6035 lines
164 KiB
JSON
6035 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.1753463089601964,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742142391204833,
|
|
"epoch": 0.000292243848266994,
|
|
"grad_norm": 4.625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7907,
|
|
"mean_token_accuracy": 9.328357991762459e-05,
|
|
"num_tokens": 11126.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742114543914795,
|
|
"epoch": 0.000584487696533988,
|
|
"grad_norm": 5.21875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.763,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 20949.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742155933380127,
|
|
"epoch": 0.000876731544800982,
|
|
"grad_norm": 5.125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7441,
|
|
"mean_token_accuracy": 0.00010582010727375746,
|
|
"num_tokens": 31135.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.74212188720703,
|
|
"epoch": 0.001168975393067976,
|
|
"grad_norm": 4.46875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6777,
|
|
"mean_token_accuracy": 9.990009712055326e-05,
|
|
"num_tokens": 40153.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742057132720948,
|
|
"epoch": 0.00146121924133497,
|
|
"grad_norm": 4.84375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5935,
|
|
"mean_token_accuracy": 0.005301665072329343,
|
|
"num_tokens": 49826.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741645336151123,
|
|
"epoch": 0.001753463089601964,
|
|
"grad_norm": 4.0625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4734,
|
|
"mean_token_accuracy": 0.037928895093500614,
|
|
"num_tokens": 60054.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740150451660156,
|
|
"epoch": 0.002045706937868958,
|
|
"grad_norm": 3.390625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3332,
|
|
"mean_token_accuracy": 0.045763476938009265,
|
|
"num_tokens": 69706.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.735286903381347,
|
|
"epoch": 0.002337950786135952,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2072,
|
|
"mean_token_accuracy": 0.04907714687287808,
|
|
"num_tokens": 80908.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.72396411895752,
|
|
"epoch": 0.0026301946344029457,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.116,
|
|
"mean_token_accuracy": 0.0504226915538311,
|
|
"num_tokens": 91354.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.707611751556396,
|
|
"epoch": 0.00292243848266994,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.0357,
|
|
"mean_token_accuracy": 0.05158382542431354,
|
|
"num_tokens": 101448.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.695165920257569,
|
|
"epoch": 0.0032146823309369336,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.987,
|
|
"mean_token_accuracy": 0.04974161125719547,
|
|
"num_tokens": 111679.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.690466690063477,
|
|
"epoch": 0.003506926179203928,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.903,
|
|
"mean_token_accuracy": 0.05024520866572857,
|
|
"num_tokens": 122426.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.688275337219238,
|
|
"epoch": 0.0037991700274709215,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8425,
|
|
"mean_token_accuracy": 0.04879201892763376,
|
|
"num_tokens": 132600.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.682420635223389,
|
|
"epoch": 0.004091413875737916,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.803,
|
|
"mean_token_accuracy": 0.054092536121606825,
|
|
"num_tokens": 143485.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.67214002609253,
|
|
"epoch": 0.00438365772400491,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7878,
|
|
"mean_token_accuracy": 0.05034251846373081,
|
|
"num_tokens": 154697.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.66068525314331,
|
|
"epoch": 0.004675901572271904,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6894,
|
|
"mean_token_accuracy": 0.05171992778778076,
|
|
"num_tokens": 164883.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.651456069946288,
|
|
"epoch": 0.004968145420538897,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.6326,
|
|
"mean_token_accuracy": 0.05340392328798771,
|
|
"num_tokens": 175534.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.638569164276124,
|
|
"epoch": 0.0052603892688058915,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.6175,
|
|
"mean_token_accuracy": 0.05033664517104626,
|
|
"num_tokens": 187495.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.627774047851563,
|
|
"epoch": 0.005552633117072886,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.5236,
|
|
"mean_token_accuracy": 0.0469218760728836,
|
|
"num_tokens": 197961.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.619970226287842,
|
|
"epoch": 0.00584487696533988,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.4519,
|
|
"mean_token_accuracy": 0.05195427779108286,
|
|
"num_tokens": 209553.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.597136783599854,
|
|
"epoch": 0.006137120813606874,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.3622,
|
|
"mean_token_accuracy": 0.0559786681085825,
|
|
"num_tokens": 220019.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.556843280792236,
|
|
"epoch": 0.006429364661873867,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.2362,
|
|
"mean_token_accuracy": 0.06384762041270733,
|
|
"num_tokens": 231542.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.514682292938232,
|
|
"epoch": 0.006721608510140861,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.1411,
|
|
"mean_token_accuracy": 0.060038824751973155,
|
|
"num_tokens": 242192.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.46776523590088,
|
|
"epoch": 0.007013852358407856,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.0408,
|
|
"mean_token_accuracy": 0.06044495329260826,
|
|
"num_tokens": 252711.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.397676372528077,
|
|
"epoch": 0.00730609620667485,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.958,
|
|
"mean_token_accuracy": 0.05820495039224625,
|
|
"num_tokens": 262768.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.315190315246582,
|
|
"epoch": 0.007598340054941843,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.8458,
|
|
"mean_token_accuracy": 0.06275271475315095,
|
|
"num_tokens": 274374.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.261965942382812,
|
|
"epoch": 0.007890583903208837,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.7538,
|
|
"mean_token_accuracy": 0.06524680852890015,
|
|
"num_tokens": 285449.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.157284355163574,
|
|
"epoch": 0.008182827751475831,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.6492,
|
|
"mean_token_accuracy": 0.06321290582418441,
|
|
"num_tokens": 296785.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.050414943695069,
|
|
"epoch": 0.008475071599742826,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.5334,
|
|
"mean_token_accuracy": 0.06643750704824924,
|
|
"num_tokens": 307781.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.926141262054443,
|
|
"epoch": 0.00876731544800982,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.3954,
|
|
"mean_token_accuracy": 0.06747495792806149,
|
|
"num_tokens": 318713.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.822235202789306,
|
|
"epoch": 0.009059559296276814,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.3951,
|
|
"mean_token_accuracy": 0.05827293880283833,
|
|
"num_tokens": 329779.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.66906967163086,
|
|
"epoch": 0.009351803144543808,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.2444,
|
|
"mean_token_accuracy": 0.0628633838146925,
|
|
"num_tokens": 339733.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.44699068069458,
|
|
"epoch": 0.009644046992810802,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.1742,
|
|
"mean_token_accuracy": 0.06834175810217857,
|
|
"num_tokens": 350232.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.28164119720459,
|
|
"epoch": 0.009936290841077795,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.1224,
|
|
"mean_token_accuracy": 0.06834722384810447,
|
|
"num_tokens": 361051.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.148957920074462,
|
|
"epoch": 0.010228534689344789,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.0815,
|
|
"mean_token_accuracy": 0.0646477747708559,
|
|
"num_tokens": 371699.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.94444236755371,
|
|
"epoch": 0.010520778537611783,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.9754,
|
|
"mean_token_accuracy": 0.06804421916604042,
|
|
"num_tokens": 381817.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.812028980255127,
|
|
"epoch": 0.010813022385878777,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.9774,
|
|
"mean_token_accuracy": 0.06625252738595008,
|
|
"num_tokens": 393788.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.69484453201294,
|
|
"epoch": 0.011105266234145771,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.8944,
|
|
"mean_token_accuracy": 0.07389901392161846,
|
|
"num_tokens": 404690.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.55163288116455,
|
|
"epoch": 0.011397510082412765,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.855,
|
|
"mean_token_accuracy": 0.06855227462947369,
|
|
"num_tokens": 414627.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.51447925567627,
|
|
"epoch": 0.01168975393067976,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.9112,
|
|
"mean_token_accuracy": 0.06506339274346828,
|
|
"num_tokens": 425523.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.489240837097167,
|
|
"epoch": 0.011981997778946754,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.8654,
|
|
"mean_token_accuracy": 0.06927913017570972,
|
|
"num_tokens": 436340.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.421260166168214,
|
|
"epoch": 0.012274241627213748,
|
|
"grad_norm": 0.6953125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.875,
|
|
"mean_token_accuracy": 0.07178668864071369,
|
|
"num_tokens": 447677.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.410223293304444,
|
|
"epoch": 0.01256648547548074,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.852,
|
|
"mean_token_accuracy": 0.06873307004570961,
|
|
"num_tokens": 457470.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.345960903167725,
|
|
"epoch": 0.012858729323747735,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.841,
|
|
"mean_token_accuracy": 0.06865699663758278,
|
|
"num_tokens": 467056.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.390186595916749,
|
|
"epoch": 0.013150973172014729,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.8266,
|
|
"mean_token_accuracy": 0.07396755889058113,
|
|
"num_tokens": 476964.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.322624683380127,
|
|
"epoch": 0.013443217020281723,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.7907,
|
|
"mean_token_accuracy": 0.07042324244976043,
|
|
"num_tokens": 488273.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.322233295440673,
|
|
"epoch": 0.013735460868548717,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.8396,
|
|
"mean_token_accuracy": 0.06669306643307209,
|
|
"num_tokens": 499188.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.32526388168335,
|
|
"epoch": 0.014027704716815711,
|
|
"grad_norm": 0.6796875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.7654,
|
|
"mean_token_accuracy": 0.07240047603845597,
|
|
"num_tokens": 509750.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.2788516998291,
|
|
"epoch": 0.014319948565082705,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.7089,
|
|
"mean_token_accuracy": 0.07294566892087459,
|
|
"num_tokens": 519680.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.086223363876343,
|
|
"epoch": 0.0146121924133497,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.8038,
|
|
"mean_token_accuracy": 0.07106421366333962,
|
|
"num_tokens": 531028.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.35772066116333,
|
|
"epoch": 0.014904436261616694,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.7634,
|
|
"mean_token_accuracy": 0.07067177146673202,
|
|
"num_tokens": 540911.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.142296409606933,
|
|
"epoch": 0.015196680109883686,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.7239,
|
|
"mean_token_accuracy": 0.07250927239656449,
|
|
"num_tokens": 551217.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.26462230682373,
|
|
"epoch": 0.01548892395815068,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.7371,
|
|
"mean_token_accuracy": 0.06948361918330193,
|
|
"num_tokens": 560741.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.203502750396728,
