fe33fe6f22
Model: fpadovani/swe-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.2427506213753108,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 4088562477696000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|