,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004986278572211933,
|
|
"loss": 6.2518,
|
|
"mean_token_accuracy": 0.12930550798773766,
|
|
"num_tokens": 2535586.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.2914097785949705,
|
|
"epoch": 1.1557580778790388,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004985923939947987,
|
|
"loss": 6.3282,
|
|
"mean_token_accuracy": 0.12142003327608109,
|
|
"num_tokens": 2546685.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.185162925720215,
|
|
"epoch": 1.15990057995029,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004985564797581115,
|
|
"loss": 6.0433,
|
|
"mean_token_accuracy": 0.12628076002001762,
|
|
"num_tokens": 2556873.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.250091075897217,
|
|
"epoch": 1.164043082021541,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004985201145835747,
|
|
"loss": 6.0778,
|
|
"mean_token_accuracy": 0.13431011289358138,
|
|
"num_tokens": 2565222.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.09890832901001,
|
|
"epoch": 1.1681855840927922,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000498483298544541,
|
|
"loss": 6.0752,
|
|
"mean_token_accuracy": 0.1278785362839699,
|
|
"num_tokens": 2575467.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.252477359771729,
|
|
"epoch": 1.172328086164043,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000498446031715272,
|
|
"loss": 6.0932,
|
|
"mean_token_accuracy": 0.1310752049088478,
|
|
"num_tokens": 2583963.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.18706784248352,
|
|
"epoch": 1.1764705882352942,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004984083141709391,
|
|
"loss": 6.1308,
|
|
"mean_token_accuracy": 0.1298580549657345,
|
|
"num_tokens": 2593350.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.089730072021484,
|
|
"epoch": 1.180613090306545,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000498370145987623,
|
|
"loss": 6.0316,
|
|
"mean_token_accuracy": 0.13288710117340088,
|
|
"num_tokens": 2602191.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.324441194534302,
|
|
"epoch": 1.1847555923777962,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004983315272423127,
|
|
"loss": 6.176,
|
|
"mean_token_accuracy": 0.12697831466794013,
|
|
"num_tokens": 2613590.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.06857738494873,
|
|
"epoch": 1.1888980944490473,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004982924580129066,
|
|
"loss": 5.9812,
|
|
"mean_token_accuracy": 0.13065367564558983,
|
|
"num_tokens": 2622072.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.248331546783447,
|
|
"epoch": 1.1930405965202984,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004982529383782116,
|
|
"loss": 6.1865,
|
|
"mean_token_accuracy": 0.1266859605908394,
|
|
"num_tokens": 2630904.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.229864883422851,
|
|
"epoch": 1.1971830985915493,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004982129684179432,
|
|
"loss": 6.1479,
|
|
"mean_token_accuracy": 0.12835109010338783,
|
|
"num_tokens": 2639925.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.201044178009033,
|
|
"epoch": 1.2013256006628004,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004981725482127251,
|
|
"loss": 6.0555,
|
|
"mean_token_accuracy": 0.13077252507209777,
|
|
"num_tokens": 2649397.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.251532125473022,
|
|
"epoch": 1.2054681027340513,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004981316778440892,
|
|
"loss": 6.1979,
|
|
"mean_token_accuracy": 0.1274854823946953,
|
|
"num_tokens": 2658877.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.10891489982605,
|
|
"epoch": 1.2096106048053024,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004980903573944756,
|
|
"loss": 6.0963,
|
|
"mean_token_accuracy": 0.1251230150461197,
|
|
"num_tokens": 2668518.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.220654296875,
|
|
"epoch": 1.2137531068765535,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004980485869472322,
|
|
"loss": 6.2062,
|
|
"mean_token_accuracy": 0.1263741858303547,
|
|
"num_tokens": 2677459.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.302640914916992,
|
|
"epoch": 1.2178956089478046,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004980063665866143,
|
|
"loss": 6.1744,
|
|
"mean_token_accuracy": 0.1222662165760994,
|
|
"num_tokens": 2686668.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.034982395172119,
|
|
"epoch": 1.2220381110190555,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004979636963977853,
|
|
"loss": 5.9536,
|
|
"mean_token_accuracy": 0.13268350437283516,
|
|
"num_tokens": 2695545.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.225825643539428,
|
|
"epoch": 1.2261806130903066,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004979205764668156,
|
|
"loss": 6.1317,
|
|
"mean_token_accuracy": 0.12943535372614862,
|
|
"num_tokens": 2705514.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.211910057067871,
|
|
"epoch": 1.2303231151615575,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004978770068806825,
|
|
"loss": 6.0386,
|
|
"mean_token_accuracy": 0.12719125896692277,
|
|
"num_tokens": 2714373.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.15532283782959,
|
|
"epoch": 1.2344656172328086,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497832987727271,
|
|
"loss": 6.1529,
|
|
"mean_token_accuracy": 0.12199215218424797,
|
|
"num_tokens": 2723887.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.166454887390136,
|
|
"epoch": 1.2386081193040597,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004977885190953724,
|
|
"loss": 6.0362,
|
|
"mean_token_accuracy": 0.12878855913877488,
|
|
"num_tokens": 2733591.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.207261800765991,
|
|
"epoch": 1.2427506213753108,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004977436010746846,
|
|
"loss": 6.1471,
|
|
"mean_token_accuracy": 0.13749015629291533,
|
|
"num_tokens": 2741638.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2427506213753108,
|
|
"eval_entropy": 6.123490674155099,
|
|
"eval_loss": 6.331637382507324,
|
|
"eval_mean_token_accuracy": 0.12770517323730088,
|
|
"eval_num_tokens": 2741638.0,
|
|
"eval_runtime": 2.9708,
|
|
"eval_samples_per_second": 1261.266,
|
|
"eval_steps_per_second": 157.869,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.160883140563965,
|
|
"epoch": 1.2468931234465617,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004976982337558124,
|
|
"loss": 6.0387,
|
|
"mean_token_accuracy": 0.133138719946146,
|
|
"num_tokens": 2749711.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.207182788848877,
|
|
"epoch": 1.2510356255178128,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004976524172302665,
|
|
"loss": 6.2191,
|
|
"mean_token_accuracy": 0.12747407108545303,
|
|
"num_tokens": 2758121.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.156457328796387,
|
|
"epoch": 1.2551781275890637,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004976061515904641,
|
|
"loss": 6.2125,
|
|
"mean_token_accuracy": 0.12646155208349227,
|
|
"num_tokens": 2766764.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.284233474731446,
|
|
"epoch": 1.2593206296603148,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004975594369297277,
|
|
"loss": 6.1469,
|
|
"mean_token_accuracy": 0.1318532146513462,
|
|
"num_tokens": 2775561.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.1375274658203125,
|
|
"epoch": 1.2634631317315659,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004975122733422861,
|
|
"loss": 6.1051,
|
|
"mean_token_accuracy": 0.1313116282224655,
|
|
"num_tokens": 2784691.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.224356698989868,
|
|
"epoch": 1.267605633802817,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004974646609232732,
|
|
"loss": 6.0634,
|
|
"mean_token_accuracy": 0.13405025228857995,
|
|
"num_tokens": 2793893.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.1460289478302,
|
|
"epoch": 1.2717481358740679,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004974165997687288,
|
|
"loss": 6.1463,
|
|
"mean_token_accuracy": 0.12966038882732392,
|
|
"num_tokens": 2804479.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.288549613952637,
|
|
"epoch": 1.275890637945319,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004973680899755972,
|
|
"loss": 6.1883,
|
|
"mean_token_accuracy": 0.12688734605908394,
|
|
"num_tokens": 2813966.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.1371747016906735,
|
|
"epoch": 1.2800331400165699,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000497319131641728,
|
|
"loss": 5.998,
|
|
"mean_token_accuracy": 0.13688533008098602,
|
|
"num_tokens": 2822328.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.090888452529907,
|
|
"epoch": 1.284175642087821,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004972697248658755,
|
|
"loss": 5.9801,
|
|
"mean_token_accuracy": 0.13497933447360994,
|
|
"num_tokens": 2829845.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.0702019214630125,
|
|
"epoch": 1.288318144159072,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004972198697476987,
|
|
"loss": 5.9519,
|
|
"mean_token_accuracy": 0.13229596316814424,
|
|
"num_tokens": 2839040.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.079361152648926,
|
|
"epoch": 1.2924606462303232,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004971695663877607,
|
|
"loss": 6.043,
|
|
"mean_token_accuracy": 0.13861832544207572,
|
|
"num_tokens": 2848574.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.124926471710205,
|
|
"epoch": 1.296603148301574,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004971188148875289,
|
|
"loss": 6.0208,
|
|
"mean_token_accuracy": 0.14211280643939972,
|
|
"num_tokens": 2856812.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.180253314971924,
|
|
"epoch": 1.3007456503728252,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004970676153493748,
|
|
"loss": 6.1061,
|
|
"mean_token_accuracy": 0.12911001816391945,
|
|
"num_tokens": 2866001.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.168315410614014,
|
|
"epoch": 1.3048881524440763,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004970159678765731,
|
|
"loss": 6.1004,
|
|
"mean_token_accuracy": 0.12929908260703088,
|
|
"num_tokens": 2875181.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.167870616912841,
|
|
"epoch": 1.3090306545153272,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004969638725733028,
|
|
"loss": 6.0246,
|
|
"mean_token_accuracy": 0.13357625305652618,
|
|
"num_tokens": 2885103.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.009994029998779,
|
|
"epoch": 1.3131731565865783,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004969113295446455,
|
|
"loss": 5.9852,
|
|
"mean_token_accuracy": 0.13527049124240875,
|
|
"num_tokens": 2894055.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.2158726215362545,
|
|
"epoch": 1.3173156586578294,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004968583388965863,
|
|
"loss": 6.0608,
|
|
"mean_token_accuracy": 0.13581639677286148,
|
|
"num_tokens": 2903391.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.093079280853272,
|
|
"epoch": 1.3214581607290803,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004968049007360132,
|
|
"loss": 6.0605,
|
|
"mean_token_accuracy": 0.13181073293089868,
|
|
"num_tokens": 2911760.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.063480806350708,
|
|
"epoch": 1.3256006628003314,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004967510151707169,
|
|
"loss": 6.0138,
|
|
"mean_token_accuracy": 0.13460284769535064,
|
|
"num_tokens": 2920589.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.224989795684815,
|
|
"epoch": 1.3297431648715825,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004966966823093903,
|
|
"loss": 6.1592,
|
|
"mean_token_accuracy": 0.12850213572382926,
|
|
"num_tokens": 2929635.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.084767436981201,
|
|
"epoch": 1.3338856669428334,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004966419022616287,
|
|
"loss": 6.0021,
|
|
"mean_token_accuracy": 0.13327318727970122,
|
|
"num_tokens": 2938778.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.075594139099121,
|
|
"epoch": 1.3380281690140845,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004965866751379296,
|
|
"loss": 6.1079,
|
|
"mean_token_accuracy": 0.1267368644475937,
|
|
"num_tokens": 2947836.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.235728120803833,
|
|
"epoch": 1.3421706710853356,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004965310010496922,
|
|
"loss": 6.0749,
|
|
"mean_token_accuracy": 0.12694141641259193,
|
|
"num_tokens": 2957028.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.089836168289184,
|
|
"epoch": 1.3463131731565865,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004964748801092168,
|
|
"loss": 6.0329,
|
|
"mean_token_accuracy": 0.12828338146209717,
|
|
"num_tokens": 2966088.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.1745452880859375,
|
|
"epoch": 1.3504556752278376,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004964183124297059,
|
|
"loss": 6.1172,
|
|
"mean_token_accuracy": 0.1309424474835396,
|
|
"num_tokens": 2975271.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.151106929779052,
|
|
"epoch": 1.3545981772990887,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004963612981252625,
|
|
"loss": 6.1097,
|
|
"mean_token_accuracy": 0.13138212487101555,
|
|
"num_tokens": 2983799.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.1551590919494625,
|
|
"epoch": 1.3587406793703396,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004963038373108908,
|
|
"loss": 6.0853,
|
|
"mean_token_accuracy": 0.13040598779916762,
|
|
"num_tokens": 2992675.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.182181072235108,
|
|
"epoch": 1.3628831814415907,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004962459301024954,
|
|
"loss": 6.1637,
|
|
"mean_token_accuracy": 0.12975483611226082,
|
|
"num_tokens": 3002483.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.180190896987915,
|
|
"epoch": 1.3670256835128418,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004961875766168816,
|
|
"loss": 6.0541,
|
|
"mean_token_accuracy": 0.13903512582182884,
|
|
"num_tokens": 3010611.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.102008962631226,
|
|
"epoch": 1.371168185584093,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004961287769717547,
|
|
"loss": 6.0193,
|
|
"mean_token_accuracy": 0.12874715775251389,
|
|
"num_tokens": 3019928.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.066634035110473,
|
|
"epoch": 1.3753106876553438,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004960695312857199,
|
|
"loss": 6.0043,
|
|
"mean_token_accuracy": 0.13317709937691688,
|
|
"num_tokens": 3029003.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.0940875053405765,
|
|
"epoch": 1.379453189726595,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004960098396782824,
|
|
"loss": 6.0235,
|
|
"mean_token_accuracy": 0.13606003373861314,
|
|
"num_tokens": 3038049.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.055018329620362,
|
|
"epoch": 1.3835956917978458,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004959497022698466,
|
|
"loss": 6.0211,
|
|
"mean_token_accuracy": 0.1442451775074005,
|
|
"num_tokens": 3046674.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.138688325881958,
|
|
"epoch": 1.387738193869097,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004958891191817163,
|
|
"loss": 6.0594,
|
|
"mean_token_accuracy": 0.1267915315926075,
|
|
"num_tokens": 3056459.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.11493444442749,
|
|
"epoch": 1.391880695940348,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000495828090536094,
|
|
"loss": 6.0525,
|
|
"mean_token_accuracy": 0.1292160801589489,
|
|
"num_tokens": 3066294.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.1094911098480225,
|
|
"epoch": 1.3960231980115991,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004957666164560815,
|
|
"loss": 6.017,
|
|
"mean_token_accuracy": 0.13060638532042504,
|
|
"num_tokens": 3075420.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.1826433658599855,
|
|
"epoch": 1.40016570008285,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004957046970656784,
|
|
"loss": 6.1106,
|
|
"mean_token_accuracy": 0.1357931189239025,
|
|
"num_tokens": 3084373.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.090426683425903,
|
|
"epoch": 1.4043082021541011,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004956423324897831,
|
|
"loss": 6.0819,
|
|
"mean_token_accuracy": 0.1346761591732502,
|
|
"num_tokens": 3094212.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.164022970199585,
|
|
"epoch": 1.408450704225352,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004955795228541917,
|
|
"loss": 6.0205,
|
|
"mean_token_accuracy": 0.14607309103012084,
|
|
"num_tokens": 3102769.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.0860377788543705,
|
|
"epoch": 1.4125932062966031,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000495516268285598,
|
|
"loss": 6.0509,
|
|
"mean_token_accuracy": 0.13397998213768006,
|
|
"num_tokens": 3111465.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.0558473587036135,
|
|
"epoch": 1.4167357083678542,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004954525689115935,
|
|
"loss": 5.9885,
|
|
"mean_token_accuracy": 0.13343949466943741,
|
|
"num_tokens": 3121106.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.11056923866272,
|
|