|
|
"epoch": 0.015781167806417674,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.7069,
|
|
"mean_token_accuracy": 0.07292059324681759,
|
|
"num_tokens": 571004.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.08554220199585,
|
|
"epoch": 0.01607341165468467,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.6602,
|
|
"mean_token_accuracy": 0.07287065088748931,
|
|
"num_tokens": 581035.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.230623817443847,
|
|
"epoch": 0.016365655502951663,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.7703,
|
|
"mean_token_accuracy": 0.06783514469861984,
|
|
"num_tokens": 591460.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.123512077331544,
|
|
"epoch": 0.016657899351218655,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.7867,
|
|
"mean_token_accuracy": 0.06507540903985501,
|
|
"num_tokens": 602104.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.162607574462891,
|
|
"epoch": 0.01695014319948565,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.7244,
|
|
"mean_token_accuracy": 0.06926270946860313,
|
|
"num_tokens": 613497.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.155611419677735,
|
|
"epoch": 0.017242387047752643,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.7376,
|
|
"mean_token_accuracy": 0.06834047213196755,
|
|
"num_tokens": 623432.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.172423458099365,
|
|
"epoch": 0.01753463089601964,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.72,
|
|
"mean_token_accuracy": 0.07163975387811661,
|
|
"num_tokens": 633890.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.120147132873536,
|
|
"epoch": 0.017826874744286632,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.6996,
|
|
"mean_token_accuracy": 0.07336684912443162,
|
|
"num_tokens": 644020.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.08779067993164,
|
|
"epoch": 0.018119118592553628,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.644,
|
|
"mean_token_accuracy": 0.07471407577395439,
|
|
"num_tokens": 654177.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.040860605239867,
|
|
"epoch": 0.01841136244082062,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.5986,
|
|
"mean_token_accuracy": 0.08076093941926957,
|
|
"num_tokens": 665317.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.058250904083252,
|
|
"epoch": 0.018703606289087616,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.7048,
|
|
"mean_token_accuracy": 0.07110530957579612,
|
|
"num_tokens": 676248.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.157335948944091,
|
|
"epoch": 0.01899585013735461,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.7111,
|
|
"mean_token_accuracy": 0.06896476708352565,
|
|
"num_tokens": 687573.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.00673680305481,
|
|
"epoch": 0.019288093985621604,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.7276,
|
|
"mean_token_accuracy": 0.07006649971008301,
|
|
"num_tokens": 698919.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.120959758758545,
|
|
"epoch": 0.019580337833888597,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.6748,
|
|
"mean_token_accuracy": 0.06717267856001854,
|
|
"num_tokens": 709450.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.110147094726562,
|
|
"epoch": 0.01987258168215559,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.6476,
|
|
"mean_token_accuracy": 0.07495066076517105,
|
|
"num_tokens": 719909.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.031400108337403,
|
|
"epoch": 0.020164825530422585,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.5699,
|
|
"mean_token_accuracy": 0.07540844045579434,
|
|
"num_tokens": 729451.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.043409729003907,
|
|
"epoch": 0.020457069378689578,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.5831,
|
|
"mean_token_accuracy": 0.07626536451280116,
|
|
"num_tokens": 739256.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.9351013660430905,
|
|
"epoch": 0.020749313226956573,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.5958,
|
|
"mean_token_accuracy": 0.07205557823181152,
|
|
"num_tokens": 749690.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.968925666809082,
|
|
"epoch": 0.021041557075223566,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.6478,
|
|
"mean_token_accuracy": 0.0736626997590065,
|
|
"num_tokens": 760665.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.029117727279663,
|
|
"epoch": 0.021333800923490562,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.5629,
|
|
"mean_token_accuracy": 0.07812464088201523,
|
|
"num_tokens": 770506.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.863681507110596,
|
|
"epoch": 0.021626044771757554,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.6357,
|
|
"mean_token_accuracy": 0.07435660734772682,
|
|
"num_tokens": 780939.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.057669687271119,
|
|
"epoch": 0.02191828862002455,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.6061,
|
|
"mean_token_accuracy": 0.0721366185694933,
|
|
"num_tokens": 792336.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.00227198600769,
|
|
"epoch": 0.022210532468291543,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.6259,
|
|
"mean_token_accuracy": 0.07205860875546932,
|
|
"num_tokens": 803357.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.049015951156616,
|
|
"epoch": 0.022502776316558535,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.5829,
|
|
"mean_token_accuracy": 0.07036133743822574,
|
|
"num_tokens": 813964.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.892692565917969,
|
|
"epoch": 0.02279502016482553,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.5666,
|
|
"mean_token_accuracy": 0.07797688394784927,
|
|
"num_tokens": 824042.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.043365049362183,
|
|
"epoch": 0.023087264013092523,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.5108,
|
|
"mean_token_accuracy": 0.08256808444857597,
|
|
"num_tokens": 834080.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.010589456558227,
|
|
"epoch": 0.02337950786135952,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.617,
|
|
"mean_token_accuracy": 0.06876562051475048,
|
|
"num_tokens": 844483.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.925695133209229,
|
|
"epoch": 0.02367175170962651,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.6103,
|
|
"mean_token_accuracy": 0.07202962972223759,
|
|
"num_tokens": 855896.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.904125165939331,
|
|
"epoch": 0.023963995557893508,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.5299,
|
|
"mean_token_accuracy": 0.07748726978898049,
|
|
"num_tokens": 866631.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.941328859329223,
|
|
"epoch": 0.0242562394061605,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.5184,
|
|
"mean_token_accuracy": 0.0782412201166153,
|
|
"num_tokens": 877779.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.92393913269043,
|
|
"epoch": 0.024548483254427496,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.5434,
|
|
"mean_token_accuracy": 0.07804081477224827,
|
|
"num_tokens": 888478.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.941523599624634,
|
|
"epoch": 0.02484072710269449,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.5063,
|
|
"mean_token_accuracy": 0.07843287661671638,
|
|
"num_tokens": 898164.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.827020597457886,
|
|
"epoch": 0.02513297095096148,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.444,
|
|
"mean_token_accuracy": 0.08076250851154328,
|
|
"num_tokens": 908426.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.868816757202149,
|
|
"epoch": 0.025425214799228477,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.5119,
|
|
"mean_token_accuracy": 0.07991167530417442,
|
|
"num_tokens": 919164.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.792380475997925,
|
|
"epoch": 0.02571745864749547,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.4187,
|
|
"mean_token_accuracy": 0.08300766497850418,
|
|
"num_tokens": 931152.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.8588916778564455,
|
|
"epoch": 0.026009702495762465,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.5027,
|
|
"mean_token_accuracy": 0.08102720528841019,
|
|
"num_tokens": 941629.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.851666688919067,
|
|
"epoch": 0.026301946344029457,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.4577,
|
|
"mean_token_accuracy": 0.08078472912311555,
|
|
"num_tokens": 952665.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.90864748954773,
|
|
"epoch": 0.026594190192296453,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.4166,
|
|
"mean_token_accuracy": 0.07709830179810524,
|
|
"num_tokens": 963395.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.783758735656738,
|
|
"epoch": 0.026886434040563446,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.402,
|
|
"mean_token_accuracy": 0.08202188611030578,
|
|
"num_tokens": 972789.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.851816082000733,
|
|
"epoch": 0.02717867788883044,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.4988,
|
|
"mean_token_accuracy": 0.07351949512958526,
|
|
"num_tokens": 984240.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.863189268112182,
|
|
"epoch": 0.027470921737097434,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.4893,
|
|
"mean_token_accuracy": 0.07835234403610229,
|
|
"num_tokens": 994698.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.70669264793396,
|
|
"epoch": 0.027763165585364426,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.4247,
|
|
"mean_token_accuracy": 0.08151589259505272,
|
|
"num_tokens": 1005324.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.905373239517212,
|
|
"epoch": 0.028055409433631422,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.4238,
|
|
"mean_token_accuracy": 0.07885461077094078,
|
|
"num_tokens": 1016587.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.845970726013183,
|
|
"epoch": 0.028347653281898415,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.4595,
|
|
"mean_token_accuracy": 0.07829558476805687,
|
|
"num_tokens": 1026998.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.805553197860718,
|
|
"epoch": 0.02863989713016541,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.4385,
|
|
"mean_token_accuracy": 0.0780658096075058,
|
|
"num_tokens": 1038475.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.88462347984314,
|
|
"epoch": 0.028932140978432403,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.3556,
|
|
"mean_token_accuracy": 0.08420942425727844,
|
|
"num_tokens": 1049631.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.75970048904419,
|
|
"epoch": 0.0292243848266994,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.4304,
|
|
"mean_token_accuracy": 0.07490956112742424,
|
|
"num_tokens": 1060052.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.750205373764038,
|
|
"epoch": 0.02951662867496639,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.3578,
|
|
"mean_token_accuracy": 0.08051066994667053,
|
|
"num_tokens": 1070516.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.676132297515869,
|
|
"epoch": 0.029808872523233387,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.2888,
|
|
"mean_token_accuracy": 0.08652116805315017,
|
|
"num_tokens": 1081521.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.763888645172119,
|
|
"epoch": 0.03010111637150038,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.3703,
|
|
"mean_token_accuracy": 0.08117554560303689,
|
|
"num_tokens": 1092032.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.78075270652771,
|
|
"epoch": 0.030393360219767372,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.344,
|
|
"mean_token_accuracy": 0.07733950912952423,
|
|
"num_tokens": 1103165.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.714309215545654,
|
|
"epoch": 0.030685604068034368,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.3237,
|
|
"mean_token_accuracy": 0.08414753824472428,
|
|
"num_tokens": 1113184.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.6896360397338865,
|
|
"epoch": 0.03097784791630136,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.3481,
|
|
"mean_token_accuracy": 0.07835544645786285,
|
|
"num_tokens": 1123690.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.715307807922363,
|
|
"epoch": 0.03127009176456835,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.3444,
|
|
"mean_token_accuracy": 0.08257319033145905,
|
|
"num_tokens": 1134299.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.731861400604248,
|
|
"epoch": 0.03156233561283535,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.3126,
