fe33fe6f22
Model: fpadovani/swe-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.4142502071251036,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 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
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 12060,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1382675779584000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|