"epoch": 1.4208782104391053,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004953884248606669,
|
|
"loss": 6.0095,
|
|
"mean_token_accuracy": 0.14755313694477082,
|
|
"num_tokens": 3130662.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.031003999710083,
|
|
"epoch": 1.4250207125103562,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004953238362622034,
|
|
"loss": 6.0571,
|
|
"mean_token_accuracy": 0.1311265155673027,
|
|
"num_tokens": 3139538.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.109453105926514,
|
|
"epoch": 1.4291632145816073,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004952588032464855,
|
|
"loss": 6.1066,
|
|
"mean_token_accuracy": 0.1264252707362175,
|
|
"num_tokens": 3148552.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.092384767532349,
|
|
"epoch": 1.4333057166528582,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004951933259446919,
|
|
"loss": 6.0087,
|
|
"mean_token_accuracy": 0.13966907411813737,
|
|
"num_tokens": 3156549.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.223468208312989,
|
|
"epoch": 1.4374482187241093,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004951274044888975,
|
|
"loss": 6.1692,
|
|
"mean_token_accuracy": 0.12101665809750557,
|
|
"num_tokens": 3166821.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.01346492767334,
|
|
"epoch": 1.4415907207953604,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004950610390120731,
|
|
"loss": 6.0518,
|
|
"mean_token_accuracy": 0.1328336514532566,
|
|
"num_tokens": 3174962.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.186932849884033,
|
|
"epoch": 1.4457332228666115,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000494994229648085,
|
|
"loss": 5.957,
|
|
"mean_token_accuracy": 0.13281209170818328,
|
|
"num_tokens": 3184027.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.017684364318848,
|
|
"epoch": 1.4498757249378624,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004949269765316948,
|
|
"loss": 6.0971,
|
|
"mean_token_accuracy": 0.1264866605401039,
|
|
"num_tokens": 3193186.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.227017116546631,
|
|
"epoch": 1.4540182270091135,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.0004948592797985599,
|
|
"loss": 6.1057,
|
|
"mean_token_accuracy": 0.12643057182431222,
|
|
"num_tokens": 3202676.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.240481233596801,
|
|
"epoch": 1.4581607290803644,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004947911395852317,
|
|
"loss": 6.2243,
|
|
"mean_token_accuracy": 0.13314623907208442,
|
|
"num_tokens": 3211950.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.1601158618927006,
|
|
"epoch": 1.4623032311516155,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004947225560291565,
|
|
"loss": 6.1576,
|
|
"mean_token_accuracy": 0.12665070593357086,
|
|
"num_tokens": 3221135.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.144661283493042,
|
|
"epoch": 1.4664457332228666,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004946535292686746,
|
|
"loss": 5.9717,
|
|
"mean_token_accuracy": 0.13558048903942108,
|
|
"num_tokens": 3230174.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.993796110153198,
|
|
"epoch": 1.4705882352941178,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004945840594430209,
|
|
"loss": 5.948,
|
|
"mean_token_accuracy": 0.14206447452306747,
|
|
"num_tokens": 3238898.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.077493000030517,
|
|
"epoch": 1.4747307373653686,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004945141466923235,
|
|
"loss": 5.988,
|
|
"mean_token_accuracy": 0.14010899588465692,
|
|
"num_tokens": 3247550.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.233349943161011,
|
|
"epoch": 1.4788732394366197,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004944437911576041,
|
|
"loss": 6.1399,
|
|
"mean_token_accuracy": 0.12699589803814887,
|
|
"num_tokens": 3256810.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.942909288406372,
|
|
"epoch": 1.4830157415078706,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004943729929807772,
|
|
"loss": 5.9276,
|
|
"mean_token_accuracy": 0.13631228655576705,
|
|
"num_tokens": 3265886.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.0962855339050295,
|
|
"epoch": 1.4871582435791217,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004943017523046507,
|
|
"loss": 6.0033,
|
|
"mean_token_accuracy": 0.13771791383624077,
|
|
"num_tokens": 3274522.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.160940313339234,
|
|
"epoch": 1.4913007456503728,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000494230069272925,
|
|
"loss": 6.1139,
|
|
"mean_token_accuracy": 0.1388836421072483,
|
|
"num_tokens": 3283388.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.049196195602417,
|
|
"epoch": 1.495443247721624,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004941579440301923,
|
|
"loss": 5.9919,
|
|
"mean_token_accuracy": 0.13140289038419722,
|
|
"num_tokens": 3291887.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.019881868362427,
|
|
"epoch": 1.4995857497928748,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004940853767219372,
|
|
"loss": 6.0475,
|
|
"mean_token_accuracy": 0.1365261606872082,
|
|
"num_tokens": 3300674.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.014644193649292,
|
|
"epoch": 1.503728251864126,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004940123674945359,
|
|
"loss": 5.8999,
|
|
"mean_token_accuracy": 0.13877790048718452,
|
|
"num_tokens": 3309745.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.119295310974121,
|
|
"epoch": 1.5078707539353768,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004939389164952561,
|
|
"loss": 6.0591,
|
|
"mean_token_accuracy": 0.13487198427319527,
|
|
"num_tokens": 3318888.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.966858196258545,
|
|
"epoch": 1.512013256006628,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004938650238722564,
|
|
"loss": 5.9897,
|
|
"mean_token_accuracy": 0.13288637325167657,
|
|
"num_tokens": 3328274.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.136635732650757,
|
|
"epoch": 1.516155758077879,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004937906897745861,
|
|
"loss": 6.0095,
|
|
"mean_token_accuracy": 0.14079945385456086,
|
|
"num_tokens": 3337579.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.079625654220581,
|
|
"epoch": 1.5202982601491302,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004937159143521856,
|
|
"loss": 5.9861,
|
|
"mean_token_accuracy": 0.13965884745121002,
|
|
"num_tokens": 3346262.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.045686912536621,
|
|
"epoch": 1.5244407622203813,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004936406977558847,
|
|
"loss": 6.0836,
|
|
"mean_token_accuracy": 0.12937265560030936,
|
|
"num_tokens": 3354833.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.17369589805603,
|
|
"epoch": 1.5285832642916322,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004935650401374039,
|
|
"loss": 6.1543,
|
|
"mean_token_accuracy": 0.14100572615861892,
|
|
"num_tokens": 3364314.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.07831563949585,
|
|
"epoch": 1.532725766362883,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004934889416493524,
|
|
"loss": 5.9693,
|
|
"mean_token_accuracy": 0.13386041671037674,
|
|
"num_tokens": 3373648.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.98188886642456,
|
|
"epoch": 1.5368682684341342,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004934124024452295,
|
|
"loss": 5.8443,
|
|
"mean_token_accuracy": 0.14330715388059617,
|
|
"num_tokens": 3382919.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.019197273254394,
|
|
"epoch": 1.5410107705053853,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004933354226794232,
|
|
"loss": 5.9833,
|
|
"mean_token_accuracy": 0.1422867774963379,
|
|
"num_tokens": 3391814.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.095344734191895,
|
|
"epoch": 1.5451532725766364,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004932580025072098,
|
|
"loss": 6.0449,
|
|
"mean_token_accuracy": 0.13532943651080132,
|
|
"num_tokens": 3400392.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.067120552062988,
|
|
"epoch": 1.5492957746478875,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004931801420847545,
|
|
"loss": 6.0924,
|
|
"mean_token_accuracy": 0.12777185067534447,
|
|
"num_tokens": 3410430.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.015319013595581,
|
|
"epoch": 1.5534382767191384,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004931018415691102,
|
|
"loss": 5.9287,
|
|
"mean_token_accuracy": 0.13285826817154883,
|
|
"num_tokens": 3420654.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.0661335468292235,
|
|
"epoch": 1.5575807787903893,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004930231011182176,
|
|
"loss": 6.0527,
|
|
"mean_token_accuracy": 0.13392168357968331,
|
|
"num_tokens": 3430514.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.158822917938233,
|
|
"epoch": 1.5617232808616404,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004929439208909048,
|
|
"loss": 6.0031,
|
|
"mean_token_accuracy": 0.13457945808768274,
|
|
"num_tokens": 3439551.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.95737943649292,
|
|
"epoch": 1.5658657829328915,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004928643010468869,
|
|
"loss": 5.9409,
|
|
"mean_token_accuracy": 0.13493541032075881,
|
|
"num_tokens": 3448160.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.1136833190917965,
|
|
"epoch": 1.5700082850041426,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004927842417467659,
|
|
"loss": 6.0155,
|
|
"mean_token_accuracy": 0.14148347079753876,
|
|
"num_tokens": 3457296.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.018945932388306,
|
|
"epoch": 1.5741507870753937,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004927037431520302,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.14098128974437713,
|
|
"num_tokens": 3466108.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.053025722503662,
|
|
"epoch": 1.5782932891466446,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004926228054250541,
|
|
"loss": 6.0408,
|
|
"mean_token_accuracy": 0.1393430083990097,
|
|
"num_tokens": 3475336.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.099352216720581,
|
|
"epoch": 1.5824357912178955,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004925414287290979,
|
|
"loss": 6.0325,
|
|
"mean_token_accuracy": 0.1402113474905491,
|
|
"num_tokens": 3483963.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.013989686965942,
|
|
"epoch": 1.5865782932891466,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004924596132283075,
|
|
"loss": 5.9376,
|
|
"mean_token_accuracy": 0.14408371299505235,
|
|
"num_tokens": 3492947.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.973308038711548,
|
|
"epoch": 1.5907207953603977,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004923773590877135,
|
|
"loss": 6.029,
|
|
"mean_token_accuracy": 0.13019290938973427,
|
|
"num_tokens": 3501877.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.14404263496399,
|
|
"epoch": 1.5948632974316488,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004922946664732314,
|
|
"loss": 6.074,
|
|
"mean_token_accuracy": 0.12927663773298265,
|
|
"num_tokens": 3510610.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.131452226638794,
|
|
"epoch": 1.5990057995029,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004922115355516615,
|
|
"loss": 6.0328,
|
|
"mean_token_accuracy": 0.13440151512622833,
|
|
"num_tokens": 3519829.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.053449392318726,
|
|
"epoch": 1.6031483015741508,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004921279664906877,
|
|
"loss": 6.0203,
|
|
"mean_token_accuracy": 0.1330590568482876,
|
|
"num_tokens": 3529408.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.981090784072876,
|
|
"epoch": 1.6072908036454017,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004920439594588781,
|
|
"loss": 5.9579,
|
|
"mean_token_accuracy": 0.14276544600725175,
|
|
"num_tokens": 3538068.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.187561845779419,
|
|
"epoch": 1.6114333057166528,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000491959514625684,
|
|
"loss": 6.0732,
|
|
"mean_token_accuracy": 0.1400441735982895,
|
|
"num_tokens": 3547302.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.01104588508606,
|
|
"epoch": 1.6155758077879039,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00049187463216144,
|
|
"loss": 5.9397,
|
|
"mean_token_accuracy": 0.14568847864866258,
|
|
"num_tokens": 3556087.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.153070068359375,
|
|
"epoch": 1.619718309859155,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000491789312237363,
|
|
"loss": 6.0312,
|
|
"mean_token_accuracy": 0.13621370121836662,
|
|
"num_tokens": 3566235.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.994067668914795,
|
|
"epoch": 1.623860811930406,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004917035550255528,
|
|
"loss": 5.9618,
|
|
"mean_token_accuracy": 0.1392713963985443,
|
|
"num_tokens": 3574762.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.9850420475006105,
|
|
"epoch": 1.628003314001657,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000491617360698991,
|
|
"loss": 5.9424,
|
|
"mean_token_accuracy": 0.13408162221312522,
|
|
"num_tokens": 3583359.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.994701671600342,
|
|
"epoch": 1.6321458160729079,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004915307294315412,
|
|
"loss": 5.9009,
|
|
"mean_token_accuracy": 0.13760913535952568,
|
|
"num_tokens": 3591436.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.980083847045899,
|
|
"epoch": 1.636288318144159,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000491443661397948,
|
|
"loss": 5.9868,
|
|
"mean_token_accuracy": 0.1349792368710041,
|
|
"num_tokens": 3601054.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.052795314788819,
|
|
"epoch": 1.64043082021541,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004913561567738368,
|
|
"loss": 5.9674,
|
|
"mean_token_accuracy": 0.13878381252288818,
|
|
"num_tokens": 3610641.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.99411187171936,
|
|
"epoch": 1.6445733222866612,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004912682157357146,
|
|
"loss": 5.9041,
|
|
"mean_token_accuracy": 0.14029043316841125,
|
|
"num_tokens": 3619351.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.911650848388672,
|
|
"epoch": 1.6487158243579123,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004911798384609675,
|
|
"loss": 5.8879,
|
|
"mean_token_accuracy": 0.14449498876929284,
|
|
"num_tokens": 3627341.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.012162733078003,
|
|
"epoch": 1.6528583264291632,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004910910251278626,
|
|
"loss": 5.9452,
|
|
"mean_token_accuracy": 0.14296462759375572,
|
|
"num_tokens": 3637132.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.013031959533691,
|
|
"epoch": 1.6570008285004143,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004910017759155458,
|
|
"loss": 6.0021,
|
|
"mean_token_accuracy": 0.13557344153523446,
|
|
"num_tokens": 3646196.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6570008285004143,
|
|
"eval_entropy": 5.924376028178852,
|
|
"eval_loss": 6.150927543640137,
|
|
"eval_mean_token_accuracy": 0.13550341723443093,
|
|
"eval_num_tokens": 3646196.0,
|
|
"eval_runtime": 2.7516,
|
|
"eval_samples_per_second": 1361.773,
|
|
"eval_steps_per_second": 170.449,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.026213884353638,
|
|
"epoch": 1.6611433305716652,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004909120910040427,
|
|
"loss": 5.9388,
|
|
"mean_token_accuracy": 0.14352744221687316,
|
|
"num_tokens": 3654658.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.142762756347656,
|
|
"epoch": 1.6652858326429163,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004908219705742574,
|
|
"loss": 6.2148,
|
|
"mean_token_accuracy": 0.13257693275809287,
|
|
"num_tokens": 3664509.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.105361080169677,
|
|
"epoch": 1.6694283347141674,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004907314148079727,
|
|
"loss": 5.917,
|
|
"mean_token_accuracy": 0.14199795499444007,
|
|
"num_tokens": 3672983.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.827013301849365,
|
|
"epoch": 1.6735708367854185,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004906404238878496,
|
|
"loss": 5.8004,
|
|
"mean_token_accuracy": 0.14207299277186394,
|
|
"num_tokens": 3681196.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.070524263381958,
|
|
"epoch": 1.6777133388566694,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004905489979974265,
|
|
"loss": 5.9619,
|
|
"mean_token_accuracy": 0.13726090565323829,
|
|
"num_tokens": 3690077.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.060069894790649,
|
|
"epoch": 1.6818558409279205,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004904571373211197,
|
|
"loss": 6.0231,
|
|
"mean_token_accuracy": 0.13792718350887298,
|
|
"num_tokens": 3700121.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.907645797729492,
|
|
"epoch": 1.6859983429991714,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000490364842044222,
|
|
"loss": 5.8567,
|
|
"mean_token_accuracy": 0.14549633860588074,
|
|
"num_tokens": 3709618.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.038069868087769,
|
|
"epoch": 1.6901408450704225,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004902721123529031,