|
|
"mean_token_accuracy": 0.08477383702993393,
|
|
"num_tokens": 1145966.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.746872615814209,
|
|
"epoch": 0.031854579461102345,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.4176,
|
|
"mean_token_accuracy": 0.08028875067830085,
|
|
"num_tokens": 1156999.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.7179771900177006,
|
|
"epoch": 0.03214682330936934,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.3017,
|
|
"mean_token_accuracy": 0.08542975857853889,
|
|
"num_tokens": 1167420.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.68302435874939,
|
|
"epoch": 0.03243906715763633,
|
|
"grad_norm": 3.28125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.268,
|
|
"mean_token_accuracy": 0.08016745746135712,
|
|
"num_tokens": 1179271.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.762677955627441,
|
|
"epoch": 0.032731311005903325,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.3496,
|
|
"mean_token_accuracy": 0.08063844367861747,
|
|
"num_tokens": 1189881.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.573287487030029,
|
|
"epoch": 0.03302355485417032,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.246,
|
|
"mean_token_accuracy": 0.08935716524720191,
|
|
"num_tokens": 1200880.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.696466159820557,
|
|
"epoch": 0.03331579870243731,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.2508,
|
|
"mean_token_accuracy": 0.08653135001659393,
|
|
"num_tokens": 1211455.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.651762437820435,
|
|
"epoch": 0.033608042550704306,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.2974,
|
|
"mean_token_accuracy": 0.08382403329014779,
|
|
"num_tokens": 1222675.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.634721422195435,
|
|
"epoch": 0.0339002863989713,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.2447,
|
|
"mean_token_accuracy": 0.08185268826782703,
|
|
"num_tokens": 1233254.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.682222318649292,
|
|
"epoch": 0.0341925302472383,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.2589,
|
|
"mean_token_accuracy": 0.08581831827759742,
|
|
"num_tokens": 1243724.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.732604122161865,
|
|
"epoch": 0.03448477409550529,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.2926,
|
|
"mean_token_accuracy": 0.08690692782402039,
|
|
"num_tokens": 1253982.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.568682384490967,
|
|
"epoch": 0.03477701794377228,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.2482,
|
|
"mean_token_accuracy": 0.08756501823663712,
|
|
"num_tokens": 1265278.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.584898567199707,
|
|
"epoch": 0.03506926179203928,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.2001,
|
|
"mean_token_accuracy": 0.08569748774170875,
|
|
"num_tokens": 1276582.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.657285356521607,
|
|
"epoch": 0.035361505640306275,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.2281,
|
|
"mean_token_accuracy": 0.08484720215201377,
|
|
"num_tokens": 1286556.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.627065277099609,
|
|
"epoch": 0.035653749488573264,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.2133,
|
|
"mean_token_accuracy": 0.08658791929483414,
|
|
"num_tokens": 1296558.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.507645845413208,
|
|
"epoch": 0.03594599333684026,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.2109,
|
|
"mean_token_accuracy": 0.08499465212225914,
|
|
"num_tokens": 1307194.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.747908115386963,
|
|
"epoch": 0.036238237185107255,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.2899,
|
|
"mean_token_accuracy": 0.08115014433860779,
|
|
"num_tokens": 1317171.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.540807151794434,
|
|
"epoch": 0.036530481033374244,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.187,
|
|
"mean_token_accuracy": 0.0856039360165596,
|
|
"num_tokens": 1327867.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.596679210662842,
|
|
"epoch": 0.03682272488164124,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.1396,
|
|
"mean_token_accuracy": 0.08967974856495857,
|
|
"num_tokens": 1338589.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.481625366210937,
|
|
"epoch": 0.037114968729908236,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.2229,
|
|
"mean_token_accuracy": 0.08567533791065216,
|
|
"num_tokens": 1350732.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.6011754989624025,
|
|
"epoch": 0.03740721257817523,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.1723,
|
|
"mean_token_accuracy": 0.0867012806236744,
|
|
"num_tokens": 1361166.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.567913913726807,
|
|
"epoch": 0.03769945642644222,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.2163,
|
|
"mean_token_accuracy": 0.08414383605122566,
|
|
"num_tokens": 1372026.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.545151472091675,
|
|
"epoch": 0.03799170027470922,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.1912,
|
|
"mean_token_accuracy": 0.08693999275565148,
|
|
"num_tokens": 1382727.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.538869190216064,
|
|
"epoch": 0.03828394412297621,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.1684,
|
|
"mean_token_accuracy": 0.08681568801403046,
|
|
"num_tokens": 1393881.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.494607639312744,
|
|
"epoch": 0.03857618797124321,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.1153,
|
|
"mean_token_accuracy": 0.09281730018556118,
|
|
"num_tokens": 1404430.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.494344902038574,
|
|
"epoch": 0.0388684318195102,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.1512,
|
|
"mean_token_accuracy": 0.08740939907729625,
|
|
"num_tokens": 1415195.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.641942262649536,
|
|
"epoch": 0.039160675667777194,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.1686,
|
|
"mean_token_accuracy": 0.09041328504681587,
|
|
"num_tokens": 1425604.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.415121173858642,
|
|
"epoch": 0.03945291951604419,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.1303,
|
|
"mean_token_accuracy": 0.08791615143418312,
|
|
"num_tokens": 1435641.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.5405020236969,
|
|
"epoch": 0.03974516336431118,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.1614,
|
|
"mean_token_accuracy": 0.08788530007004738,
|
|
"num_tokens": 1446355.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.481500291824341,
|
|
"epoch": 0.040037407212578174,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.1119,
|
|
"mean_token_accuracy": 0.09073843285441399,
|
|
"num_tokens": 1456837.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.4543869972229,
|
|
"epoch": 0.04032965106084517,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.1028,
|
|
"mean_token_accuracy": 0.09097715318202973,
|
|
"num_tokens": 1467382.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.397175216674805,
|
|
"epoch": 0.040621894909112166,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.0417,
|
|
"mean_token_accuracy": 0.10079705789685249,
|
|
"num_tokens": 1479689.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.5507731437683105,
|
|
"epoch": 0.040914138757379155,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.1699,
|
|
"mean_token_accuracy": 0.08496656790375709,
|
|
"num_tokens": 1490284.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.476789808273315,
|
|
"epoch": 0.04120638260564615,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.0216,
|
|
"mean_token_accuracy": 0.09127195402979851,
|
|
"num_tokens": 1500991.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.356255292892456,
|
|
"epoch": 0.04149862645391315,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.0435,
|
|
"mean_token_accuracy": 0.09375435337424279,
|
|
"num_tokens": 1512023.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.4905274391174315,
|
|
"epoch": 0.041790870302180136,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.0935,
|
|
"mean_token_accuracy": 0.09124578461050988,
|
|
"num_tokens": 1522304.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.34208402633667,
|
|
"epoch": 0.04208311415044713,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.0737,
|
|
"mean_token_accuracy": 0.08837040141224861,
|
|
"num_tokens": 1533071.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.4827569961547855,
|
|
"epoch": 0.04237535799871413,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.056,
|
|
"mean_token_accuracy": 0.0910212941467762,
|
|
"num_tokens": 1543506.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.390641450881958,
|
|
"epoch": 0.042667601846981124,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.9943,
|
|
"mean_token_accuracy": 0.09501200690865516,
|
|
"num_tokens": 1553311.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.392716789245606,
|
|
"epoch": 0.04295984569524811,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.0216,
|
|
"mean_token_accuracy": 0.09818659871816635,
|
|
"num_tokens": 1563563.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.349376201629639,
|
|
"epoch": 0.04325208954351511,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.0553,
|
|
"mean_token_accuracy": 0.09552945345640182,
|
|
"num_tokens": 1574665.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.364068841934204,
|
|
"epoch": 0.043544333391782104,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.0336,
|
|
"mean_token_accuracy": 0.09296244904398918,
|
|
"num_tokens": 1585497.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.4066088676452635,
|
|
"epoch": 0.0438365772400491,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.0484,
|
|
"mean_token_accuracy": 0.09101586267352105,
|
|
"num_tokens": 1597024.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.408521699905395,
|
|
"epoch": 0.04412882108831609,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.0568,
|
|
"mean_token_accuracy": 0.09450711980462075,
|
|
"num_tokens": 1607722.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.317360258102417,
|
|
"epoch": 0.044421064936583085,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.0671,
|
|
"mean_token_accuracy": 0.09410665109753609,
|
|
"num_tokens": 1619347.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.3950025081634525,
|
|
"epoch": 0.04471330878485008,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.9465,
|
|
"mean_token_accuracy": 0.10844082832336426,
|
|
"num_tokens": 1629580.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.3762188911437985,
|
|
"epoch": 0.04500555263311707,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.0156,
|
|
"mean_token_accuracy": 0.09395859688520432,
|
|
"num_tokens": 1639380.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.337083148956299,
|
|
"epoch": 0.045297796481384066,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.9648,
|
|
"mean_token_accuracy": 0.10094484835863113,
|
|
"num_tokens": 1649984.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.357005643844604,
|
|
"epoch": 0.04559004032965106,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.04,
|
|
"mean_token_accuracy": 0.0935479886829853,
|
|
"num_tokens": 1660396.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.4329475402832035,
|
|
"epoch": 0.04588228417791806,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.0673,
|
|
"mean_token_accuracy": 0.09372782036662101,
|
|
"num_tokens": 1671541.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.288609457015991,
|
|
"epoch": 0.04617452802618505,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.893,
|
|
"mean_token_accuracy": 0.10022989436984062,
|
|
"num_tokens": 1681831.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.254759979248047,
|
|
"epoch": 0.04646677187445204,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.9837,
|
|
"mean_token_accuracy": 0.09495187327265739,
|
|
"num_tokens": 1691971.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.211888647079467,
|
|
"epoch": 0.04675901572271904,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.9761,
|
|
"mean_token_accuracy": 0.09160351902246475,
|
|
"num_tokens": 1703112.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.284194564819336,
|
|
"epoch": 0.04705125957098603,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.9933,
|
|
"mean_token_accuracy": 0.09874650239944457,
|
|
"num_tokens": 1714431.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.3689024448394775,
|
|
"epoch": 0.04734350341925302,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.013,
|
|
"mean_token_accuracy": 0.09238667488098144,