|
|
"loss": 6.0487,
|
|
"mean_token_accuracy": 0.12704455628991126,
|
|
"num_tokens": 3719570.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.154643487930298,
|
|
"epoch": 1.6942833471416736,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000490178948434209,
|
|
"loss": 5.9776,
|
|
"mean_token_accuracy": 0.14316998422145844,
|
|
"num_tokens": 3729354.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.893994522094727,
|
|
"epoch": 1.6984258492129247,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004900853504760614,
|
|
"loss": 5.8869,
|
|
"mean_token_accuracy": 0.14341987296938896,
|
|
"num_tokens": 3738453.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.962806272506714,
|
|
"epoch": 1.7025683512841756,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004899913186672577,
|
|
"loss": 5.8651,
|
|
"mean_token_accuracy": 0.14164601415395736,
|
|
"num_tokens": 3747815.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.914563226699829,
|
|
"epoch": 1.7067108533554267,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004898968531974702,
|
|
"loss": 5.9676,
|
|
"mean_token_accuracy": 0.14769200831651688,
|
|
"num_tokens": 3757298.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.056203603744507,
|
|
"epoch": 1.7108533554266776,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004898019542572462,
|
|
"loss": 5.9602,
|
|
"mean_token_accuracy": 0.13963632434606552,
|
|
"num_tokens": 3765939.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.100585651397705,
|
|
"epoch": 1.7149958574979287,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004897066220380073,
|
|
"loss": 5.995,
|
|
"mean_token_accuracy": 0.1360404372215271,
|
|
"num_tokens": 3775160.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.958986473083496,
|
|
"epoch": 1.7191383595691798,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004896108567320489,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.13895440623164176,
|
|
"num_tokens": 3785181.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.122873830795288,
|
|
"epoch": 1.723280861640431,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048951465853254,
|
|
"loss": 5.9952,
|
|
"mean_token_accuracy": 0.13867081850767135,
|
|
"num_tokens": 3794531.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.962896156311035,
|
|
"epoch": 1.7274233637116818,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004894180276335231,
|
|
"loss": 5.9768,
|
|
"mean_token_accuracy": 0.1366507463157177,
|
|
"num_tokens": 3804602.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.0686544418334964,
|
|
"epoch": 1.731565865782933,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004893209642299132,
|
|
"loss": 6.0243,
|
|
"mean_token_accuracy": 0.13472760021686553,
|
|
"num_tokens": 3813394.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.06077013015747,
|
|
"epoch": 1.7357083678541838,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004892234685174976,
|
|
"loss": 5.9941,
|
|
"mean_token_accuracy": 0.13705680817365645,
|
|
"num_tokens": 3821604.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.069777965545654,
|
|
"epoch": 1.739850869925435,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004891255406929359,
|
|
"loss": 5.995,
|
|
"mean_token_accuracy": 0.13641481548547746,
|
|
"num_tokens": 3831977.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.0059997081756595,
|
|
"epoch": 1.743993371996686,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004890271809537595,
|
|
"loss": 5.9453,
|
|
"mean_token_accuracy": 0.14286706522107123,
|
|
"num_tokens": 3840684.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.989671230316162,
|
|
"epoch": 1.7481358740679371,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004889283894983704,
|
|
"loss": 5.8682,
|
|
"mean_token_accuracy": 0.13856151327490807,
|
|
"num_tokens": 3851147.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.041515731811524,
|
|
"epoch": 1.752278376139188,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000488829166526042,
|
|
"loss": 6.0099,
|
|
"mean_token_accuracy": 0.1394909791648388,
|
|
"num_tokens": 3859881.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.0923104763031,
|
|
"epoch": 1.7564208782104391,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004887295122369176,
|
|
"loss": 6.0555,
|
|
"mean_token_accuracy": 0.13175006359815597,
|
|
"num_tokens": 3870151.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.974537181854248,
|
|
"epoch": 1.76056338028169,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000488629426832011,
|
|
"loss": 5.9759,
|
|
"mean_token_accuracy": 0.13857906460762023,
|
|
"num_tokens": 3878398.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.02483057975769,
|
|
"epoch": 1.7647058823529411,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004885289105132054,
|
|
"loss": 5.9227,
|
|
"mean_token_accuracy": 0.14162865728139878,
|
|
"num_tokens": 3888007.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.9927177906036375,
|
|
"epoch": 1.7688483844241922,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004884279634832532,
|
|
"loss": 5.9141,
|
|
"mean_token_accuracy": 0.14104892462491989,
|
|
"num_tokens": 3896945.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.011741352081299,
|
|
"epoch": 1.7729908864954433,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004883265859457754,
|
|
"loss": 5.9408,
|
|
"mean_token_accuracy": 0.14671368449926375,
|
|
"num_tokens": 3906049.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.927330160140992,
|
|
"epoch": 1.7771333885666944,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048822477810526186,
|
|
"loss": 5.8978,
|
|
"mean_token_accuracy": 0.137970220297575,
|
|
"num_tokens": 3915050.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.98696722984314,
|
|
"epoch": 1.7812758906379453,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048812254016706994,
|
|
"loss": 5.9854,
|
|
"mean_token_accuracy": 0.1412371039390564,
|
|
"num_tokens": 3924031.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.043074274063111,
|
|
"epoch": 1.7854183927091962,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004880198723374247,
|
|
"loss": 6.0205,
|
|
"mean_token_accuracy": 0.14472691193223,
|
|
"num_tokens": 3933756.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.06718225479126,
|
|
"epoch": 1.7895608947804473,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048791677482341845,
|
|
"loss": 6.0066,
|
|
"mean_token_accuracy": 0.13294869661331177,
|
|
"num_tokens": 3943129.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.958523988723755,
|
|
"epoch": 1.7937033968516984,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048781324783301016,
|
|
"loss": 5.8428,
|
|
"mean_token_accuracy": 0.1384035088121891,
|
|
"num_tokens": 3951494.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.979408073425293,
|
|
"epoch": 1.7978458989229495,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048770929157502505,
|
|
"loss": 6.014,
|
|
"mean_token_accuracy": 0.14348282739520074,
|
|
"num_tokens": 3960588.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.067771530151367,
|
|
"epoch": 1.8019884009942007,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004876049062591543,
|
|
"loss": 5.9762,
|
|
"mean_token_accuracy": 0.13583431839942933,
|
|
"num_tokens": 3970135.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.0059386730194095,
|
|
"epoch": 1.8061309030654515,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048750009209595443,
|
|
"loss": 5.9479,
|
|
"mean_token_accuracy": 0.13654519543051719,
|
|
"num_tokens": 3979569.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.87956805229187,
|
|
"epoch": 1.8102734051367024,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048739484929684714,
|
|
"loss": 5.91,
|
|
"mean_token_accuracy": 0.1421449676156044,
|
|
"num_tokens": 3987885.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.040490770339966,
|
|
"epoch": 1.8144159072079535,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048728917807411856,
|
|
"loss": 6.0586,
|
|
"mean_token_accuracy": 0.13321116119623183,
|
|
"num_tokens": 3997726.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.041323566436768,
|
|
"epoch": 1.8185584092792046,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004871830786409191,
|
|
"loss": 5.887,
|
|
"mean_token_accuracy": 0.14867799878120422,
|
|
"num_tokens": 4006437.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.968663835525513,
|
|
"epoch": 1.8227009113504558,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00048707655121126296,
|
|
"loss": 5.8754,
|
|
"mean_token_accuracy": 0.14638606756925582,
|
|
"num_tokens": 4015326.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.972012090682983,
|
|
"epoch": 1.8268434134217069,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048696959600002767,
|
|
"loss": 5.9139,
|
|
"mean_token_accuracy": 0.14176759198307992,
|
|
"num_tokens": 4024294.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.954474020004272,
|
|
"epoch": 1.8309859154929577,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048686221322295347,
|
|
"loss": 5.9376,
|
|
"mean_token_accuracy": 0.14578078538179398,
|
|
"num_tokens": 4033341.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.941730737686157,
|
|
"epoch": 1.8351284175642086,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048675440309664314,
|
|
"loss": 5.8816,
|
|
"mean_token_accuracy": 0.14992816746234894,
|
|
"num_tokens": 4042446.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.9512083530426025,
|
|
"epoch": 1.8392709196354597,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048664616583856166,
|
|
"loss": 5.9209,
|
|
"mean_token_accuracy": 0.13979735597968102,
|
|
"num_tokens": 4051849.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.053277587890625,
|
|
"epoch": 1.8434134217067109,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048653750166703516,
|
|
"loss": 6.0092,
|
|
"mean_token_accuracy": 0.14033474326133727,
|
|
"num_tokens": 4061075.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.005611562728882,
|
|
"epoch": 1.847555923777962,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004864284108012513,
|
|
"loss": 5.9982,
|
|
"mean_token_accuracy": 0.12854147851467132,
|
|
"num_tokens": 4071022.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.967685556411743,
|
|
"epoch": 1.851698425849213,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004863188934612583,
|
|
"loss": 5.9007,
|
|
"mean_token_accuracy": 0.14465969204902648,
|
|
"num_tokens": 4080728.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.987842464447022,
|
|
"epoch": 1.855840927920464,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004862089498679645,
|
|
"loss": 5.8568,
|
|
"mean_token_accuracy": 0.14544849768280982,
|
|
"num_tokens": 4088712.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.957550859451294,
|
|
"epoch": 1.8599834299917148,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004860985802431383,
|
|
"loss": 5.8859,
|
|
"mean_token_accuracy": 0.1428893633186817,
|
|
"num_tokens": 4097488.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.878279829025269,
|
|
"epoch": 1.864125932062966,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004859877848094071,
|
|
"loss": 5.8605,
|
|
"mean_token_accuracy": 0.14168896824121474,
|
|
"num_tokens": 4106485.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.049147462844848,
|
|
"epoch": 1.868268434134217,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004858765637902575,
|
|
"loss": 5.8386,
|
|
"mean_token_accuracy": 0.14742093831300734,
|
|
"num_tokens": 4115806.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.996465444564819,
|
|
"epoch": 1.8724109362054682,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004857649174100344,
|
|
"loss": 5.9829,
|
|
"mean_token_accuracy": 0.1403212234377861,
|
|
"num_tokens": 4124636.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.9591474533081055,
|
|
"epoch": 1.8765534382767193,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004856528458939409,
|
|
"loss": 5.9603,
|
|
"mean_token_accuracy": 0.13828378096222876,
|
|
"num_tokens": 4133876.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.940617370605469,
|
|
"epoch": 1.8806959403479702,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004855403494680373,
|
|
"loss": 5.8493,
|
|
"mean_token_accuracy": 0.14110578671097757,
|
|
"num_tokens": 4143555.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.966775751113891,
|
|
"epoch": 1.884838442419221,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048542742835924133,
|
|
"loss": 5.8261,
|
|
"mean_token_accuracy": 0.14103717133402824,
|
|
"num_tokens": 4153007.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.793911170959473,
|
|
"epoch": 1.8889809444904722,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048531408279532705,
|
|
"loss": 5.8588,
|
|
"mean_token_accuracy": 0.13940922021865845,
|
|
"num_tokens": 4162107.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.9467339515686035,
|
|
"epoch": 1.8931234465617233,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048520031300492507,
|
|
"loss": 5.8562,
|
|
"mean_token_accuracy": 0.1414430856704712,
|
|
"num_tokens": 4170920.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.99835844039917,
|
|
"epoch": 1.8972659486329744,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048508611921752124,
|
|
"loss": 5.8883,
|
|
"mean_token_accuracy": 0.14315918162465097,
|
|
"num_tokens": 4180226.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.902116441726685,
|
|
"epoch": 1.9014084507042255,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048497150166345703,
|
|
"loss": 5.8181,
|
|
"mean_token_accuracy": 0.1446967154741287,
|
|
"num_tokens": 4189283.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.875041151046753,
|
|
"epoch": 1.9055509527754764,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004848564605739287,
|
|
"loss": 5.8647,
|
|
"mean_token_accuracy": 0.14096341878175736,
|
|
"num_tokens": 4199017.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.04994044303894,
|
|
"epoch": 1.9096934548467275,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048474099618098644,
|
|
"loss": 5.9904,
|
|
"mean_token_accuracy": 0.13669014796614648,
|
|
"num_tokens": 4207756.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.923856973648071,
|
|
"epoch": 1.9138359569179784,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004846251087175348,
|
|
"loss": 5.9005,
|
|
"mean_token_accuracy": 0.1407478854060173,
|
|
"num_tokens": 4216721.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.897194051742554,
|
|
"epoch": 1.9179784589892295,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004845087984173312,
|
|
"loss": 5.8789,
|
|
"mean_token_accuracy": 0.14667138159275056,
|
|
"num_tokens": 4225596.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.997752904891968,
|
|
"epoch": 1.9221209610604806,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048439206551498644,
|
|
"loss": 5.9253,
|
|
"mean_token_accuracy": 0.14398268461227418,
|
|
"num_tokens": 4234500.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.8700896263122555,
|
|
"epoch": 1.9262634631317317,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004842749102459635,
|
|
"loss": 5.8781,
|
|
"mean_token_accuracy": 0.15120936185121536,
|
|
"num_tokens": 4243240.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.906232309341431,
|
|
"epoch": 1.9304059652029826,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048415733284657727,
|
|
"loss": 5.8516,
|
|
"mean_token_accuracy": 0.14731550291180612,
|
|
"num_tokens": 4251892.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.97393217086792,
|
|
"epoch": 1.9345484672742337,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004840393335539943,
|
|
"loss": 5.8253,
|
|
"mean_token_accuracy": 0.14521327316761018,
|
|
"num_tokens": 4261447.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.948915719985962,
|
|
"epoch": 1.9386909693454846,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048392091260623195,
|
|
"loss": 5.9664,
|
|
"mean_token_accuracy": 0.1478654459118843,
|
|
"num_tokens": 4270120.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.98240385055542,
|
|
"epoch": 1.9428334714167357,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048380207024215836,
|
|
"loss": 5.8421,
|
|
"mean_token_accuracy": 0.14822718501091003,
|
|
"num_tokens": 4278300.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.955031204223633,
|
|
"epoch": 1.9469759734879868,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048368280670149145,
|
|
"loss": 5.8912,
|
|
"mean_token_accuracy": 0.14697090536355972,
|
|
"num_tokens": 4286454.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.854554843902588,
|
|
"epoch": 1.951118475559238,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004835631222247989,
|
|
"loss": 5.7636,
|
|
"mean_token_accuracy": 0.14784407168626784,
|
|
"num_tokens": 4294752.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.909240818023681,
|
|
"epoch": 1.9552609776304888,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048344301705349735,
|
|
"loss": 5.8298,
|
|
"mean_token_accuracy": 0.14864546656608582,
|
|
"num_tokens": 4303195.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.995252275466919,
|
|
"epoch": 1.95940347970174,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048332249142985195,
|
|
"loss": 5.9603,
|
|
"mean_token_accuracy": 0.13669342547655106,
|
|
"num_tokens": 4311961.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.914461421966553,
|
|
"epoch": 1.9635459817729908,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004832015455969763,
|
|
"loss": 5.8152,
|
|
"mean_token_accuracy": 0.13769099935889245,