|
|
"num_tokens": 1725089.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.233988809585571,
|
|
"epoch": 0.04763574726752002,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.9197,
|
|
"mean_token_accuracy": 0.09475543424487114,
|
|
"num_tokens": 1736100.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.245414113998413,
|
|
"epoch": 0.047927991115787015,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.9406,
|
|
"mean_token_accuracy": 0.09858621507883072,
|
|
"num_tokens": 1745555.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.255463647842407,
|
|
"epoch": 0.048220234964054004,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.909,
|
|
"mean_token_accuracy": 0.10349308103322982,
|
|
"num_tokens": 1755741.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.341704893112182,
|
|
"epoch": 0.048512478812321,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.9909,
|
|
"mean_token_accuracy": 0.09519274383783341,
|
|
"num_tokens": 1766302.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.243224716186523,
|
|
"epoch": 0.048804722660587996,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.038,
|
|
"mean_token_accuracy": 0.095157690346241,
|
|
"num_tokens": 1778512.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.2741392135620115,
|
|
"epoch": 0.04909696650885499,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.9546,
|
|
"mean_token_accuracy": 0.09773377999663353,
|
|
"num_tokens": 1790279.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.1886015892028805,
|
|
"epoch": 0.04938921035712198,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.8582,
|
|
"mean_token_accuracy": 0.1006741963326931,
|
|
"num_tokens": 1801602.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.210630559921265,
|
|
"epoch": 0.04968145420538898,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.9407,
|
|
"mean_token_accuracy": 0.09713189452886581,
|
|
"num_tokens": 1812456.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.222310018539429,
|
|
"epoch": 0.04997369805365597,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.8839,
|
|
"mean_token_accuracy": 0.09924520403146744,
|
|
"num_tokens": 1822590.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.2720434188842775,
|
|
"epoch": 0.05026594190192296,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.98,
|
|
"mean_token_accuracy": 0.09584357738494872,
|
|
"num_tokens": 1833937.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.218048000335694,
|
|
"epoch": 0.05055818575018996,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.9253,
|
|
"mean_token_accuracy": 0.09750764816999435,
|
|
"num_tokens": 1844190.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.174602031707764,
|
|
"epoch": 0.05085042959845695,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.9257,
|
|
"mean_token_accuracy": 0.104092937707901,
|
|
"num_tokens": 1855337.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.183422327041626,
|
|
"epoch": 0.05114267344672395,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.8805,
|
|
"mean_token_accuracy": 0.10292773395776748,
|
|
"num_tokens": 1866098.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.173034286499023,
|
|
"epoch": 0.05143491729499094,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.8788,
|
|
"mean_token_accuracy": 0.09948200657963753,
|
|
"num_tokens": 1877001.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.238340282440186,
|
|
"epoch": 0.051727161143257934,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.9253,
|
|
"mean_token_accuracy": 0.09718706980347633,
|
|
"num_tokens": 1887507.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.206234312057495,
|
|
"epoch": 0.05201940499152493,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.8844,
|
|
"mean_token_accuracy": 0.09948105290532112,
|
|
"num_tokens": 1897638.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.217797946929932,
|
|
"epoch": 0.052311648839791926,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.8906,
|
|
"mean_token_accuracy": 0.09587719589471817,
|
|
"num_tokens": 1908572.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.146448469161987,
|
|
"epoch": 0.052603892688058915,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.9014,
|
|
"mean_token_accuracy": 0.09291641935706138,
|
|
"num_tokens": 1918537.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.208133697509766,
|
|
"epoch": 0.05289613653632591,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.793,
|
|
"mean_token_accuracy": 0.10730788707733155,
|
|
"num_tokens": 1929307.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.119743871688843,
|
|
"epoch": 0.053188380384592907,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.8879,
|
|
"mean_token_accuracy": 0.10149016827344895,
|
|
"num_tokens": 1940664.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.115985631942749,
|
|
"epoch": 0.053480624232859895,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.7875,
|
|
"mean_token_accuracy": 0.10395304337143899,
|
|
"num_tokens": 1951539.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.2284361839294435,
|
|
"epoch": 0.05377286808112689,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.9179,
|
|
"mean_token_accuracy": 0.0977307379245758,
|
|
"num_tokens": 1962185.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.103752136230469,
|
|
"epoch": 0.05406511192939389,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.819,
|
|
"mean_token_accuracy": 0.10571394339203835,
|
|
"num_tokens": 1972680.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.096303844451905,
|
|
"epoch": 0.05435735577766088,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.8222,
|
|
"mean_token_accuracy": 0.10604664534330369,
|
|
"num_tokens": 1984223.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.133648777008057,
|
|
"epoch": 0.05464959962592787,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.819,
|
|
"mean_token_accuracy": 0.10592230781912804,
|
|
"num_tokens": 1995814.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.043779277801514,
|
|
"epoch": 0.05494184347419487,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.7906,
|
|
"mean_token_accuracy": 0.10688083916902542,
|
|
"num_tokens": 2006785.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.173277759552002,
|
|
"epoch": 0.055234087322461864,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.9091,
|
|
"mean_token_accuracy": 0.1030313603579998,
|
|
"num_tokens": 2017321.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.240300178527832,
|
|
"epoch": 0.05552633117072885,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.8766,
|
|
"mean_token_accuracy": 0.10060662850737571,
|
|
"num_tokens": 2028209.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.038855934143067,
|
|
"epoch": 0.05581857501899585,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.7648,
|
|
"mean_token_accuracy": 0.1070568285882473,
|
|
"num_tokens": 2039525.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.029396915435791,
|
|
"epoch": 0.056110818867262845,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.7747,
|
|
"mean_token_accuracy": 0.10474479496479035,
|
|
"num_tokens": 2049722.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.040965557098389,
|
|
"epoch": 0.05640306271552984,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.8088,
|
|
"mean_token_accuracy": 0.10178461968898773,
|
|
"num_tokens": 2060393.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.061819076538086,
|
|
"epoch": 0.05669530656379683,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.7823,
|
|
"mean_token_accuracy": 0.10536119341850281,
|
|
"num_tokens": 2071412.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.025031852722168,
|
|
"epoch": 0.056987550412063825,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.8559,
|
|
"mean_token_accuracy": 0.09819397777318954,
|
|
"num_tokens": 2082185.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.146571302413941,
|
|
"epoch": 0.05727979426033082,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.8453,
|
|
"mean_token_accuracy": 0.09954645112156868,
|
|
"num_tokens": 2093513.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.115858840942383,
|
|
"epoch": 0.05757203810859782,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.7571,
|
|
"mean_token_accuracy": 0.10359621495008468,
|
|
"num_tokens": 2103058.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.171190929412842,
|
|
"epoch": 0.057864281956864806,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.8395,
|
|
"mean_token_accuracy": 0.10478886738419532,
|
|
"num_tokens": 2113381.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.013499212265015,
|
|
"epoch": 0.0581565258051318,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.8223,
|
|
"mean_token_accuracy": 0.10343048572540284,
|
|
"num_tokens": 2124292.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.106663417816162,
|
|
"epoch": 0.0584487696533988,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.8417,
|
|
"mean_token_accuracy": 0.10271733924746514,
|
|
"num_tokens": 2134781.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.109879779815674,
|
|
"epoch": 0.05874101350166579,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.7996,
|
|
"mean_token_accuracy": 0.10722792819142342,
|
|
"num_tokens": 2144334.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.970988988876343,
|
|
"epoch": 0.05903325734993278,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.7796,
|
|
"mean_token_accuracy": 0.10693327635526657,
|
|
"num_tokens": 2154132.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.147661733627319,
|
|
"epoch": 0.05932550119819978,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.8103,
|
|
"mean_token_accuracy": 0.10502526685595512,
|
|
"num_tokens": 2164636.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.964215326309204,
|
|
"epoch": 0.059617745046466775,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.7739,
|
|
"mean_token_accuracy": 0.10797730386257172,
|
|
"num_tokens": 2175282.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.077650690078736,
|
|
"epoch": 0.059909988894733764,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.8135,
|
|
"mean_token_accuracy": 0.10404316186904908,
|
|
"num_tokens": 2186043.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.101065444946289,
|
|
"epoch": 0.06020223274300076,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.7654,
|
|
"mean_token_accuracy": 0.10284030437469482,
|
|
"num_tokens": 2195986.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.94724440574646,
|
|
"epoch": 0.060494476591267755,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.7298,
|
|
"mean_token_accuracy": 0.10851018875837326,
|
|
"num_tokens": 2205982.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.063318300247192,
|
|
"epoch": 0.060786720439534744,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.7655,
|
|
"mean_token_accuracy": 0.10834679603576661,
|
|
"num_tokens": 2215845.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.005128240585327,
|
|
"epoch": 0.06107896428780174,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.8032,
|
|
"mean_token_accuracy": 0.10471449047327042,
|
|
"num_tokens": 2226867.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.038025045394898,
|
|
"epoch": 0.061371208136068736,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.8166,
|
|
"mean_token_accuracy": 0.10440648719668388,
|
|
"num_tokens": 2236725.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.032281970977783,
|
|
"epoch": 0.06166345198433573,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.7174,
|
|
"mean_token_accuracy": 0.10135879516601562,
|
|
"num_tokens": 2247094.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.975936508178711,
|
|
"epoch": 0.06195569583260272,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.7476,
|
|
"mean_token_accuracy": 0.10740591585636139,
|
|
"num_tokens": 2257873.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.962172317504883,
|
|
"epoch": 0.06224793968086972,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.5924,
|
|
"mean_token_accuracy": 0.12047782689332961,
|
|
"num_tokens": 2268416.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.9587071418762205,
|
|
"epoch": 0.0625401835291367,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.7671,
|
|
"mean_token_accuracy": 0.10503260344266892,
|
|
"num_tokens": 2280190.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.942209196090698,
|
|
"epoch": 0.0628324273774037,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.6321,
|
|
"mean_token_accuracy": 0.1134841226041317,
|
|
"num_tokens": 2289860.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.873535633087158,
|
|
"epoch": 0.0631246712256707,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.6771,
|
|
"mean_token_accuracy": 0.11035498306155204,
|
|
"num_tokens": 2299846.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.00258731842041,