|
|
"num_tokens": 4322267.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.856181859970093,
|
|
"epoch": 1.9676884838442419,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048308017979883114,
|
|
"loss": 5.939,
|
|
"mean_token_accuracy": 0.14319972842931747,
|
|
"num_tokens": 4330584.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.907060337066651,
|
|
"epoch": 1.971830985915493,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048295839428022454,
|
|
"loss": 5.7562,
|
|
"mean_token_accuracy": 0.14656164199113847,
|
|
"num_tokens": 4339215.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.901394891738891,
|
|
"epoch": 1.975973487986744,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048283618928681127,
|
|
"loss": 5.7878,
|
|
"mean_token_accuracy": 0.1504148483276367,
|
|
"num_tokens": 4347711.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.778523015975952,
|
|
"epoch": 1.980115990057995,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004827135650650923,
|
|
"loss": 5.7814,
|
|
"mean_token_accuracy": 0.1501213528215885,
|
|
"num_tokens": 4356982.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.905798053741455,
|
|
"epoch": 1.984258492129246,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004825905218624139,
|
|
"loss": 5.8019,
|
|
"mean_token_accuracy": 0.148068468272686,
|
|
"num_tokens": 4365378.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.008777904510498,
|
|
"epoch": 1.988400994200497,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048246705992696756,
|
|
"loss": 5.8635,
|
|
"mean_token_accuracy": 0.1390143997967243,
|
|
"num_tokens": 4374007.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.05829381942749,
|
|
"epoch": 1.992543496271748,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048234317950778975,
|
|
"loss": 5.9666,
|
|
"mean_token_accuracy": 0.13901185542345046,
|
|
"num_tokens": 4383909.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.023723077774048,
|
|
"epoch": 1.9966859983429992,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004822188808547607,
|
|
"loss": 6.0501,
|
|
"mean_token_accuracy": 0.13643686026334761,
|
|
"num_tokens": 4393145.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.890930223464966,
|
|
"epoch": 2.0008285004142503,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004820941642186043,
|
|
"loss": 5.7449,
|
|
"mean_token_accuracy": 0.14210715517401695,
|
|
"num_tokens": 4402024.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.920894384384155,
|
|
"epoch": 2.0049710024855014,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004819690298508879,
|
|
"loss": 5.5832,
|
|
"mean_token_accuracy": 0.15544317960739135,
|
|
"num_tokens": 4411377.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.783400630950927,
|
|
"epoch": 2.009113504556752,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048184347800402083,
|
|
"loss": 5.4847,
|
|
"mean_token_accuracy": 0.1676163673400879,
|
|
"num_tokens": 4419781.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.719689798355103,
|
|
"epoch": 2.013256006628003,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004817175089312551,
|
|
"loss": 5.4593,
|
|
"mean_token_accuracy": 0.1607228234410286,
|
|
"num_tokens": 4428427.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.824680519104004,
|
|
"epoch": 2.0173985086992543,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048159112288668403,
|
|
"loss": 5.4979,
|
|
"mean_token_accuracy": 0.15756488591432571,
|
|
"num_tokens": 4438156.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.819461822509766,
|
|
"epoch": 2.0215410107705054,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048146432012524215,
|
|
"loss": 5.532,
|
|
"mean_token_accuracy": 0.16391444951295853,
|
|
"num_tokens": 4446342.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.761379432678223,
|
|
"epoch": 2.0256835128417565,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048133710090270446,
|
|
"loss": 5.5658,
|
|
"mean_token_accuracy": 0.15113780349493028,
|
|
"num_tokens": 4455582.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.888222980499267,
|
|
"epoch": 2.0298260149130076,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048120946547568596,
|
|
"loss": 5.5826,
|
|
"mean_token_accuracy": 0.15940120220184326,
|
|
"num_tokens": 4464518.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.927597761154175,
|
|
"epoch": 2.0339685169842583,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048108141410164133,
|
|
"loss": 5.7022,
|
|
"mean_token_accuracy": 0.1438803918659687,
|
|
"num_tokens": 4475031.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.8743188858032225,
|
|
"epoch": 2.0381110190555094,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004809529470388642,
|
|
"loss": 5.5708,
|
|
"mean_token_accuracy": 0.1615961655974388,
|
|
"num_tokens": 4484609.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.786350297927856,
|
|
"epoch": 2.0422535211267605,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004808240645464866,
|
|
"loss": 5.6056,
|
|
"mean_token_accuracy": 0.15688657909631729,
|
|
"num_tokens": 4494238.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.79974250793457,
|
|
"epoch": 2.0463960231980116,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004806947668844787,
|
|
"loss": 5.4662,
|
|
"mean_token_accuracy": 0.16647584587335587,
|
|
"num_tokens": 4503258.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.838503170013428,
|
|
"epoch": 2.0505385252692627,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048056505431364794,
|
|
"loss": 5.5489,
|
|
"mean_token_accuracy": 0.15769330710172652,
|
|
"num_tokens": 4512071.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.770729064941406,
|
|
"epoch": 2.054681027340514,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048043492709563877,
|
|
"loss": 5.3786,
|
|
"mean_token_accuracy": 0.17338205873966217,
|
|
"num_tokens": 4520219.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.736357259750366,
|
|
"epoch": 2.0588235294117645,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048030438549293193,
|
|
"loss": 5.6183,
|
|
"mean_token_accuracy": 0.15568529814481735,
|
|
"num_tokens": 4529386.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.889974308013916,
|
|
"epoch": 2.0629660314830156,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048017342976884424,
|
|
"loss": 5.6105,
|
|
"mean_token_accuracy": 0.14977954104542732,
|
|
"num_tokens": 4538860.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.706206464767456,
|
|
"epoch": 2.0671085335542667,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048004206018752773,
|
|
"loss": 5.5646,
|
|
"mean_token_accuracy": 0.15797781199216843,
|
|
"num_tokens": 4547842.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.816332674026489,
|
|
"epoch": 2.071251035625518,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047991027701396895,
|
|
"loss": 5.5431,
|
|
"mean_token_accuracy": 0.15634830445051193,
|
|
"num_tokens": 4556238.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.071251035625518,
|
|
"eval_entropy": 5.749874321382437,
|
|
"eval_loss": 6.056258201599121,
|
|
"eval_mean_token_accuracy": 0.14146501896605054,
|
|
"eval_num_tokens": 4556238.0,
|
|
"eval_runtime": 2.7516,
|
|
"eval_samples_per_second": 1361.734,
|
|
"eval_steps_per_second": 170.444,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.786322116851807,
|
|
"epoch": 2.075393537696769,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004797780805139892,
|
|
"loss": 5.5554,
|
|
"mean_token_accuracy": 0.16125964000821114,
|
|
"num_tokens": 4565684.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.756158685684204,
|
|
"epoch": 2.07953603976802,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004796454709542432,
|
|
"loss": 5.5027,
|
|
"mean_token_accuracy": 0.1608697935938835,
|
|
"num_tokens": 4575001.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.776489973068237,
|
|
"epoch": 2.083678541839271,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004795124486022189,
|
|
"loss": 5.6243,
|
|
"mean_token_accuracy": 0.15420307517051696,
|
|
"num_tokens": 4584128.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.726174306869507,
|
|
"epoch": 2.087821043910522,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047937901372623694,
|
|
"loss": 5.4739,
|
|
"mean_token_accuracy": 0.16476461738348008,
|
|
"num_tokens": 4593011.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.785531711578369,
|
|
"epoch": 2.091963545981773,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047924516659545005,
|
|
"loss": 5.5068,
|
|
"mean_token_accuracy": 0.1623775690793991,
|
|
"num_tokens": 4602676.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.84798231124878,
|
|
"epoch": 2.096106048053024,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004791109074798425,
|
|
"loss": 5.5896,
|
|
"mean_token_accuracy": 0.15097921043634416,
|
|
"num_tokens": 4611915.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.7544732093811035,
|
|
"epoch": 2.100248550124275,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004789762366502296,
|
|
"loss": 5.5609,
|
|
"mean_token_accuracy": 0.15663281530141832,
|
|
"num_tokens": 4621396.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.795808458328247,
|
|
"epoch": 2.1043910521955262,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004788411543782571,
|
|
"loss": 5.5689,
|
|
"mean_token_accuracy": 0.15645280182361604,
|
|
"num_tokens": 4630533.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.74350414276123,
|
|
"epoch": 2.108533554266777,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004787056609364008,
|
|
"loss": 5.4951,
|
|
"mean_token_accuracy": 0.1582939773797989,
|
|
"num_tokens": 4639023.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.719309282302857,
|
|
"epoch": 2.112676056338028,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004785697565979657,
|
|
"loss": 5.5339,
|
|
"mean_token_accuracy": 0.16151303052902222,
|
|
"num_tokens": 4647975.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.737587833404541,
|
|
"epoch": 2.116818558409279,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004784334416370857,
|
|
"loss": 5.5491,
|
|
"mean_token_accuracy": 0.15539203137159346,
|
|
"num_tokens": 4657555.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.7617528438568115,
|
|
"epoch": 2.1209610604805302,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004782967163287231,
|
|
"loss": 5.4565,
|
|
"mean_token_accuracy": 0.15736441761255265,
|
|
"num_tokens": 4667228.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.704279279708862,
|
|
"epoch": 2.1251035625517813,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047815958094866756,
|
|
"loss": 5.5323,
|
|
"mean_token_accuracy": 0.15277542173862457,
|
|
"num_tokens": 4676723.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.795382833480835,
|
|
"epoch": 2.1292460646230325,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004780220357735363,
|
|
"loss": 5.5207,
|
|
"mean_token_accuracy": 0.1547121912240982,
|
|
"num_tokens": 4686241.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.7523492813110355,
|
|
"epoch": 2.1333885666942836,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047788408108077294,
|
|
"loss": 5.5718,
|
|
"mean_token_accuracy": 0.15764677077531813,
|
|
"num_tokens": 4695195.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.813769340515137,
|
|
"epoch": 2.1375310687655342,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004777457171486471,
|
|
"loss": 5.63,
|
|
"mean_token_accuracy": 0.15644090324640275,
|
|
"num_tokens": 4704716.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.719523763656616,
|
|
"epoch": 2.1416735708367853,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004776069442562541,
|
|
"loss": 5.5895,
|
|
"mean_token_accuracy": 0.15595096796751023,
|
|
"num_tokens": 4713720.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.820378255844116,
|
|
"epoch": 2.1458160729080364,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004774677626835138,
|
|
"loss": 5.5662,
|
|
"mean_token_accuracy": 0.15348037481307983,
|
|
"num_tokens": 4722428.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.800876331329346,
|
|
"epoch": 2.1499585749792876,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047732817271117084,
|
|
"loss": 5.5822,
|
|
"mean_token_accuracy": 0.15440100207924842,
|
|
"num_tokens": 4732395.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.759647798538208,
|
|
"epoch": 2.1541010770505387,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004771881746207934,
|
|
"loss": 5.5637,
|
|
"mean_token_accuracy": 0.15513806194067,
|
|
"num_tokens": 4741071.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.729876518249512,
|
|
"epoch": 2.1582435791217898,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047704776869477285,
|
|
"loss": 5.5579,
|
|
"mean_token_accuracy": 0.15268941819667817,
|
|
"num_tokens": 4749876.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.7958738803863525,
|
|
"epoch": 2.1623860811930404,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047690695521632344,
|
|
"loss": 5.5113,
|
|
"mean_token_accuracy": 0.15810582637786866,
|
|
"num_tokens": 4758721.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.697278213500977,
|
|
"epoch": 2.1665285832642915,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047676573446948114,
|
|
"loss": 5.5424,
|
|
"mean_token_accuracy": 0.16137579083442688,
|
|
"num_tokens": 4767162.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.794365882873535,
|
|
"epoch": 2.1706710853355426,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047662410673910376,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.16126850992441177,
|
|
"num_tokens": 4776397.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.6881626605987545,
|
|
"epoch": 2.1748135874067938,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047648207231087,
|
|
"loss": 5.507,
|
|
"mean_token_accuracy": 0.1613411232829094,
|
|
"num_tokens": 4784822.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.762795734405517,
|
|
"epoch": 2.178956089478045,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004763396314712786,
|
|
"loss": 5.573,
|
|
"mean_token_accuracy": 0.16014576703310013,
|
|
"num_tokens": 4794272.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.7245035648345945,
|
|
"epoch": 2.183098591549296,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004761967845076484,
|
|
"loss": 5.5137,
|
|
"mean_token_accuracy": 0.15319861993193626,
|
|
"num_tokens": 4804314.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.677399730682373,
|
|
"epoch": 2.1872410936205466,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004760535317081174,
|
|
"loss": 5.519,
|
|
"mean_token_accuracy": 0.15785249769687654,
|
|
"num_tokens": 4812685.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.780713081359863,
|
|
"epoch": 2.1913835956917977,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047590987336164207,
|
|
"loss": 5.59,
|
|
"mean_token_accuracy": 0.15907835513353347,
|
|
"num_tokens": 4821761.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.792177724838257,
|
|
"epoch": 2.195526097763049,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047576580975799705,
|
|
"loss": 5.5878,
|
|
"mean_token_accuracy": 0.1585995614528656,
|
|
"num_tokens": 4830515.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.790842866897583,
|
|
"epoch": 2.1996685998343,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004756213411877743,
|
|
"loss": 5.5382,
|
|
"mean_token_accuracy": 0.15319958925247193,
|
|
"num_tokens": 4840741.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.652156352996826,
|
|
"epoch": 2.203811101905551,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047547646794238277,
|
|
"loss": 5.4936,
|
|
"mean_token_accuracy": 0.16105724722146988,
|
|
"num_tokens": 4849571.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.680575323104859,
|
|
"epoch": 2.207953603976802,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047533119031404757,
|
|
"loss": 5.4472,
|
|
"mean_token_accuracy": 0.16898063719272613,
|
|
"num_tokens": 4857194.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.678529882431031,
|
|
"epoch": 2.212096106048053,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047518550859580966,
|
|
"loss": 5.539,
|
|
"mean_token_accuracy": 0.15916522443294526,
|
|
"num_tokens": 4865894.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.780280065536499,
|
|
"epoch": 2.216238608119304,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004750394230815249,
|
|
"loss": 5.5643,
|
|
"mean_token_accuracy": 0.15810834765434265,
|
|
"num_tokens": 4875131.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.654790019989013,
|
|
"epoch": 2.220381110190555,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047489293406586375,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.1665143996477127,
|
|
"num_tokens": 4884098.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.683253002166748,
|
|
"epoch": 2.224523612261806,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004747460418443106,
|
|
"loss": 5.4703,
|
|
"mean_token_accuracy": 0.15905565470457078,
|
|
"num_tokens": 4892963.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.701637554168701,
|
|
"epoch": 2.2286661143330573,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047459874671316305,
|
|
"loss": 5.5293,
|
|
"mean_token_accuracy": 0.15367296785116197,
|
|
"num_tokens": 4902097.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.647844886779785,
|
|
"epoch": 2.2328086164043084,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047445104896953153,
|
|
"loss": 5.4472,
|
|
"mean_token_accuracy": 0.16798048615455627,
|
|