|
|
"epoch": 0.0634169150739377,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.6881,
|
|
"mean_token_accuracy": 0.10880518928170205,
|
|
"num_tokens": 2310284.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.9021313190460205,
|
|
"epoch": 0.06370915892220469,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.7233,
|
|
"mean_token_accuracy": 0.11462967023253441,
|
|
"num_tokens": 2320268.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.942895889282227,
|
|
"epoch": 0.06400140277047169,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.7369,
|
|
"mean_token_accuracy": 0.11447333618998527,
|
|
"num_tokens": 2331739.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.974298715591431,
|
|
"epoch": 0.06429364661873868,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.7475,
|
|
"mean_token_accuracy": 0.11269782036542893,
|
|
"num_tokens": 2342597.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.936334705352783,
|
|
"epoch": 0.06458589046700566,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.7078,
|
|
"mean_token_accuracy": 0.10611579194664955,
|
|
"num_tokens": 2353875.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.9239002704620365,
|
|
"epoch": 0.06487813431527266,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.6635,
|
|
"mean_token_accuracy": 0.10749338716268539,
|
|
"num_tokens": 2365357.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.850267648696899,
|
|
"epoch": 0.06517037816353966,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.628,
|
|
"mean_token_accuracy": 0.11054924726486207,
|
|
"num_tokens": 2375010.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.882474184036255,
|
|
"epoch": 0.06546262201180665,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.6789,
|
|
"mean_token_accuracy": 0.10820299237966538,
|
|
"num_tokens": 2385519.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.889027214050293,
|
|
"epoch": 0.06575486586007365,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.6862,
|
|
"mean_token_accuracy": 0.11094608381390572,
|
|
"num_tokens": 2396761.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.0117217063903805,
|
|
"epoch": 0.06604710970834064,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.6488,
|
|
"mean_token_accuracy": 0.11947292312979699,
|
|
"num_tokens": 2407419.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.862841606140137,
|
|
"epoch": 0.06633935355660764,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.6722,
|
|
"mean_token_accuracy": 0.10760655030608177,
|
|
"num_tokens": 2419133.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.965993213653564,
|
|
"epoch": 0.06663159740487462,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.685,
|
|
"mean_token_accuracy": 0.11355137154459953,
|
|
"num_tokens": 2429589.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.851849794387817,
|
|
"epoch": 0.06692384125314162,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.5958,
|
|
"mean_token_accuracy": 0.11682025790214538,
|
|
"num_tokens": 2440922.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.874046993255615,
|
|
"epoch": 0.06721608510140861,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.687,
|
|
"mean_token_accuracy": 0.10524867624044418,
|
|
"num_tokens": 2452131.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.846616935729981,
|
|
"epoch": 0.06750832894967561,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.6188,
|
|
"mean_token_accuracy": 0.11801559552550316,
|
|
"num_tokens": 2461350.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.870991897583008,
|
|
"epoch": 0.0678005727979426,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.6404,
|
|
"mean_token_accuracy": 0.11024663522839546,
|
|
"num_tokens": 2472131.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.840404462814331,
|
|
"epoch": 0.0680928166462096,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.6472,
|
|
"mean_token_accuracy": 0.11330607905983925,
|
|
"num_tokens": 2482104.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.849715948104858,
|
|
"epoch": 0.0683850604944766,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.6573,
|
|
"mean_token_accuracy": 0.11178938299417496,
|
|
"num_tokens": 2492398.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.962659215927124,
|
|
"epoch": 0.06867730434274359,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.6426,
|
|
"mean_token_accuracy": 0.10926170796155929,
|
|
"num_tokens": 2503154.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.855165195465088,
|
|
"epoch": 0.06896954819101057,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.7251,
|
|
"mean_token_accuracy": 0.10576548874378204,
|
|
"num_tokens": 2514003.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.911726379394532,
|
|
"epoch": 0.06926179203927757,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.6173,
|
|
"mean_token_accuracy": 0.11194900646805764,
|
|
"num_tokens": 2523503.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.926169490814209,
|
|
"epoch": 0.06955403588754457,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.744,
|
|
"mean_token_accuracy": 0.11214761435985565,
|
|
"num_tokens": 2534220.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.81519570350647,
|
|
"epoch": 0.06984627973581156,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.6087,
|
|
"mean_token_accuracy": 0.11375177204608918,
|
|
"num_tokens": 2544746.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.817518424987793,
|
|
"epoch": 0.07013852358407856,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.631,
|
|
"mean_token_accuracy": 0.11508884876966477,
|
|
"num_tokens": 2556154.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.984947299957275,
|
|
"epoch": 0.07043076743234555,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.6979,
|
|
"mean_token_accuracy": 0.11085271090269089,
|
|
"num_tokens": 2566600.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.798012971878052,
|
|
"epoch": 0.07072301128061255,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.5785,
|
|
"mean_token_accuracy": 0.11072144955396652,
|
|
"num_tokens": 2576374.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.940262937545777,
|
|
"epoch": 0.07101525512887953,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.6457,
|
|
"mean_token_accuracy": 0.11224333122372628,
|
|
"num_tokens": 2587299.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.823904705047608,
|
|
"epoch": 0.07130749897714653,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.5806,
|
|
"mean_token_accuracy": 0.11313971057534218,
|
|
"num_tokens": 2598750.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.860695171356201,
|
|
"epoch": 0.07159974282541352,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.5857,
|
|
"mean_token_accuracy": 0.1150270365178585,
|
|
"num_tokens": 2609499.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.810735893249512,
|
|
"epoch": 0.07189198667368052,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.5096,
|
|
"mean_token_accuracy": 0.11989947631955147,
|
|
"num_tokens": 2619673.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.825433397293091,
|
|
"epoch": 0.07218423052194751,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.5906,
|
|
"mean_token_accuracy": 0.11345770210027695,
|
|
"num_tokens": 2629940.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.777529716491699,
|
|
"epoch": 0.07247647437021451,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.5867,
|
|
"mean_token_accuracy": 0.11988462135195732,
|
|
"num_tokens": 2640687.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.786758613586426,
|
|
"epoch": 0.0727687182184815,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.5018,
|
|
"mean_token_accuracy": 0.12359912693500519,
|
|
"num_tokens": 2650662.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.786625051498413,
|
|
"epoch": 0.07306096206674849,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.5444,
|
|
"mean_token_accuracy": 0.12096817344427109,
|
|
"num_tokens": 2661729.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.795501613616944,
|
|
"epoch": 0.07335320591501548,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.5936,
|
|
"mean_token_accuracy": 0.1116180919110775,
|
|
"num_tokens": 2673139.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.7598155498504635,
|
|
"epoch": 0.07364544976328248,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.6019,
|
|
"mean_token_accuracy": 0.118670953810215,
|
|
"num_tokens": 2683578.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.8184223651885985,
|
|
"epoch": 0.07393769361154948,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.5418,
|
|
"mean_token_accuracy": 0.11984733566641807,
|
|
"num_tokens": 2694386.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.820712375640869,
|
|
"epoch": 0.07422993745981647,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.5585,
|
|
"mean_token_accuracy": 0.11072490140795707,
|
|
"num_tokens": 2704806.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.749563884735108,
|
|
"epoch": 0.07452218130808347,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.551,
|
|
"mean_token_accuracy": 0.11777527481317521,
|
|
"num_tokens": 2715262.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.865103101730346,
|
|
"epoch": 0.07481442515635046,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.5641,
|
|
"mean_token_accuracy": 0.11636885330080986,
|
|
"num_tokens": 2726158.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.731928920745849,
|
|
"epoch": 0.07510666900461745,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.5509,
|
|
"mean_token_accuracy": 0.1174699142575264,
|
|
"num_tokens": 2736553.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.80519118309021,
|
|
"epoch": 0.07539891285288444,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.5062,
|
|
"mean_token_accuracy": 0.12198017165064812,
|
|
"num_tokens": 2746842.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.70200023651123,
|
|
"epoch": 0.07569115670115144,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.5264,
|
|
"mean_token_accuracy": 0.11584797650575637,
|
|
"num_tokens": 2756976.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.736905813217163,
|
|
"epoch": 0.07598340054941843,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.5953,
|
|
"mean_token_accuracy": 0.1133107990026474,
|
|
"num_tokens": 2769278.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.882327318191528,
|
|
"epoch": 0.07627564439768543,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.5397,
|
|
"mean_token_accuracy": 0.11898713633418083,
|
|
"num_tokens": 2779888.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.664877223968506,
|
|
"epoch": 0.07656788824595243,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.5644,
|
|
"mean_token_accuracy": 0.11823364794254303,
|
|
"num_tokens": 2791403.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.878745174407959,
|
|
"epoch": 0.07686013209421942,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.5908,
|
|
"mean_token_accuracy": 0.11406810358166694,
|
|
"num_tokens": 2801988.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.7007993221282955,
|
|
"epoch": 0.07715237594248642,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.5193,
|
|
"mean_token_accuracy": 0.1187627412378788,
|
|
"num_tokens": 2812614.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.739048719406128,
|
|
"epoch": 0.0774446197907534,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.4688,
|
|
"mean_token_accuracy": 0.12098750770092011,
|
|
"num_tokens": 2822341.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.714245843887329,
|
|
"epoch": 0.0777368636390204,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.4272,
|
|
"mean_token_accuracy": 0.12731068730354309,
|
|
"num_tokens": 2833319.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.747768211364746,
|
|
"epoch": 0.07802910748728739,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.4319,
|
|
"mean_token_accuracy": 0.11928986981511117,
|
|
"num_tokens": 2844688.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.7483831405639645,
|
|
"epoch": 0.07832135133555439,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.5507,
|
|
"mean_token_accuracy": 0.11463962271809577,
|
|
"num_tokens": 2853975.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.797797775268554,
|
|
"epoch": 0.07861359518382138,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.4863,
|
|
"mean_token_accuracy": 0.12166559845209121,
|
|
"num_tokens": 2863231.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.657674551010132,
|
|
"epoch": 0.07890583903208838,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.4512,
|
|
"mean_token_accuracy": 0.12450483366847039,
|
|
"num_tokens": 2874523.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.764217472076416,
|
|
"epoch": 0.07919808288035537,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.462,