"num_tokens": 4910469.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.876229000091553,
|
|
"epoch": 2.236951118475559,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047430294891133846,
|
|
"loss": 5.7309,
|
|
"mean_token_accuracy": 0.1496710143983364,
|
|
"num_tokens": 4921024.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.72487506866455,
|
|
"epoch": 2.24109362054681,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004741544468373179,
|
|
"loss": 5.5663,
|
|
"mean_token_accuracy": 0.15733755230903626,
|
|
"num_tokens": 4930297.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.729690361022949,
|
|
"epoch": 2.2452361226180613,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047400554304701474,
|
|
"loss": 5.53,
|
|
"mean_token_accuracy": 0.15848677456378937,
|
|
"num_tokens": 4938971.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.639805555343628,
|
|
"epoch": 2.2493786246893124,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047385623784078427,
|
|
"loss": 5.5279,
|
|
"mean_token_accuracy": 0.15552860796451567,
|
|
"num_tokens": 4948531.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.787559366226196,
|
|
"epoch": 2.2535211267605635,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004737065315197913,
|
|
"loss": 5.6031,
|
|
"mean_token_accuracy": 0.16300928741693496,
|
|
"num_tokens": 4957430.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.746896505355835,
|
|
"epoch": 2.2576636288318146,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047355642438600987,
|
|
"loss": 5.5968,
|
|
"mean_token_accuracy": 0.15746028870344161,
|
|
"num_tokens": 4966681.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.689626312255859,
|
|
"epoch": 2.2618061309030653,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004734059167422225,
|
|
"loss": 5.4557,
|
|
"mean_token_accuracy": 0.1679423928260803,
|
|
"num_tokens": 4975327.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.666088104248047,
|
|
"epoch": 2.2659486329743164,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047325500889201956,
|
|
"loss": 5.5233,
|
|
"mean_token_accuracy": 0.16318378522992133,
|
|
"num_tokens": 4983741.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.736676263809204,
|
|
"epoch": 2.2700911350455675,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004731037011397986,
|
|
"loss": 5.6099,
|
|
"mean_token_accuracy": 0.15960921347141266,
|
|
"num_tokens": 4993171.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.83419246673584,
|
|
"epoch": 2.2742336371168186,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047295199379076403,
|
|
"loss": 5.639,
|
|
"mean_token_accuracy": 0.151589272916317,
|
|
"num_tokens": 5002810.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.685939884185791,
|
|
"epoch": 2.2783761391880697,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000472799887150926,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.15614852607250213,
|
|
"num_tokens": 5011875.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.7269433498382565,
|
|
"epoch": 2.282518641259321,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004726473815271003,
|
|
"loss": 5.5689,
|
|
"mean_token_accuracy": 0.16298677772283554,
|
|
"num_tokens": 5020780.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.73256287574768,
|
|
"epoch": 2.286661143330572,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047249447722690747,
|
|
"loss": 5.4877,
|
|
"mean_token_accuracy": 0.158497454226017,
|
|
"num_tokens": 5029896.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.687757587432861,
|
|
"epoch": 2.2908036454018226,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047234117455877215,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.15659505352377892,
|
|
"num_tokens": 5038956.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.689195823669434,
|
|
"epoch": 2.2949461474730737,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004721874738319225,
|
|
"loss": 5.4933,
|
|
"mean_token_accuracy": 0.16915134489536285,
|
|
"num_tokens": 5047005.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.674070739746094,
|
|
"epoch": 2.299088649544325,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004720333753563898,
|
|
"loss": 5.5645,
|
|
"mean_token_accuracy": 0.15010802298784257,
|
|
"num_tokens": 5056475.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.762642955780029,
|
|
"epoch": 2.303231151615576,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047187887944300747,
|
|
"loss": 5.5326,
|
|
"mean_token_accuracy": 0.16023070067167283,
|
|
"num_tokens": 5064978.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.707161235809326,
|
|
"epoch": 2.307373653686827,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004717239864034107,
|
|
"loss": 5.6213,
|
|
"mean_token_accuracy": 0.15539530366659166,
|
|
"num_tokens": 5074235.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.734616947174072,
|
|
"epoch": 2.3115161557580777,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047156869655003554,
|
|
"loss": 5.4672,
|
|
"mean_token_accuracy": 0.16697436273097993,
|
|
"num_tokens": 5082386.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.639080142974853,
|
|
"epoch": 2.315658657829329,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047141301019611875,
|
|
"loss": 5.4486,
|
|
"mean_token_accuracy": 0.1636665090918541,
|
|
"num_tokens": 5090634.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.8338916301727295,
|
|
"epoch": 2.31980115990058,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004712569276556967,
|
|
"loss": 5.6254,
|
|
"mean_token_accuracy": 0.15228739976882935,
|
|
"num_tokens": 5099003.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.685038995742798,
|
|
"epoch": 2.323943661971831,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047110044924360484,
|
|
"loss": 5.7285,
|
|
"mean_token_accuracy": 0.1518704243004322,
|
|
"num_tokens": 5108631.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.7691676139831545,
|
|
"epoch": 2.328086164043082,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047094357527547725,
|
|
"loss": 5.492,
|
|
"mean_token_accuracy": 0.15908450484275818,
|
|
"num_tokens": 5117309.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.748100280761719,
|
|
"epoch": 2.332228666114333,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004707863060677459,
|
|
"loss": 5.5936,
|
|
"mean_token_accuracy": 0.1596307374536991,
|
|
"num_tokens": 5126305.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.909375190734863,
|
|
"epoch": 2.3363711681855843,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047062864193764,
|
|
"loss": 5.6575,
|
|
"mean_token_accuracy": 0.15526335835456848,
|
|
"num_tokens": 5135338.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.662262630462647,
|
|
"epoch": 2.340513670256835,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047047058320318525,
|
|
"loss": 5.5526,
|
|
"mean_token_accuracy": 0.16301657408475875,
|
|
"num_tokens": 5144518.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.736520290374756,
|
|
"epoch": 2.344656172328086,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004703121301832034,
|
|
"loss": 5.6585,
|
|
"mean_token_accuracy": 0.15638706237077712,
|
|
"num_tokens": 5154102.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.77762246131897,
|
|
"epoch": 2.348798674399337,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047015328319731153,
|
|
"loss": 5.5953,
|
|
"mean_token_accuracy": 0.15536246746778487,
|
|
"num_tokens": 5163719.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.71826581954956,
|
|
"epoch": 2.3529411764705883,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004699940425659215,
|
|
"loss": 5.5533,
|
|
"mean_token_accuracy": 0.1597944587469101,
|
|
"num_tokens": 5173192.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.660325384140014,
|
|
"epoch": 2.3570836785418394,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046983440861023877,
|
|
"loss": 5.5111,
|
|
"mean_token_accuracy": 0.16499586999416352,
|
|
"num_tokens": 5182438.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.761060523986816,
|
|
"epoch": 2.36122618061309,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046967438165226275,
|
|
"loss": 5.6253,
|
|
"mean_token_accuracy": 0.15501290336251258,
|
|
"num_tokens": 5191595.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.683302545547486,
|
|
"epoch": 2.365368682684341,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004695139620147851,
|
|
"loss": 5.5101,
|
|
"mean_token_accuracy": 0.16470725685358048,
|
|
"num_tokens": 5200174.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.7557755470275875,
|
|
"epoch": 2.3695111847555923,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046935315002138994,
|
|
"loss": 5.6179,
|
|
"mean_token_accuracy": 0.15405628085136414,
|
|
"num_tokens": 5210454.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.797947883605957,
|
|
"epoch": 2.3736536868268434,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004691919459964525,
|
|
"loss": 5.5461,
|
|
"mean_token_accuracy": 0.15969752371311188,
|
|
"num_tokens": 5219243.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.599535417556763,
|
|
"epoch": 2.3777961888980945,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046903035026513894,
|
|
"loss": 5.4581,
|
|
"mean_token_accuracy": 0.16127343028783797,
|
|
"num_tokens": 5228656.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.649978446960449,
|
|
"epoch": 2.3819386909693456,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004688683631534056,
|
|
"loss": 5.5285,
|
|
"mean_token_accuracy": 0.15923455357551575,
|
|
"num_tokens": 5237807.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.794434547424316,
|
|
"epoch": 2.3860811930405967,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004687059849879981,
|
|
"loss": 5.629,
|
|
"mean_token_accuracy": 0.15792540162801744,
|
|
"num_tokens": 5247432.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.662710428237915,
|
|
"epoch": 2.3902236951118474,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046854321609645093,
|
|
"loss": 5.4666,
|
|
"mean_token_accuracy": 0.15640246868133545,
|
|
"num_tokens": 5256680.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.708359909057617,
|
|
"epoch": 2.3943661971830985,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 0.00046838005680708677,
|
|
"loss": 5.5622,
|
|
"mean_token_accuracy": 0.14960763677954675,
|
|
"num_tokens": 5266793.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.644489765167236,
|
|
"epoch": 2.3985086992543496,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046821650744901573,
|
|
"loss": 5.4844,
|
|
"mean_token_accuracy": 0.16081234514713288,
|
|
"num_tokens": 5275028.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.687662172317505,
|
|
"epoch": 2.4026512013256007,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004680525683521347,
|
|
"loss": 5.5543,
|
|
"mean_token_accuracy": 0.15716781169176103,
|
|
"num_tokens": 5284162.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.802298831939697,
|
|
"epoch": 2.406793703396852,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004678882398471269,
|
|
"loss": 5.6351,
|
|
"mean_token_accuracy": 0.14897942692041397,
|
|
"num_tokens": 5293263.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.7214367389678955,
|
|
"epoch": 2.4109362054681025,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046772352226546066,
|
|
"loss": 5.5221,
|
|
"mean_token_accuracy": 0.16579788774251938,
|
|
"num_tokens": 5301373.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.654629802703857,
|
|
"epoch": 2.4150787075393536,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004675584159393895,
|
|
"loss": 5.5472,
|
|
"mean_token_accuracy": 0.15706405341625213,
|
|
"num_tokens": 5311009.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.7292389392852785,
|
|
"epoch": 2.4192212096106047,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046739292120195083,
|
|
"loss": 5.553,
|
|
"mean_token_accuracy": 0.15957032442092894,
|
|
"num_tokens": 5320341.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.703303718566895,
|
|
"epoch": 2.423363711681856,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004672270383869657,
|
|
"loss": 5.525,
|
|
"mean_token_accuracy": 0.16170006692409516,
|
|
"num_tokens": 5329250.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.660790014266968,
|
|
"epoch": 2.427506213753107,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004670607678290379,
|
|
"loss": 5.486,
|
|
"mean_token_accuracy": 0.16461261510848998,
|
|
"num_tokens": 5336515.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.666818332672119,
|
|
"epoch": 2.431648715824358,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004668941098635532,
|
|
"loss": 5.5664,
|
|
"mean_token_accuracy": 0.16323853135108948,
|
|
"num_tokens": 5346043.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.721542930603027,
|
|
"epoch": 2.435791217895609,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046672706482667897,
|
|
"loss": 5.5287,
|
|
"mean_token_accuracy": 0.1586231455206871,
|
|
"num_tokens": 5355124.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.67028284072876,
|
|
"epoch": 2.43993371996686,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004665596330553633,
|
|
"loss": 5.5369,
|
|
"mean_token_accuracy": 0.15888493806123732,
|
|
"num_tokens": 5364556.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.712566614151001,
|
|
"epoch": 2.444076222038111,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046639181488733445,
|
|
"loss": 5.5755,
|
|
"mean_token_accuracy": 0.15751136243343353,
|
|
"num_tokens": 5373778.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.770811271667481,
|
|
"epoch": 2.448218724109362,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004662236106610999,
|
|
"loss": 5.5566,
|
|
"mean_token_accuracy": 0.1569676160812378,
|
|
"num_tokens": 5382633.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.576747179031372,
|
|
"epoch": 2.452361226180613,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046605502071594604,
|
|
"loss": 5.4624,
|
|
"mean_token_accuracy": 0.16361840963363647,
|
|
"num_tokens": 5391442.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.590614557266235,
|
|
"epoch": 2.4565037282518642,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004658860453919371,
|
|
"loss": 5.4619,
|
|
"mean_token_accuracy": 0.16674424558877946,
|
|
"num_tokens": 5400504.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.691436147689819,
|
|
"epoch": 2.460646230323115,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004657166850299149,
|
|
"loss": 5.4869,
|
|
"mean_token_accuracy": 0.16380043774843217,
|
|
"num_tokens": 5408939.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.66795802116394,
|
|
"epoch": 2.464788732394366,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004655469399714978,
|
|
"loss": 5.5004,
|
|
"mean_token_accuracy": 0.16163172274827958,
|
|
"num_tokens": 5418934.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.624106788635254,
|
|
"epoch": 2.468931234465617,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004653768105590799,
|
|
"loss": 5.5379,
|
|
"mean_token_accuracy": 0.16006896495819092,
|
|
"num_tokens": 5427894.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.696897411346436,
|
|
"epoch": 2.4730737365368682,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046520629713583115,
|
|
"loss": 5.5192,
|
|
"mean_token_accuracy": 0.16015419960021973,
|
|
"num_tokens": 5436830.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.637984657287598,
|
|
"epoch": 2.4772162386081193,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046503540004569555,
|
|
"loss": 5.5262,
|
|
"mean_token_accuracy": 0.16234136521816253,
|
|
"num_tokens": 5445504.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.732030582427979,
|
|
"epoch": 2.4813587406793705,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046486411963339125,
|
|
"loss": 5.5659,
|
|
"mean_token_accuracy": 0.15476318076252937,
|
|
"num_tokens": 5455378.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.726173019409179,
|
|
"epoch": 2.4855012427506216,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046469245624440955,
|
|
"loss": 5.6408,
|
|
"mean_token_accuracy": 0.1584235519170761,
|
|
"num_tokens": 5465334.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.4855012427506216,
|
|
"eval_entropy": 5.639674754030923,
|
|
"eval_loss": 5.9833292961120605,
|
|
"eval_mean_token_accuracy": 0.1469943482738568,
|
|
"eval_num_tokens": 5465334.0,
|
|
"eval_runtime": 2.967,
|
|
"eval_samples_per_second": 1262.891,
|
|
"eval_steps_per_second": 158.072,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.791637706756592,
|
|
"epoch": 2.4896437448218722,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046452041022501423,
|
|
"loss": 5.6258,
|
|
"mean_token_accuracy": 0.15460191220045089,
|
|
"num_tokens": 5475701.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.727213287353516,
|
|
"epoch": 2.4937862468931233,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004643479819222411,
|
|
"loss": 5.5093,
|
|
"mean_token_accuracy": 0.16252206414937972,
|
|
"num_tokens": 5485244.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.609255790710449,
|
|
"epoch": 2.4979287489643744,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046417517168389664,
|
|
"loss": 5.4792,
|
|
"mean_token_accuracy": 0.16647332459688186,
|
|
"num_tokens": 5493587.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.652401065826416,
|
|
"epoch": 2.5020712510356256,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004640019798585581,
|
|
"loss": 5.6175,
|
|
"mean_token_accuracy": 0.1556309536099434,
|
|