|
|
"mean_token_accuracy": 0.11863645762205124,
|
|
"num_tokens": 2884908.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.647523546218872,
|
|
"epoch": 0.07949032672862236,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.5025,
|
|
"mean_token_accuracy": 0.1177070640027523,
|
|
"num_tokens": 2896656.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.677059030532837,
|
|
"epoch": 0.07978257057688935,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.5434,
|
|
"mean_token_accuracy": 0.11750056520104409,
|
|
"num_tokens": 2906569.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.767497634887695,
|
|
"epoch": 0.08007481442515635,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.4349,
|
|
"mean_token_accuracy": 0.1241106316447258,
|
|
"num_tokens": 2916780.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.716397714614868,
|
|
"epoch": 0.08036705827342334,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.4407,
|
|
"mean_token_accuracy": 0.1245139479637146,
|
|
"num_tokens": 2927731.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.6510984897613525,
|
|
"epoch": 0.08065930212169034,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.4914,
|
|
"mean_token_accuracy": 0.11960818469524384,
|
|
"num_tokens": 2938674.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.709840297698975,
|
|
"epoch": 0.08095154596995734,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.13028942197561263,
|
|
"num_tokens": 2948096.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.687193155288696,
|
|
"epoch": 0.08124378981822433,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.4485,
|
|
"mean_token_accuracy": 0.12074602544307708,
|
|
"num_tokens": 2958427.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.690118646621704,
|
|
"epoch": 0.08153603366649131,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.5266,
|
|
"mean_token_accuracy": 0.12162336185574532,
|
|
"num_tokens": 2968260.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.74784836769104,
|
|
"epoch": 0.08182827751475831,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.4324,
|
|
"mean_token_accuracy": 0.12377095222473145,
|
|
"num_tokens": 2978036.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.659767961502075,
|
|
"epoch": 0.0821205213630253,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.4694,
|
|
"mean_token_accuracy": 0.11742118075489998,
|
|
"num_tokens": 2988142.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.694945001602173,
|
|
"epoch": 0.0824127652112923,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.5228,
|
|
"mean_token_accuracy": 0.11679866909980774,
|
|
"num_tokens": 2998846.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.732224845886231,
|
|
"epoch": 0.0827050090595593,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.4709,
|
|
"mean_token_accuracy": 0.11908368766307831,
|
|
"num_tokens": 3009192.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.658540391921997,
|
|
"epoch": 0.0829972529078263,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.4806,
|
|
"mean_token_accuracy": 0.1223271630704403,
|
|
"num_tokens": 3020985.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.777828359603882,
|
|
"epoch": 0.08328949675609329,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.4893,
|
|
"mean_token_accuracy": 0.12406448647379875,
|
|
"num_tokens": 3031105.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.652883434295655,
|
|
"epoch": 0.08358174060436027,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.4921,
|
|
"mean_token_accuracy": 0.11444137990474701,
|
|
"num_tokens": 3043082.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.7215594291687015,
|
|
"epoch": 0.08387398445262727,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.4227,
|
|
"mean_token_accuracy": 0.12754354998469353,
|
|
"num_tokens": 3052951.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.624235820770264,
|
|
"epoch": 0.08416622830089426,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.4615,
|
|
"mean_token_accuracy": 0.12389512956142426,
|
|
"num_tokens": 3063443.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.732293891906738,
|
|
"epoch": 0.08445847214916126,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.4316,
|
|
"mean_token_accuracy": 0.12258192375302315,
|
|
"num_tokens": 3073356.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.644233226776123,
|
|
"epoch": 0.08475071599742826,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.5274,
|
|
"mean_token_accuracy": 0.12170976623892785,
|
|
"num_tokens": 3083482.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.651962184906006,
|
|
"epoch": 0.08504295984569525,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.4405,
|
|
"mean_token_accuracy": 0.12288702800869941,
|
|
"num_tokens": 3094258.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.6918994903564455,
|
|
"epoch": 0.08533520369396225,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.4286,
|
|
"mean_token_accuracy": 0.13141337484121324,
|
|
"num_tokens": 3105585.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.666267776489258,
|
|
"epoch": 0.08562744754222923,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.4369,
|
|
"mean_token_accuracy": 0.12149759083986282,
|
|
"num_tokens": 3116732.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.6255169868469235,
|
|
"epoch": 0.08591969139049623,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.4212,
|
|
"mean_token_accuracy": 0.1260172627866268,
|
|
"num_tokens": 3127628.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.695675086975098,
|
|
"epoch": 0.08621193523876322,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.4009,
|
|
"mean_token_accuracy": 0.13144792318344117,
|
|
"num_tokens": 3139390.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.626928186416626,
|
|
"epoch": 0.08650417908703022,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.4448,
|
|
"mean_token_accuracy": 0.12692544609308243,
|
|
"num_tokens": 3149543.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.6834962368011475,
|
|
"epoch": 0.08679642293529721,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.4452,
|
|
"mean_token_accuracy": 0.12478880658745765,
|
|
"num_tokens": 3159499.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.647976207733154,
|
|
"epoch": 0.08708866678356421,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.4241,
|
|
"mean_token_accuracy": 0.12237094715237617,
|
|
"num_tokens": 3170174.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.664702844619751,
|
|
"epoch": 0.0873809106318312,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.3459,
|
|
"mean_token_accuracy": 0.13025859296321868,
|
|
"num_tokens": 3180821.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.60100302696228,
|
|
"epoch": 0.0876731544800982,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.402,
|
|
"mean_token_accuracy": 0.1265116386115551,
|
|
"num_tokens": 3191166.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.6825783252716064,
|
|
"epoch": 0.08796539832836518,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.4261,
|
|
"mean_token_accuracy": 0.11982446685433387,
|
|
"num_tokens": 3202072.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.590072822570801,
|
|
"epoch": 0.08825764217663218,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.3761,
|
|
"mean_token_accuracy": 0.12612907961010933,
|
|
"num_tokens": 3211779.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.652530908584595,
|
|
"epoch": 0.08854988602489917,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.4482,
|
|
"mean_token_accuracy": 0.11996340826153755,
|
|
"num_tokens": 3222763.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.613873863220215,
|
|
"epoch": 0.08884212987316617,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.4019,
|
|
"mean_token_accuracy": 0.12952780574560166,
|
|
"num_tokens": 3232434.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.633971309661865,
|
|
"epoch": 0.08913437372143317,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.3858,
|
|
"mean_token_accuracy": 0.12593087404966355,
|
|
"num_tokens": 3243040.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.633696794509888,
|
|
"epoch": 0.08942661756970016,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.3209,
|
|
"mean_token_accuracy": 0.1274397224187851,
|
|
"num_tokens": 3252926.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.603791093826294,
|
|
"epoch": 0.08971886141796716,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.4487,
|
|
"mean_token_accuracy": 0.12076061442494393,
|
|
"num_tokens": 3264764.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.689486312866211,
|
|
"epoch": 0.09001110526623414,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.4377,
|
|
"mean_token_accuracy": 0.12479345500469208,
|
|
"num_tokens": 3275399.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.62721815109253,
|
|
"epoch": 0.09030334911450114,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.4017,
|
|
"mean_token_accuracy": 0.12434872388839721,
|
|
"num_tokens": 3285530.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.609357786178589,
|
|
"epoch": 0.09059559296276813,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.3888,
|
|
"mean_token_accuracy": 0.12555827125906943,
|
|
"num_tokens": 3296737.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.558108901977539,
|
|
"epoch": 0.09088783681103513,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.3645,
|
|
"mean_token_accuracy": 0.1258600175380707,
|
|
"num_tokens": 3307161.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.612364006042481,
|
|
"epoch": 0.09118008065930212,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.3147,
|
|
"mean_token_accuracy": 0.13195257112383843,
|
|
"num_tokens": 3319639.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.510131216049194,
|
|
"epoch": 0.09147232450756912,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.4048,
|
|
"mean_token_accuracy": 0.12524430304765702,
|
|
"num_tokens": 3329910.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.781470203399659,
|
|
"epoch": 0.09176456835583612,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 6.4516,
|
|
"mean_token_accuracy": 0.12193214818835259,
|
|
"num_tokens": 3340235.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.569652318954468,
|
|
"epoch": 0.0920568122041031,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.3401,
|
|
"mean_token_accuracy": 0.12737754583358765,
|
|
"num_tokens": 3350422.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.578609991073608,
|
|
"epoch": 0.0923490560523701,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 6.3491,
|
|
"mean_token_accuracy": 0.12496926188468933,
|
|
"num_tokens": 3361151.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.601184511184693,
|
|
"epoch": 0.09264129990063709,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.3571,
|
|
"mean_token_accuracy": 0.12629662379622458,
|
|
"num_tokens": 3372008.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.5719188213348385,
|
|
"epoch": 0.09293354374890408,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.3238,
|
|
"mean_token_accuracy": 0.1277265168726444,
|
|
"num_tokens": 3383257.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.522926998138428,
|
|
"epoch": 0.09322578759717108,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.383,
|
|
"mean_token_accuracy": 0.12176450192928315,
|
|
"num_tokens": 3394546.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.633662509918213,
|
|
"epoch": 0.09351803144543808,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 6.34,
|
|
"mean_token_accuracy": 0.12867723405361176,
|
|
"num_tokens": 3404822.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.62365984916687,
|
|
"epoch": 0.09381027529370507,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 6.2563,
|
|
"mean_token_accuracy": 0.1342356614768505,
|
|
"num_tokens": 3415409.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.538460159301758,
|
|
"epoch": 0.09410251914197205,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.4488,
|
|
"mean_token_accuracy": 0.1176258459687233,
|
|
"num_tokens": 3427058.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.667861986160278,
|
|
"epoch": 0.09439476299023905,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.2981,
|
|
"mean_token_accuracy": 0.1294393613934517,
|
|
"num_tokens": 3437752.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.535409927368164,
|
|
"epoch": 0.09468700683850605,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.3298,
|
|
"mean_token_accuracy": 0.12445673272013665,
|
|
"num_tokens": 3447951.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.557067728042602,
|
|
"epoch": 0.09497925068677304,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.315,
|
|
"mean_token_accuracy": 0.13457685336470604,
|
|
"num_tokens": 3458981.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.5102777004241945,