"num_tokens": 5503117.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.7778074741363525,
|
|
"epoch": 2.5062137531068767,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046382840679557233,
|
|
"loss": 5.5637,
|
|
"mean_token_accuracy": 0.16106179654598235,
|
|
"num_tokens": 5512742.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.636971282958984,
|
|
"epoch": 2.5103562551781273,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046365445284505516,
|
|
"loss": 5.5246,
|
|
"mean_token_accuracy": 0.15782141983509063,
|
|
"num_tokens": 5521862.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.651063251495361,
|
|
"epoch": 2.5144987572493784,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004634801183578907,
|
|
"loss": 5.5415,
|
|
"mean_token_accuracy": 0.1628969594836235,
|
|
"num_tokens": 5531244.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.7396568775177,
|
|
"epoch": 2.5186412593206295,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046330540368573063,
|
|
"loss": 5.5713,
|
|
"mean_token_accuracy": 0.15206096023321153,
|
|
"num_tokens": 5540786.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.720060348510742,
|
|
"epoch": 2.5227837613918807,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046313030918099365,
|
|
"loss": 5.5721,
|
|
"mean_token_accuracy": 0.1614146649837494,
|
|
"num_tokens": 5549961.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.705565547943115,
|
|
"epoch": 2.5269262634631318,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046295483519686437,
|
|
"loss": 5.5061,
|
|
"mean_token_accuracy": 0.16351179033517838,
|
|
"num_tokens": 5558941.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.665171957015991,
|
|
"epoch": 2.531068765534383,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046277898208729306,
|
|
"loss": 5.5118,
|
|
"mean_token_accuracy": 0.16655909419059753,
|
|
"num_tokens": 5567630.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.6335536479949955,
|
|
"epoch": 2.535211267605634,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004626027502069947,
|
|
"loss": 5.4747,
|
|
"mean_token_accuracy": 0.16044125258922576,
|
|
"num_tokens": 5576430.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.707994222640991,
|
|
"epoch": 2.539353769676885,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046242613991144814,
|
|
"loss": 5.5872,
|
|
"mean_token_accuracy": 0.16125080436468125,
|
|
"num_tokens": 5584785.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.570116853713989,
|
|
"epoch": 2.5434962717481358,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046224915155689586,
|
|
"loss": 5.5074,
|
|
"mean_token_accuracy": 0.1572348579764366,
|
|
"num_tokens": 5594583.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.726114130020141,
|
|
"epoch": 2.547638773819387,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046207178550034244,
|
|
"loss": 5.4746,
|
|
"mean_token_accuracy": 0.162801356613636,
|
|
"num_tokens": 5603498.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.751227521896363,
|
|
"epoch": 2.551781275890638,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004618940420995547,
|
|
"loss": 5.6074,
|
|
"mean_token_accuracy": 0.15416808128356935,
|
|
"num_tokens": 5613101.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.66825041770935,
|
|
"epoch": 2.555923777961889,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004617159217130606,
|
|
"loss": 5.585,
|
|
"mean_token_accuracy": 0.15699986293911933,
|
|
"num_tokens": 5621821.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.771618795394898,
|
|
"epoch": 2.5600662800331397,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004615374247001483,
|
|
"loss": 5.5981,
|
|
"mean_token_accuracy": 0.15983579754829408,
|
|
"num_tokens": 5631006.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.717100048065186,
|
|
"epoch": 2.564208782104391,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004613585514208659,
|
|
"loss": 5.5649,
|
|
"mean_token_accuracy": 0.15318279713392258,
|
|
"num_tokens": 5640159.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.654997730255127,
|
|
"epoch": 2.568351284175642,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046117930223602016,
|
|
"loss": 5.4329,
|
|
"mean_token_accuracy": 0.16797446459531784,
|
|
"num_tokens": 5648369.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.580140495300293,
|
|
"epoch": 2.572493786246893,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004609996775071765,
|
|
"loss": 5.4869,
|
|
"mean_token_accuracy": 0.15920215398073195,
|
|
"num_tokens": 5657669.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.672556781768799,
|
|
"epoch": 2.576636288318144,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004608196775966575,
|
|
"loss": 5.5745,
|
|
"mean_token_accuracy": 0.16189952045679093,
|
|
"num_tokens": 5666601.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.708433198928833,
|
|
"epoch": 2.5807787903893953,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046063930286754267,
|
|
"loss": 5.4922,
|
|
"mean_token_accuracy": 0.16438818722963333,
|
|
"num_tokens": 5675243.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.607330846786499,
|
|
"epoch": 2.5849212924606464,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046045855368366756,
|
|
"loss": 5.5122,
|
|
"mean_token_accuracy": 0.15873268097639084,
|
|
"num_tokens": 5683646.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.735720300674439,
|
|
"epoch": 2.5890637945318975,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000460277430409623,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.1679003655910492,
|
|
"num_tokens": 5693069.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.556051111221313,
|
|
"epoch": 2.593206296603148,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046009593341075444,
|
|
"loss": 5.4092,
|
|
"mean_token_accuracy": 0.16623837649822235,
|
|
"num_tokens": 5701527.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.590375089645386,
|
|
"epoch": 2.5973487986743993,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004599140630531612,
|
|
"loss": 5.5604,
|
|
"mean_token_accuracy": 0.16305261999368667,
|
|
"num_tokens": 5710379.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.747953128814697,
|
|
"epoch": 2.6014913007456504,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004597318197036956,
|
|
"loss": 5.48,
|
|
"mean_token_accuracy": 0.1645922839641571,
|
|
"num_tokens": 5719217.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.658558940887451,
|
|
"epoch": 2.6056338028169015,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004595492037299624,
|
|
"loss": 5.5902,
|
|
"mean_token_accuracy": 0.15660670399665833,
|
|
"num_tokens": 5728331.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.670481443405151,
|
|
"epoch": 2.6097763048881526,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045936621550031806,
|
|
"loss": 5.5444,
|
|
"mean_token_accuracy": 0.16000198125839232,
|
|
"num_tokens": 5737696.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.761334562301636,
|
|
"epoch": 2.6139188069594033,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045918285538386973,
|
|
"loss": 5.616,
|
|
"mean_token_accuracy": 0.16404765099287033,
|
|
"num_tokens": 5746191.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.724868297576904,
|
|
"epoch": 2.6180613090306544,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045899912375047485,
|
|
"loss": 5.6065,
|
|
"mean_token_accuracy": 0.16577061861753464,
|
|
"num_tokens": 5754824.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.630853986740112,
|
|
"epoch": 2.6222038111019055,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045881502097074023,
|
|
"loss": 5.5528,
|
|
"mean_token_accuracy": 0.1559559613466263,
|
|
"num_tokens": 5764863.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.6735029220581055,
|
|
"epoch": 2.6263463131731566,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004586305474160212,
|
|
"loss": 5.5744,
|
|
"mean_token_accuracy": 0.15252353399991989,
|
|
"num_tokens": 5774273.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.654740190505981,
|
|
"epoch": 2.6304888152444077,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045844570345842116,
|
|
"loss": 5.5444,
|
|
"mean_token_accuracy": 0.1586697816848755,
|
|
"num_tokens": 5784085.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.673130083084106,
|
|
"epoch": 2.634631317315659,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004582604894707906,
|
|
"loss": 5.5743,
|
|
"mean_token_accuracy": 0.1666999340057373,
|
|
"num_tokens": 5793921.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.719259262084961,
|
|
"epoch": 2.63877381938691,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045807490582672634,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.1527248203754425,
|
|
"num_tokens": 5803902.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.607631492614746,
|
|
"epoch": 2.6429163214581606,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045788895290057083,
|
|
"loss": 5.4507,
|
|
"mean_token_accuracy": 0.16747543811798096,
|
|
"num_tokens": 5812089.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.708329534530639,
|
|
"epoch": 2.6470588235294117,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045770263106741137,
|
|
"loss": 5.5981,
|
|
"mean_token_accuracy": 0.15919173806905745,
|
|
"num_tokens": 5821072.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.637457323074341,
|
|
"epoch": 2.651201325600663,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004575159407030797,
|
|
"loss": 5.4373,
|
|
"mean_token_accuracy": 0.16173361092805863,
|
|
"num_tokens": 5829725.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.542225885391235,
|
|
"epoch": 2.655343827671914,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004573288821841505,
|
|
"loss": 5.5196,
|
|
"mean_token_accuracy": 0.16196130812168122,
|
|
"num_tokens": 5838153.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.6365491390228275,
|
|
"epoch": 2.659486329743165,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004571414558879413,
|
|
"loss": 5.4599,
|
|
"mean_token_accuracy": 0.15795936137437822,
|
|
"num_tokens": 5847270.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.676420450210571,
|
|
"epoch": 2.6636288318144157,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004569536621925114,
|
|
"loss": 5.5895,
|
|
"mean_token_accuracy": 0.15637888610363007,
|
|
"num_tokens": 5855954.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.689839649200439,
|
|
"epoch": 2.667771333885667,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004567655014766612,
|
|
"loss": 5.4853,
|
|
"mean_token_accuracy": 0.1627836525440216,
|
|
"num_tokens": 5864435.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.640894746780395,
|
|
"epoch": 2.671913835956918,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004565769741199314,
|
|
"loss": 5.5571,
|
|
"mean_token_accuracy": 0.15790077894926072,
|
|
"num_tokens": 5873964.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.725470733642578,
|
|
"epoch": 2.676056338028169,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045638808050260237,
|
|
"loss": 5.5074,
|
|
"mean_token_accuracy": 0.16316789835691453,
|
|
"num_tokens": 5882738.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.639863920211792,
|
|
"epoch": 2.68019884009942,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045619882100569317,
|
|
"loss": 5.6406,
|
|
"mean_token_accuracy": 0.15904040038585662,
|
|
"num_tokens": 5891749.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.716983032226563,
|
|
"epoch": 2.684341342170671,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004560091960109608,
|
|
"loss": 5.5776,
|
|
"mean_token_accuracy": 0.15503172799944878,
|
|
"num_tokens": 5900551.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.711254215240478,
|
|
"epoch": 2.6884838442419223,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004558192059008996,
|
|
"loss": 5.595,
|
|
"mean_token_accuracy": 0.154455304145813,
|
|
"num_tokens": 5909254.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.6522393226623535,
|
|
"epoch": 2.692626346313173,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004556288510587406,
|
|
"loss": 5.5395,
|
|
"mean_token_accuracy": 0.15719025135040282,
|
|
"num_tokens": 5918357.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.7096232891082765,
|
|
"epoch": 2.696768848384424,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045543813186845004,
|
|
"loss": 5.5289,
|
|
"mean_token_accuracy": 0.15661482959985734,
|
|
"num_tokens": 5927742.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.649437093734742,
|
|
"epoch": 2.700911350455675,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004552470487147294,
|
|
"loss": 5.5935,
|
|
"mean_token_accuracy": 0.15934477895498275,
|
|
"num_tokens": 5937025.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.805260419845581,
|
|
"epoch": 2.7050538525269263,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045505560198301443,
|
|
"loss": 5.6019,
|
|
"mean_token_accuracy": 0.15056679993867875,
|
|
"num_tokens": 5946176.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.671889066696167,
|
|
"epoch": 2.7091963545981774,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045486379205947396,
|
|
"loss": 5.5669,
|
|
"mean_token_accuracy": 0.15695112347602844,
|
|
"num_tokens": 5955550.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.621965122222901,
|
|
"epoch": 2.713338856669428,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004546716193310098,
|
|
"loss": 5.4955,
|
|
"mean_token_accuracy": 0.15768464505672455,
|
|
"num_tokens": 5964467.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.697195816040039,
|
|
"epoch": 2.717481358740679,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000454479084185255,
|
|
"loss": 5.5231,
|
|
"mean_token_accuracy": 0.15613993704319,
|
|
"num_tokens": 5973906.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.676262474060058,
|
|
"epoch": 2.7216238608119303,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004542861870105743,
|
|
"loss": 5.5458,
|
|
"mean_token_accuracy": 0.16207694113254548,
|
|
"num_tokens": 5983701.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.698145246505737,
|
|
"epoch": 2.7257663628831814,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045409292819606224,
|
|
"loss": 5.5323,
|
|
"mean_token_accuracy": 0.1577382504940033,
|
|
"num_tokens": 5993367.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.663068532943726,
|
|
"epoch": 2.7299088649544325,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045389930813154297,
|
|
"loss": 5.4883,
|
|
"mean_token_accuracy": 0.16210660040378572,
|
|
"num_tokens": 6003752.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.766804790496826,
|
|
"epoch": 2.7340513670256836,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004537053272075693,
|
|
"loss": 5.5558,
|
|
"mean_token_accuracy": 0.16685177832841874,
|
|
"num_tokens": 6012486.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.5676192283630375,
|
|
"epoch": 2.7381938690969347,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045351098581542204,
|
|
"loss": 5.5081,
|
|
"mean_token_accuracy": 0.16729753464460373,
|
|
"num_tokens": 6020742.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.710718393325806,
|
|
"epoch": 2.742336371168186,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004533162843471089,
|
|
"loss": 5.5413,
|
|
"mean_token_accuracy": 0.1567221537232399,
|
|
"num_tokens": 6030248.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.695099592208862,
|
|
"epoch": 2.7464788732394365,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004531212231953641,
|
|
"loss": 5.5763,
|
|
"mean_token_accuracy": 0.1570330873131752,
|
|
"num_tokens": 6039293.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.642256498336792,
|
|
"epoch": 2.7506213753106876,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004529258027536473,
|
|
"loss": 5.4417,
|
|
"mean_token_accuracy": 0.16368594020605087,
|
|
"num_tokens": 6048081.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.669494771957398,
|
|
"epoch": 2.7547638773819387,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004527300234161428,
|
|
"loss": 5.5984,
|
|
"mean_token_accuracy": 0.16303794980049133,
|
|
"num_tokens": 6057549.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.64669942855835,
|
|
"epoch": 2.75890637945319,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004525338855777591,
|
|
"loss": 5.6012,
|
|
"mean_token_accuracy": 0.15381733626127242,
|
|
"num_tokens": 6067372.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.687849664688111,
|
|
"epoch": 2.7630488815244405,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045233738963412746,
|
|
"loss": 5.457,
|
|
"mean_token_accuracy": 0.16840497851371766,
|
|
"num_tokens": 6075532.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.5901408195495605,
|
|
"epoch": 2.7671913835956916,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045214053598160173,
|
|
"loss": 5.4745,
|
|
"mean_token_accuracy": 0.16109947264194488,
|
|
"num_tokens": 6084389.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.622766542434692,
|
|
"epoch": 2.7713338856669427,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045194332501725725,
|
|
"loss": 5.4957,
|
|
"mean_token_accuracy": 0.16854244321584702,
|
|
"num_tokens": 6093536.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.716071701049804,
|
|
"epoch": 2.775476387738194,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045174575713889014,
|
|
"loss": 5.5674,
|
|
"mean_token_accuracy": 0.16249414160847664,
|
|
"num_tokens": 6102677.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.686060762405395,
|
|