|
|
"epoch": 0.09527149453504004,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.2908,
|
|
"mean_token_accuracy": 0.13025625869631768,
|
|
"num_tokens": 3469686.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.639285945892334,
|
|
"epoch": 0.09556373838330703,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.3604,
|
|
"mean_token_accuracy": 0.1277376592159271,
|
|
"num_tokens": 3481051.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.542368793487549,
|
|
"epoch": 0.09585598223157403,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.3951,
|
|
"mean_token_accuracy": 0.12594515830278397,
|
|
"num_tokens": 3491293.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.588239955902099,
|
|
"epoch": 0.09614822607984103,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 6.2757,
|
|
"mean_token_accuracy": 0.13865521401166916,
|
|
"num_tokens": 3501588.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.510730791091919,
|
|
"epoch": 0.09644046992810801,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.3238,
|
|
"mean_token_accuracy": 0.1317336842417717,
|
|
"num_tokens": 3511400.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.536106920242309,
|
|
"epoch": 0.096732713776375,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 6.3036,
|
|
"mean_token_accuracy": 0.1279335044324398,
|
|
"num_tokens": 3522328.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.496180248260498,
|
|
"epoch": 0.097024957624642,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.3877,
|
|
"mean_token_accuracy": 0.12443601489067077,
|
|
"num_tokens": 3533894.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.60569806098938,
|
|
"epoch": 0.097317201472909,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.3395,
|
|
"mean_token_accuracy": 0.12905040979385377,
|
|
"num_tokens": 3545037.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.533877229690551,
|
|
"epoch": 0.09760944532117599,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.2669,
|
|
"mean_token_accuracy": 0.1287287577986717,
|
|
"num_tokens": 3555342.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.511089706420899,
|
|
"epoch": 0.09790168916944299,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.2979,
|
|
"mean_token_accuracy": 0.1334322690963745,
|
|
"num_tokens": 3565697.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.60947060585022,
|
|
"epoch": 0.09819393301770998,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 6.3702,
|
|
"mean_token_accuracy": 0.12965244501829148,
|
|
"num_tokens": 3577563.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.5115879535675045,
|
|
"epoch": 0.09848617686597697,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 6.4125,
|
|
"mean_token_accuracy": 0.13131027817726135,
|
|
"num_tokens": 3587441.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.546698999404907,
|
|
"epoch": 0.09877842071424396,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.3067,
|
|
"mean_token_accuracy": 0.13236589655280112,
|
|
"num_tokens": 3598266.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.6162333488464355,
|
|
"epoch": 0.09907066456251096,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.3046,
|
|
"mean_token_accuracy": 0.1278063587844372,
|
|
"num_tokens": 3609032.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.577450227737427,
|
|
"epoch": 0.09936290841077795,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.4108,
|
|
"mean_token_accuracy": 0.12083216980099679,
|
|
"num_tokens": 3620132.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.537747097015381,
|
|
"epoch": 0.09965515225904495,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.2533,
|
|
"mean_token_accuracy": 0.14148433282971382,
|
|
"num_tokens": 3629802.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.495581102371216,
|
|
"epoch": 0.09994739610731194,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 6.3484,
|
|
"mean_token_accuracy": 0.1262526549398899,
|
|
"num_tokens": 3640183.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.59174656867981,
|
|
"epoch": 0.10023963995557894,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.2379,
|
|
"mean_token_accuracy": 0.1428661122918129,
|
|
"num_tokens": 3651913.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.4783319473266605,
|
|
"epoch": 0.10053188380384592,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 6.2727,
|
|
"mean_token_accuracy": 0.13544789254665374,
|
|
"num_tokens": 3661779.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.51190152168274,
|
|
"epoch": 0.10082412765211292,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 6.2916,
|
|
"mean_token_accuracy": 0.13354387730360032,
|
|
"num_tokens": 3672108.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.483732795715332,
|
|
"epoch": 0.10111637150037991,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.2756,
|
|
"mean_token_accuracy": 0.13410299718379975,
|
|
"num_tokens": 3682549.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.520133447647095,
|
|
"epoch": 0.10140861534864691,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.2651,
|
|
"mean_token_accuracy": 0.13530457243323327,
|
|
"num_tokens": 3693360.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.458718729019165,
|
|
"epoch": 0.1017008591969139,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.2081,
|
|
"mean_token_accuracy": 0.13625198900699614,
|
|
"num_tokens": 3704561.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.550940704345703,
|
|
"epoch": 0.1019931030451809,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 6.2991,
|
|
"mean_token_accuracy": 0.12806941121816634,
|
|
"num_tokens": 3714638.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.566270256042481,
|
|
"epoch": 0.1022853468934479,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.1224788896739483,
|
|
"num_tokens": 3726043.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.543083238601684,
|
|
"epoch": 0.10257759074171488,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.2892,
|
|
"mean_token_accuracy": 0.13211994245648384,
|
|
"num_tokens": 3737666.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.486248445510864,
|
|
"epoch": 0.10286983458998188,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.3026,
|
|
"mean_token_accuracy": 0.1367783397436142,
|
|
"num_tokens": 3748928.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.484316396713257,
|
|
"epoch": 0.10316207843824887,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.3185,
|
|
"mean_token_accuracy": 0.13130587711930275,
|
|
"num_tokens": 3759863.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.529352617263794,
|
|
"epoch": 0.10345432228651587,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.2434,
|
|
"mean_token_accuracy": 0.12802275642752647,
|
|
"num_tokens": 3771064.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.467418003082275,
|
|
"epoch": 0.10374656613478286,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.3083,
|
|
"mean_token_accuracy": 0.13164296075701715,
|
|
"num_tokens": 3782925.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.478915071487426,
|
|
"epoch": 0.10403880998304986,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 6.1925,
|
|
"mean_token_accuracy": 0.14120660200715066,
|
|
"num_tokens": 3793538.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.479070043563842,
|
|
"epoch": 0.10433105383131686,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.2666,
|
|
"mean_token_accuracy": 0.1388741835951805,
|
|
"num_tokens": 3804855.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.515535116195679,
|
|
"epoch": 0.10462329767958385,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.2751,
|
|
"mean_token_accuracy": 0.13155153393745422,
|
|
"num_tokens": 3815807.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.459535837173462,
|
|
"epoch": 0.10491554152785083,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 6.2362,
|
|
"mean_token_accuracy": 0.14252902269363404,
|
|
"num_tokens": 3826331.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.498953866958618,
|
|
"epoch": 0.10520778537611783,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.2061,
|
|
"mean_token_accuracy": 0.13486761301755906,
|
|
"num_tokens": 3836401.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.3942736148834225,
|
|
"epoch": 0.10550002922438483,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.2477,
|
|
"mean_token_accuracy": 0.13660379573702813,
|
|
"num_tokens": 3847034.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.53309006690979,
|
|
"epoch": 0.10579227307265182,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 6.2074,
|
|
"mean_token_accuracy": 0.13580649495124816,
|
|
"num_tokens": 3857443.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.519369792938233,
|
|
"epoch": 0.10608451692091882,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 6.2373,
|
|
"mean_token_accuracy": 0.13940913006663322,
|
|
"num_tokens": 3867975.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.377342891693115,
|
|
"epoch": 0.10637676076918581,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.2535,
|
|
"mean_token_accuracy": 0.13145749270915985,
|
|
"num_tokens": 3878701.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.5430989265441895,
|
|
"epoch": 0.10666900461745281,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 6.2915,
|
|
"mean_token_accuracy": 0.12789264768362046,
|
|
"num_tokens": 3889486.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.435117864608765,
|
|
"epoch": 0.10696124846571979,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 6.2334,
|
|
"mean_token_accuracy": 0.13487916439771652,
|
|
"num_tokens": 3900342.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.490340042114258,
|
|
"epoch": 0.10725349231398679,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.3068,
|
|
"mean_token_accuracy": 0.13169697895646096,
|
|
"num_tokens": 3911917.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.497747135162354,
|
|
"epoch": 0.10754573616225378,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.1706,
|
|
"mean_token_accuracy": 0.14493299797177314,
|
|
"num_tokens": 3922381.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.487737464904785,
|
|
"epoch": 0.10783798001052078,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.2368,
|
|
"mean_token_accuracy": 0.13707713410258293,
|
|
"num_tokens": 3933235.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.4321588516235355,
|
|
"epoch": 0.10813022385878777,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 6.2871,
|
|
"mean_token_accuracy": 0.13197886794805527,
|
|
"num_tokens": 3944337.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.553337144851684,
|
|
"epoch": 0.10842246770705477,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.2303,
|
|
"mean_token_accuracy": 0.1394609570503235,
|
|
"num_tokens": 3954104.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.467433643341065,
|
|
"epoch": 0.10871471155532177,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.2813,
|
|
"mean_token_accuracy": 0.1347561240196228,
|
|
"num_tokens": 3965779.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.449726581573486,
|
|
"epoch": 0.10900695540358875,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.1865,
|
|
"mean_token_accuracy": 0.13389561101794242,
|
|
"num_tokens": 3976874.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.509667587280274,
|
|
"epoch": 0.10929919925185574,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 6.2666,
|
|
"mean_token_accuracy": 0.13473353460431098,
|
|
"num_tokens": 3987463.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.440557336807251,
|
|
"epoch": 0.10959144310012274,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.2584,
|
|
"mean_token_accuracy": 0.13597837686538697,
|
|
"num_tokens": 3998218.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.481018161773681,
|
|
"epoch": 0.10988368694838974,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.1727,
|
|
"mean_token_accuracy": 0.13502540439367294,
|
|
"num_tokens": 4009012.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.390183782577514,
|
|
"epoch": 0.11017593079665673,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 6.1638,
|
|
"mean_token_accuracy": 0.143779406696558,
|
|
"num_tokens": 4018561.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.452798843383789,
|
|
"epoch": 0.11046817464492373,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.2398,
|
|
"mean_token_accuracy": 0.13349425047636032,
|
|
"num_tokens": 4029557.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.494185447692871,
|
|
"epoch": 0.11076041849319072,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.208,
|
|
"mean_token_accuracy": 0.13723380044102668,
|
|
"num_tokens": 4039451.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.496969747543335,
|
|
"epoch": 0.1110526623414577,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 6.2248,