"epoch": 2.779618889809445,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004515478327450164,
|
|
"loss": 5.6364,
|
|
"mean_token_accuracy": 0.15851696133613585,
|
|
"num_tokens": 6112430.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.725155210494995,
|
|
"epoch": 2.783761391880696,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045134955223487093,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.16037144213914872,
|
|
"num_tokens": 6121408.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.6291951656341555,
|
|
"epoch": 2.787903893951947,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045115091600840757,
|
|
"loss": 5.5594,
|
|
"mean_token_accuracy": 0.1527995854616165,
|
|
"num_tokens": 6131339.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.695040321350097,
|
|
"epoch": 2.7920463960231983,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045095192446629697,
|
|
"loss": 5.5165,
|
|
"mean_token_accuracy": 0.16150978207588196,
|
|
"num_tokens": 6140590.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.7630918502807615,
|
|
"epoch": 2.796188898094449,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000450752578009927,
|
|
"loss": 5.6064,
|
|
"mean_token_accuracy": 0.16191209107637405,
|
|
"num_tokens": 6149958.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.5992176055908205,
|
|
"epoch": 2.8003314001657,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045055287704140115,
|
|
"loss": 5.4601,
|
|
"mean_token_accuracy": 0.16068203300237655,
|
|
"num_tokens": 6159411.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.603084850311279,
|
|
"epoch": 2.804473902236951,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004503528219635381,
|
|
"loss": 5.5476,
|
|
"mean_token_accuracy": 0.15345613062381744,
|
|
"num_tokens": 6169264.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.713048839569092,
|
|
"epoch": 2.8086164043082023,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004501524131798709,
|
|
"loss": 5.5272,
|
|
"mean_token_accuracy": 0.15406270325183868,
|
|
"num_tokens": 6178374.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.732380008697509,
|
|
"epoch": 2.812758906379453,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000449951651094646,
|
|
"loss": 5.5217,
|
|
"mean_token_accuracy": 0.1573767378926277,
|
|
"num_tokens": 6186918.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.575137805938721,
|
|
"epoch": 2.816901408450704,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044975053611282217,
|
|
"loss": 5.5749,
|
|
"mean_token_accuracy": 0.1632239282131195,
|
|
"num_tokens": 6196400.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.593220329284668,
|
|
"epoch": 2.821043910521955,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004495490686400706,
|
|
"loss": 5.4258,
|
|
"mean_token_accuracy": 0.16297936737537383,
|
|
"num_tokens": 6204849.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.606288814544678,
|
|
"epoch": 2.8251864125932062,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044934724908277313,
|
|
"loss": 5.5297,
|
|
"mean_token_accuracy": 0.160382042825222,
|
|
"num_tokens": 6214121.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.674342155456543,
|
|
"epoch": 2.8293289146644574,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044914507784802184,
|
|
"loss": 5.5168,
|
|
"mean_token_accuracy": 0.16667052805423738,
|
|
"num_tokens": 6223163.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.644378805160523,
|
|
"epoch": 2.8334714167357085,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004489425553436182,
|
|
"loss": 5.5595,
|
|
"mean_token_accuracy": 0.15832413882017135,
|
|
"num_tokens": 6231581.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.681396913528443,
|
|
"epoch": 2.8376139188069596,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044873968197807213,
|
|
"loss": 5.6319,
|
|
"mean_token_accuracy": 0.15717055946588515,
|
|
"num_tokens": 6240015.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.746704483032227,
|
|
"epoch": 2.8417564208782107,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044853645816060156,
|
|
"loss": 5.618,
|
|
"mean_token_accuracy": 0.15753672495484353,
|
|
"num_tokens": 6249083.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.720417594909668,
|
|
"epoch": 2.8458989229494613,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044833288430113104,
|
|
"loss": 5.5847,
|
|
"mean_token_accuracy": 0.15910181850194932,
|
|
"num_tokens": 6258798.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.616397666931152,
|
|
"epoch": 2.8500414250207124,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044812896081029136,
|
|
"loss": 5.4288,
|
|
"mean_token_accuracy": 0.17185353338718415,
|
|
"num_tokens": 6267819.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.6485264778137205,
|
|
"epoch": 2.8541839270919636,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004479246880994183,
|
|
"loss": 5.5208,
|
|
"mean_token_accuracy": 0.16037193909287453,
|
|
"num_tokens": 6276915.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.701795244216919,
|
|
"epoch": 2.8583264291632147,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004477200665805525,
|
|
"loss": 5.5497,
|
|
"mean_token_accuracy": 0.16226860135793686,
|
|
"num_tokens": 6285880.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.694223499298095,
|
|
"epoch": 2.8624689312344658,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044751509666643777,
|
|
"loss": 5.6782,
|
|
"mean_token_accuracy": 0.15221977159380912,
|
|
"num_tokens": 6295408.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.652013063430786,
|
|
"epoch": 2.8666114333057164,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044730977877052084,
|
|
"loss": 5.4072,
|
|
"mean_token_accuracy": 0.16112574934959412,
|
|
"num_tokens": 6304110.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.673219871520996,
|
|
"epoch": 2.8707539353769675,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044710411330695045,
|
|
"loss": 5.5651,
|
|
"mean_token_accuracy": 0.15716703981161118,
|
|
"num_tokens": 6312577.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.815633296966553,
|
|
"epoch": 2.8748964374482187,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044689810069057615,
|
|
"loss": 5.6407,
|
|
"mean_token_accuracy": 0.15225453972816466,
|
|
"num_tokens": 6321766.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.69540114402771,
|
|
"epoch": 2.8790389395194698,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004466917413369479,
|
|
"loss": 5.5972,
|
|
"mean_token_accuracy": 0.15052569210529326,
|
|
"num_tokens": 6330327.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.617451906204224,
|
|
"epoch": 2.883181441590721,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004464850356623152,
|
|
"loss": 5.5372,
|
|
"mean_token_accuracy": 0.161219622194767,
|
|
"num_tokens": 6339596.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.652078914642334,
|
|
"epoch": 2.887323943661972,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044627798408362597,
|
|
"loss": 5.4389,
|
|
"mean_token_accuracy": 0.16472167521715164,
|
|
"num_tokens": 6348080.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.672922134399414,
|
|
"epoch": 2.891466445733223,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004460705870185258,
|
|
"loss": 5.5601,
|
|
"mean_token_accuracy": 0.15549688041210175,
|
|
"num_tokens": 6357725.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.644254064559936,
|
|
"epoch": 2.8956089478044738,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004458628448853572,
|
|
"loss": 5.5009,
|
|
"mean_token_accuracy": 0.16381119638681413,
|
|
"num_tokens": 6366821.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.726661252975464,
|
|
"epoch": 2.899751449875725,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004456547581031589,
|
|
"loss": 5.579,
|
|
"mean_token_accuracy": 0.15758128613233566,
|
|
"num_tokens": 6375641.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.899751449875725,
|
|
"eval_entropy": 5.521543442821706,
|
|
"eval_loss": 5.915812969207764,
|
|
"eval_mean_token_accuracy": 0.150046447971101,
|
|
"eval_num_tokens": 6375641.0,
|
|
"eval_runtime": 2.7561,
|
|
"eval_samples_per_second": 1359.533,
|
|
"eval_steps_per_second": 170.168,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.617849683761596,
|
|
"epoch": 2.903893951946976,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004454463270916646,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.16418685615062714,
|
|
"num_tokens": 6384596.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.794553995132446,
|
|
"epoch": 2.908036454018227,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004452375522713023,
|
|
"loss": 5.777,
|
|
"mean_token_accuracy": 0.14264359697699547,
|
|
"num_tokens": 6395048.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.6611549854278564,
|
|
"epoch": 2.912178956089478,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004450284340631938,
|
|
"loss": 5.6103,
|
|
"mean_token_accuracy": 0.15344774425029756,
|
|
"num_tokens": 6404828.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.663045120239258,
|
|
"epoch": 2.916321458160729,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044481897288915334,
|
|
"loss": 5.4779,
|
|
"mean_token_accuracy": 0.15861677676439284,
|
|
"num_tokens": 6414670.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.5174237251281735,
|
|
"epoch": 2.92046396023198,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044460916917168697,
|
|
"loss": 5.4254,
|
|
"mean_token_accuracy": 0.16657451838254927,
|
|
"num_tokens": 6423446.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.599793004989624,
|
|
"epoch": 2.924606462303231,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004443990233339918,
|
|
"loss": 5.4952,
|
|
"mean_token_accuracy": 0.1578425571322441,
|
|
"num_tokens": 6433412.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.707996892929077,
|
|
"epoch": 2.928748964374482,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004441885357999548,
|
|
"loss": 5.4906,
|
|
"mean_token_accuracy": 0.1628717914223671,
|
|
"num_tokens": 6443520.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.676200866699219,
|
|
"epoch": 2.9328914664457333,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004439777069941524,
|
|
"loss": 5.5963,
|
|
"mean_token_accuracy": 0.15572925359010698,
|
|
"num_tokens": 6452933.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.624213981628418,
|
|
"epoch": 2.9370339685169844,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044376653734184935,
|
|
"loss": 5.4056,
|
|
"mean_token_accuracy": 0.16887385100126268,
|
|
"num_tokens": 6462113.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.549249124526978,
|
|
"epoch": 2.9411764705882355,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004435550272689979,
|
|
"loss": 5.4574,
|
|
"mean_token_accuracy": 0.16682465970516205,
|
|
"num_tokens": 6471411.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.7074384689331055,
|
|
"epoch": 2.945318972659486,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044334317720223706,
|
|
"loss": 5.5627,
|
|
"mean_token_accuracy": 0.15771554186940193,
|
|
"num_tokens": 6480682.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.644071817398071,
|
|
"epoch": 2.9494614747307373,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044313098756889154,
|
|
"loss": 5.6091,
|
|
"mean_token_accuracy": 0.15549242049455642,
|
|
"num_tokens": 6490614.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.680739021301269,
|
|
"epoch": 2.9536039768019884,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004429184587969708,
|
|
"loss": 5.4566,
|
|
"mean_token_accuracy": 0.1625344857573509,
|
|
"num_tokens": 6499079.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.67983226776123,
|
|
"epoch": 2.9577464788732395,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004427055913151689,
|
|
"loss": 5.573,
|
|
"mean_token_accuracy": 0.14837478697299958,
|
|
"num_tokens": 6508556.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.678392362594605,
|
|
"epoch": 2.9618889809444906,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004424923855528626,
|
|
"loss": 5.59,
|
|
"mean_token_accuracy": 0.155683271586895,
|
|
"num_tokens": 6517934.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.643442249298095,
|
|
"epoch": 2.9660314830157413,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004422788419401114,
|
|
"loss": 5.4576,
|
|
"mean_token_accuracy": 0.16748119741678238,
|
|
"num_tokens": 6526995.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.613794136047363,
|
|
"epoch": 2.9701739850869924,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044206496090765595,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.15781812220811844,
|
|
"num_tokens": 6535264.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.682992553710937,
|
|
"epoch": 2.9743164871582435,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044185074288691776,
|
|
"loss": 5.4229,
|
|
"mean_token_accuracy": 0.17119230926036835,
|
|
"num_tokens": 6544498.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.585231161117553,
|
|
"epoch": 2.9784589892294946,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044163618830999794,
|
|
"loss": 5.54,
|
|
"mean_token_accuracy": 0.15998248010873795,
|
|
"num_tokens": 6554761.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.627832126617432,
|
|
"epoch": 2.9826014913007457,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044142129760967643,
|
|
"loss": 5.596,
|
|
"mean_token_accuracy": 0.1564616858959198,
|
|
"num_tokens": 6564039.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.723354148864746,
|
|
"epoch": 2.986743993371997,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004412060712194115,
|
|
"loss": 5.5456,
|
|
"mean_token_accuracy": 0.1615665450692177,
|
|
"num_tokens": 6572559.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.61319408416748,
|
|
"epoch": 2.990886495443248,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004409905095733381,
|
|
"loss": 5.5347,
|
|
"mean_token_accuracy": 0.158642216026783,
|
|
"num_tokens": 6581621.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.608583641052246,
|
|
"epoch": 2.995028997514499,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044077461310626764,
|
|
"loss": 5.5114,
|
|
"mean_token_accuracy": 0.1594192534685135,
|
|
"num_tokens": 6590353.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.687379932403564,
|
|
"epoch": 2.9991714995857497,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000440558382253687,
|
|
"loss": 5.5038,
|
|
"mean_token_accuracy": 0.15936833918094634,
|
|
"num_tokens": 6598822.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.65473690032959,
|
|
"epoch": 3.003314001657001,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004403418174517573,
|
|
"loss": 5.2736,
|
|
"mean_token_accuracy": 0.1684224545955658,
|
|
"num_tokens": 6606728.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.608128547668457,
|
|
"epoch": 3.007456503728252,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044012491913731353,
|
|
"loss": 5.1922,
|
|
"mean_token_accuracy": 0.171491277217865,
|
|
"num_tokens": 6616226.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.528424024581909,
|
|
"epoch": 3.011599005799503,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004399076877478632,
|
|
"loss": 5.0393,
|
|
"mean_token_accuracy": 0.18953252136707305,
|
|
"num_tokens": 6624801.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.629195356369019,
|
|
"epoch": 3.015741507870754,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004396901237215858,
|
|
"loss": 5.2488,
|
|
"mean_token_accuracy": 0.17343629151582718,
|
|
"num_tokens": 6633043.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.576674175262451,
|
|
"epoch": 3.019884009942005,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043947222749733173,
|
|
"loss": 5.1919,
|
|
"mean_token_accuracy": 0.17088026255369188,
|
|
"num_tokens": 6641197.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.592171764373779,
|
|
"epoch": 3.024026512013256,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043925399951462153,
|
|
"loss": 5.2255,
|
|
"mean_token_accuracy": 0.17408303022384644,
|
|
"num_tokens": 6650411.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.478409099578857,
|
|
"epoch": 3.028169014084507,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004390354402136449,
|
|
"loss": 5.1133,
|
|
"mean_token_accuracy": 0.1816069394350052,
|
|
"num_tokens": 6659405.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.61852650642395,
|
|
"epoch": 3.032311516155758,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004388165500352597,
|
|
"loss": 5.1873,
|
|
"mean_token_accuracy": 0.18107183575630187,
|
|
"num_tokens": 6667701.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.6003812789917,
|
|
"epoch": 3.036454018227009,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004385973294209915,
|
|
"loss": 5.2141,
|
|
"mean_token_accuracy": 0.1739385098218918,
|
|
"num_tokens": 6676801.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.48479471206665,
|
|
"epoch": 3.0405965202982603,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004383777788130321,
|
|
"loss": 5.0888,
|
|
"mean_token_accuracy": 0.1855471447110176,
|
|
"num_tokens": 6686005.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.496511220932007,
|
|
"epoch": 3.044739022369511,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004381578986542393,
|
|
"loss": 5.1157,
|
|
"mean_token_accuracy": 0.17760300189256667,
|
|
"num_tokens": 6694884.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.483774185180664,
|
|
"epoch": 3.048881524440762,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004379376893881352,
|
|
"loss": 5.2303,
|
|
"mean_token_accuracy": 0.18040781468153,
|
|
"num_tokens": 6703476.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.610590934753418,
|
|
"epoch": 3.053024026512013,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004377171514589059,
|
|
"loss": 5.2702,
|
|
"mean_token_accuracy": 0.17431289851665496,
|
|
"num_tokens": 6712576.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.457471132278442,
|
|
"epoch": 3.0571665285832643,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004374962853114007,
|
|
"loss": 5.0868,
|
|
"mean_token_accuracy": 0.17861308753490449,
|
|
"num_tokens": 6721532.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.460675001144409,
|
|
"epoch": 3.0613090306545154,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004372750913911306,
|
|
"loss": 5.1198,
|
|
"mean_token_accuracy": 0.1840505540370941,
|
|
"num_tokens": 6730055.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.507229948043824,
|
|
"epoch": 3.0654515327257665,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004370535701442679,
|
|
"loss": 5.1823,
|
|
"mean_token_accuracy": 0.17917860001325608,
|
|
"num_tokens": 6739310.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.547648572921753,
|
|
"epoch": 3.069594034797017,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043683172201764537,
|
|
"loss": 5.2717,
|
|
"mean_token_accuracy": 0.1702258437871933,
|
|
"num_tokens": 6748976.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.529606866836548,
|
|
"epoch": 3.0737365368682683,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004366095474587547,
|
|
"loss": 5.1182,
|
|
"mean_token_accuracy": 0.17836759090423585,
|
|
"num_tokens": 6758043.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.521498918533325,
|
|
"epoch": 3.0778790389395194,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043638704691574646,
|
|
"loss": 5.1912,
|
|
"mean_token_accuracy": 0.17908665388822556,
|
|
"num_tokens": 6767554.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.60535740852356,
|
|
"epoch": 3.0820215410107705,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004361642208374283,
|
|
"loss": 5.2083,
|
|
"mean_token_accuracy": 0.17299294024705886,
|
|
"num_tokens": 6776161.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.466820573806762,
|
|
"epoch": 3.0861640430820216,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043594106967326495,
|
|
"loss": 5.1217,
|
|
"mean_token_accuracy": 0.17925718873739244,
|
|
"num_tokens": 6785371.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.5335582256317135,
|
|
"epoch": 3.0903065451532727,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043571759387337665,
|
|
"loss": 5.1634,
|
|
"mean_token_accuracy": 0.1835888460278511,
|
|
"num_tokens": 6795242.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.547745561599731,
|
|
"epoch": 3.0944490472245234,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004354937938885385,
|
|
"loss": 5.249,
|
|
"mean_token_accuracy": 0.1728634551167488,
|
|
"num_tokens": 6805423.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.529517984390258,
|
|
"epoch": 3.0985915492957745,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004352696701701796,
|
|
"loss": 5.1689,
|
|
"mean_token_accuracy": 0.17931755483150483,
|
|
"num_tokens": 6814377.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.491488599777222,
|
|
"epoch": 3.1027340513670256,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004350452231703819,
|
|
"loss": 5.2307,
|
|
"mean_token_accuracy": 0.17575548440217972,
|
|
"num_tokens": 6824025.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.503986501693726,
|
|
"epoch": 3.1068765534382767,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043482045334187954,
|
|
"loss": 5.1449,
|
|
"mean_token_accuracy": 0.18518491089344025,
|
|
"num_tokens": 6833071.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.433632755279541,
|
|
"epoch": 3.111019055509528,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043459536113805785,
|
|
"loss": 5.1352,
|
|
"mean_token_accuracy": 0.1775007888674736,
|
|
"num_tokens": 6841983.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.491881084442139,
|
|
"epoch": 3.115161557580779,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043436994701295237,
|
|
"loss": 5.1331,
|
|
"mean_token_accuracy": 0.18157943040132524,
|
|
"num_tokens": 6851436.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.493215036392212,
|
|
"epoch": 3.1193040596520296,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043414421142124804,
|
|
"loss": 5.1783,
|
|
"mean_token_accuracy": 0.1788194492459297,
|
|
"num_tokens": 6861810.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.526950836181641,
|
|
"epoch": 3.1234465617232807,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043391815481827814,
|
|
"loss": 5.0796,
|
|
"mean_token_accuracy": 0.18713754117488862,
|
|
"num_tokens": 6870418.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.399437761306762,
|
|
"epoch": 3.127589063794532,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004336917776600237,
|
|
"loss": 5.0979,
|
|
"mean_token_accuracy": 0.18566158562898635,
|
|
"num_tokens": 6879419.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.424571800231933,
|
|
"epoch": 3.131731565865783,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004334650804031122,
|
|
"loss": 5.2139,
|
|
"mean_token_accuracy": 0.17417935729026796,
|
|
"num_tokens": 6888768.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.586405467987061,
|
|
"epoch": 3.135874067937034,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004332380635048165,
|
|
"loss": 5.1836,
|
|
"mean_token_accuracy": 0.17237520962953568,
|
|
"num_tokens": 6897133.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.5270240783691404,
|
|
"epoch": 3.140016570008285,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043301072742305474,
|
|
"loss": 5.1976,
|
|
"mean_token_accuracy": 0.17571339756250381,
|
|
"num_tokens": 6904976.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.558376359939575,
|
|
"epoch": 3.1441590720795363,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004327830726163886,
|
|
"loss": 5.196,
|
|
"mean_token_accuracy": 0.1767548218369484,
|
|
"num_tokens": 6915138.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.543389368057251,
|
|
"epoch": 3.148301574150787,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004325550995440226,
|
|
"loss": 5.1739,
|
|
"mean_token_accuracy": 0.17608616650104522,
|
|
"num_tokens": 6923962.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.395559072494507,
|
|
"epoch": 3.152444076222038,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043232680866580356,
|
|
"loss": 5.1517,
|
|
"mean_token_accuracy": 0.17694357335567473,
|
|
"num_tokens": 6932102.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.490355110168457,
|
|
"epoch": 3.156586578293289,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00043209820044221893,
|
|
"loss": 5.1846,
|
|
"mean_token_accuracy": 0.18131535947322847,
|
|
"num_tokens": 6940354.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.519707345962525,
|
|
"epoch": 3.1607290803645403,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043186927533439654,
|
|
"loss": 5.1749,
|
|
"mean_token_accuracy": 0.17507922053337097,
|
|
"num_tokens": 6949779.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.551403045654297,
|
|
"epoch": 3.1648715824357914,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043164003380410335,
|
|
"loss": 5.2449,
|
|
"mean_token_accuracy": 0.17628766000270843,
|
|
"num_tokens": 6958760.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.479237985610962,
|
|
"epoch": 3.169014084507042,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004314104763137445,
|
|
"loss": 5.1042,
|
|
"mean_token_accuracy": 0.18383505642414094,
|
|
"num_tokens": 6967535.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.487510824203492,
|
|
"epoch": 3.173156586578293,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043118060332636264,
|
|
"loss": 5.2234,
|
|
"mean_token_accuracy": 0.17976367324590684,
|
|
"num_tokens": 6976211.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.569280815124512,
|
|
"epoch": 3.1772990886495442,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043095041530563665,
|
|
"loss": 5.2344,
|
|
"mean_token_accuracy": 0.176742222905159,
|
|
"num_tokens": 6986001.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.433036470413208,
|
|
"epoch": 3.1814415907207954,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043071991271588104,
|
|
"loss": 5.2003,
|
|
"mean_token_accuracy": 0.18441623449325562,
|
|
"num_tokens": 6994568.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.5656046867370605,
|
|
"epoch": 3.1855840927920465,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004304890960220446,
|
|
"loss": 5.322,
|
|
"mean_token_accuracy": 0.16879098266363143,
|
|
"num_tokens": 7004744.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.5625487804412845,
|
|
"epoch": 3.1897265948632976,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043025796568970975,
|
|
"loss": 5.3119,
|
|
"mean_token_accuracy": 0.167134191095829,
|
|
"num_tokens": 7014315.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.508922958374024,
|
|
"epoch": 3.1938690969345487,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043002652218509184,
|
|
"loss": 5.1492,
|
|
"mean_token_accuracy": 0.18394915163516998,
|
|
"num_tokens": 7022258.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.534724378585816,
|
|
"epoch": 3.1980115990057993,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004297947659750376,
|
|
"loss": 5.259,
|
|
"mean_token_accuracy": 0.17378944158554077,
|
|
"num_tokens": 7031732.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.489095020294189,
|
|
"epoch": 3.2021541010770505,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004295626975270246,
|
|
"loss": 5.193,
|
|
"mean_token_accuracy": 0.17550500780344008,
|
|
"num_tokens": 7040478.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.512714910507202,
|
|
"epoch": 3.2062966031483016,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042933031730916024,
|
|
"loss": 5.1635,
|
|
"mean_token_accuracy": 0.1809170812368393,
|
|
"num_tokens": 7049360.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.5673205852508545,
|
|
"epoch": 3.2104391052195527,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000429097625790181,
|
|
"loss": 5.2467,
|
|
"mean_token_accuracy": 0.17765278667211531,
|
|
"num_tokens": 7059271.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.450319290161133,
|
|
"epoch": 3.2145816072908038,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004288646234394508,
|
|
"loss": 5.2269,
|
|
"mean_token_accuracy": 0.17899532318115235,
|
|
"num_tokens": 7069038.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.491114950180053,
|
|
"epoch": 3.218724109362055,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042863131072696115,
|
|
"loss": 5.1796,
|
|
"mean_token_accuracy": 0.1805991604924202,
|
|
"num_tokens": 7077971.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.488774967193604,
|
|
"epoch": 3.2228666114333056,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042839768812332905,
|
|
"loss": 5.1946,
|
|
"mean_token_accuracy": 0.16922565698623657,
|
|
"num_tokens": 7086821.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.46287989616394,
|
|
"epoch": 3.2270091135045567,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000428163756099797,
|
|
"loss": 5.2217,
|
|
"mean_token_accuracy": 0.17228440195322037,
|
|
"num_tokens": 7095407.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.468328332901001,
|
|
"epoch": 3.2311516155758078,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004279295151282314,
|
|
"loss": 5.1369,
|
|
"mean_token_accuracy": 0.18177154362201692,
|
|
"num_tokens": 7103964.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.498992919921875,
|
|
"epoch": 3.235294117647059,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004276949656811217,
|
|
"loss": 5.2569,
|
|
"mean_token_accuracy": 0.1760358139872551,
|
|
"num_tokens": 7113571.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.572669744491577,
|
|
"epoch": 3.23943661971831,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042746010823158006,
|
|
"loss": 5.2735,
|
|
"mean_token_accuracy": 0.17531929761171341,
|
|
"num_tokens": 7122844.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.450087594985962,
|
|
"epoch": 3.243579121789561,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042722494325333945,
|
|
"loss": 5.172,
|
|
"mean_token_accuracy": 0.17707158029079437,
|
|
"num_tokens": 7132123.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.493407535552978,
|
|
"epoch": 3.2477216238608118,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004269894712207533,
|
|
"loss": 5.208,
|
|
"mean_token_accuracy": 0.17930082976818085,
|
|
"num_tokens": 7141068.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.45019793510437,
|
|
"epoch": 3.251864125932063,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004267536926087947,
|
|
"loss": 5.2217,
|
|
"mean_token_accuracy": 0.17073139399290085,
|
|
"num_tokens": 7150829.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.46892352104187,
|
|
"epoch": 3.256006628003314,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004265176078930545,
|
|
"loss": 5.1771,
|
|
"mean_token_accuracy": 0.18690198361873628,
|
|
"num_tokens": 7160814.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.543881416320801,
|
|
"epoch": 3.260149130074565,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042628121754974166,
|
|
"loss": 5.2803,
|
|
"mean_token_accuracy": 0.1723069131374359,
|
|
"num_tokens": 7170085.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.560903167724609,
|
|
"epoch": 3.264291632145816,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004260445220556811,
|
|
"loss": 5.2425,
|
|
"mean_token_accuracy": 0.17817744761705398,
|
|
"num_tokens": 7179366.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.578168535232544,
|
|
"epoch": 3.2684341342170673,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00042580752188831376,
|
|
"loss": 5.304,
|
|
"mean_token_accuracy": 0.17222840785980226,
|
|
"num_tokens": 7188342.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.578952741622925,
|
|
"epoch": 3.272576636288318,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004255702175256946,
|
|
"loss": 5.3315,
|
|
"mean_token_accuracy": 0.1791801705956459,
|
|
"num_tokens": 7197466.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.46775631904602,
|
|
"epoch": 3.276719138359569,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042533260944649273,
|
|
"loss": 5.1905,
|
|
"mean_token_accuracy": 0.17830806821584702,
|
|
"num_tokens": 7205462.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.4653631210327145,
|
|
"epoch": 3.28086164043082,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004250946981299893,
|
|
"loss": 5.2049,
|
|
"mean_token_accuracy": 0.17918088734149934,
|
|
"num_tokens": 7215115.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.471656036376953,
|
|
"epoch": 3.2850041425020713,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042485648405607776,
|
|
"loss": 5.1558,
|
|
"mean_token_accuracy": 0.1752260610461235,
|
|
"num_tokens": 7224909.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.363120985031128,
|
|
"epoch": 3.2891466445733224,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004246179677052617,
|
|
"loss": 5.082,
|
|
"mean_token_accuracy": 0.19152185171842576,
|
|
"num_tokens": 7232946.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.45828218460083,
|
|
"epoch": 3.2932891466445735,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004243791495586547,
|
|
"loss": 5.1925,
|
|
"mean_token_accuracy": 0.18146124333143235,
|
|
"num_tokens": 7241104.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.594405555725098,
|
|
"epoch": 3.297431648715824,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004241400300979791,
|
|
"loss": 5.2484,
|
|
"mean_token_accuracy": 0.17478115111589432,
|
|
"num_tokens": 7250749.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.433809566497803,
|
|
"epoch": 3.3015741507870753,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004239006098055651,
|
|
"loss": 5.1399,
|
|
"mean_token_accuracy": 0.1764796867966652,
|
|
"num_tokens": 7258687.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.400339555740357,
|
|
"epoch": 3.3057166528583264,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004236608891643494,
|
|
"loss": 5.179,
|
|
"mean_token_accuracy": 0.17788545340299605,
|
|
"num_tokens": 7267316.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.495499086380005,
|
|
"epoch": 3.3098591549295775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004234208686578748,
|
|
"loss": 5.1947,
|
|
"mean_token_accuracy": 0.17531768828630448,
|
|
"num_tokens": 7276734.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.459540557861328,
|
|
"epoch": 3.3140016570008286,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00042318054877028876,
|
|
"loss": 5.1834,
|
|
"mean_token_accuracy": 0.18060380220413208,
|
|
"num_tokens": 7285916.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.3140016570008286,
|
|
"eval_entropy": 5.368396492655089,
|
|
"eval_loss": 5.911151885986328,
|
|
"eval_mean_token_accuracy": 0.1531766704055292,
|
|
"eval_num_tokens": 7285916.0,
|
|
"eval_runtime": 2.7544,
|
|
"eval_samples_per_second": 1360.35,
|
|
"eval_steps_per_second": 170.271,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 12060,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.086948315648e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|