|
|
"mean_token_accuracy": 0.13692051470279692,
|
|
"num_tokens": 4049981.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.398998737335205,
|
|
"epoch": 0.1113449061897247,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 6.2453,
|
|
"mean_token_accuracy": 0.13756425604224204,
|
|
"num_tokens": 4060064.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.4625708103179935,
|
|
"epoch": 0.1116371500379917,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.19,
|
|
"mean_token_accuracy": 0.13855075016617774,
|
|
"num_tokens": 4070630.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.389349794387817,
|
|
"epoch": 0.1119293938862587,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 6.1278,
|
|
"mean_token_accuracy": 0.14593113958835602,
|
|
"num_tokens": 4083486.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.421242904663086,
|
|
"epoch": 0.11222163773452569,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.151,
|
|
"mean_token_accuracy": 0.14287944734096528,
|
|
"num_tokens": 4093787.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.442874526977539,
|
|
"epoch": 0.11251388158279269,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.174,
|
|
"mean_token_accuracy": 0.13947273269295693,
|
|
"num_tokens": 4104827.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.444666814804077,
|
|
"epoch": 0.11280612543105968,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.2085,
|
|
"mean_token_accuracy": 0.1391216039657593,
|
|
"num_tokens": 4115837.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.4151026725769045,
|
|
"epoch": 0.11309836927932668,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 6.2417,
|
|
"mean_token_accuracy": 0.1367591217160225,
|
|
"num_tokens": 4126877.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.445105504989624,
|
|
"epoch": 0.11339061312759366,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.2694,
|
|
"mean_token_accuracy": 0.13381358608603477,
|
|
"num_tokens": 4138197.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.53773865699768,
|
|
"epoch": 0.11368285697586065,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.1969,
|
|
"mean_token_accuracy": 0.1340734951198101,
|
|
"num_tokens": 4149800.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.488288116455078,
|
|
"epoch": 0.11397510082412765,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 6.1925,
|
|
"mean_token_accuracy": 0.13274357169866563,
|
|
"num_tokens": 4160463.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.435004329681396,
|
|
"epoch": 0.11426734467239465,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 6.0906,
|
|
"mean_token_accuracy": 0.14487750232219695,
|
|
"num_tokens": 4170460.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.3771653175354,
|
|
"epoch": 0.11455958852066164,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.2655,
|
|
"mean_token_accuracy": 0.13562354519963266,
|
|
"num_tokens": 4181740.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.50157322883606,
|
|
"epoch": 0.11485183236892864,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 6.1981,
|
|
"mean_token_accuracy": 0.14167827740311623,
|
|
"num_tokens": 4192973.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.36492600440979,
|
|
"epoch": 0.11514407621719563,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 6.1844,
|
|
"mean_token_accuracy": 0.1378167189657688,
|
|
"num_tokens": 4202880.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.3981608867645265,
|
|
"epoch": 0.11543632006546262,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 6.1043,
|
|
"mean_token_accuracy": 0.1442752942442894,
|
|
"num_tokens": 4213690.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.361389064788819,
|
|
"epoch": 0.11572856391372961,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 6.1584,
|
|
"mean_token_accuracy": 0.14419703781604767,
|
|
"num_tokens": 4224061.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.43035044670105,
|
|
"epoch": 0.11602080776199661,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.2247,
|
|
"mean_token_accuracy": 0.13586804941296576,
|
|
"num_tokens": 4234874.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.451950883865356,
|
|
"epoch": 0.1163130516102636,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 6.1707,
|
|
"mean_token_accuracy": 0.13686590492725373,
|
|
"num_tokens": 4245595.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.388254690170288,
|
|
"epoch": 0.1166052954585306,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.149,
|
|
"mean_token_accuracy": 0.13962028473615645,
|
|
"num_tokens": 4255432.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.401205730438233,
|
|
"epoch": 0.1168975393067976,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 6.1829,
|
|
"mean_token_accuracy": 0.14039822965860366,
|
|
"num_tokens": 4265800.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.484803485870361,
|
|
"epoch": 0.11718978315506459,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 6.1871,
|
|
"mean_token_accuracy": 0.1321099206805229,
|
|
"num_tokens": 4276748.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.357201814651489,
|
|
"epoch": 0.11748202700333157,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 6.0863,
|
|
"mean_token_accuracy": 0.14674914330244065,
|
|
"num_tokens": 4287724.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.371573257446289,
|
|
"epoch": 0.11777427085159857,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 6.0497,
|
|
"mean_token_accuracy": 0.14962859824299812,
|
|
"num_tokens": 4297467.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.392589378356933,
|
|
"epoch": 0.11806651469986557,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 6.1843,
|
|
"mean_token_accuracy": 0.13824831917881966,
|
|
"num_tokens": 4308280.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.420476722717285,
|
|
"epoch": 0.11835875854813256,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 6.1712,
|
|
"mean_token_accuracy": 0.1369972661137581,
|
|
"num_tokens": 4318727.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.41554741859436,
|
|
"epoch": 0.11865100239639956,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 6.1535,
|
|
"mean_token_accuracy": 0.13764682933688163,
|
|
"num_tokens": 4328787.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.409768676757812,
|
|
"epoch": 0.11894324624466655,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 6.1493,
|
|
"mean_token_accuracy": 0.1398159608244896,
|
|
"num_tokens": 4339659.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.461029767990112,
|
|
"epoch": 0.11923549009293355,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 6.2292,
|
|
"mean_token_accuracy": 0.13543767631053924,
|
|
"num_tokens": 4350009.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.394851160049439,
|
|
"epoch": 0.11952773394120053,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 6.1387,
|
|
"mean_token_accuracy": 0.14069160670042039,
|
|
"num_tokens": 4361740.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.380158138275147,
|
|
"epoch": 0.11981997778946753,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 6.186,
|
|
"mean_token_accuracy": 0.13409000858664513,
|
|
"num_tokens": 4373321.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.392633390426636,
|
|
"epoch": 0.12011222163773452,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 6.1687,
|
|
"mean_token_accuracy": 0.1405694454908371,
|
|
"num_tokens": 4384547.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.380757570266724,
|
|
"epoch": 0.12040446548600152,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 6.2024,
|
|
"mean_token_accuracy": 0.1324629656970501,
|
|
"num_tokens": 4396137.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.300692510604859,
|
|
"epoch": 0.12069670933426851,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 6.1746,
|
|
"mean_token_accuracy": 0.13813069686293603,
|
|
"num_tokens": 4406369.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.467877721786499,
|
|
"epoch": 0.12098895318253551,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 6.1149,
|
|
"mean_token_accuracy": 0.1388881355524063,
|
|
"num_tokens": 4416731.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.3494175434112545,
|
|
"epoch": 0.1212811970308025,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 6.2169,
|
|
"mean_token_accuracy": 0.13779886662960053,
|
|
"num_tokens": 4427946.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.43696231842041,
|
|
"epoch": 0.12157344087906949,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 6.0811,
|
|
"mean_token_accuracy": 0.1449224531650543,
|
|
"num_tokens": 4438096.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.364881896972657,
|
|
"epoch": 0.12186568472733648,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 6.0859,
|
|
"mean_token_accuracy": 0.143484915047884,
|
|
"num_tokens": 4448116.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.387076139450073,
|
|
"epoch": 0.12215792857560348,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 6.1363,
|
|
"mean_token_accuracy": 0.13883766382932664,
|
|
"num_tokens": 4459155.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.415256690979004,
|
|
"epoch": 0.12245017242387048,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 6.1477,
|
|
"mean_token_accuracy": 0.1440335601568222,
|
|
"num_tokens": 4470947.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.392708349227905,
|
|
"epoch": 0.12274241627213747,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 6.1439,
|
|
"mean_token_accuracy": 0.13736520558595658,
|
|
"num_tokens": 4481526.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.44613208770752,
|
|
"epoch": 0.12303466012040447,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 6.0882,
|
|
"mean_token_accuracy": 0.14730113446712495,
|
|
"num_tokens": 4490942.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.343643760681152,
|
|
"epoch": 0.12332690396867146,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 6.186,
|
|
"mean_token_accuracy": 0.14002663940191268,
|
|
"num_tokens": 4501880.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.446917009353638,
|
|
"epoch": 0.12361914781693846,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 6.1438,
|
|
"mean_token_accuracy": 0.13863350450992584,
|
|
"num_tokens": 4512557.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.377980470657349,
|
|
"epoch": 0.12391139166520544,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 6.1563,
|
|
"mean_token_accuracy": 0.1392514392733574,
|
|
"num_tokens": 4522334.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.481888055801392,
|
|
"epoch": 0.12420363551347244,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 6.1394,
|
|
"mean_token_accuracy": 0.14525696635246277,
|
|
"num_tokens": 4533356.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.341272830963135,
|
|
"epoch": 0.12449587936173943,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 6.1005,
|
|
"mean_token_accuracy": 0.14135508090257645,
|
|
"num_tokens": 4544225.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.431925916671753,
|
|
"epoch": 0.12478812321000643,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 6.1013,
|
|
"mean_token_accuracy": 0.1419544517993927,
|
|
"num_tokens": 4555373.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.378590774536133,
|
|
"epoch": 0.1250803670582734,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 6.1104,
|
|
"mean_token_accuracy": 0.13993178755044938,
|
|
"num_tokens": 4566938.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.28928427696228,
|
|
"epoch": 0.12537261090654042,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 6.1048,
|
|
"mean_token_accuracy": 0.14255341291427612,
|
|
"num_tokens": 4578052.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.385062742233276,
|
|
"epoch": 0.1256648547548074,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 6.17,
|
|
"mean_token_accuracy": 0.14376790225505828,
|
|
"num_tokens": 4588468.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.409221410751343,
|
|
"epoch": 0.1259570986030744,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 6.0747,
|
|
"mean_token_accuracy": 0.14142340719699859,
|
|
"num_tokens": 4599116.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.379788208007812,
|
|
"epoch": 0.1262493424513414,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 6.0961,
|
|
"mean_token_accuracy": 0.14298314303159715,
|
|
"num_tokens": 4610237.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.3200249671936035,
|
|
"epoch": 0.1265415862996084,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 6.0421,
|
|
"mean_token_accuracy": 0.15251021236181259,
|
|
"num_tokens": 4619703.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.3149495124816895,
|
|
"epoch": 0.1268338301478754,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 6.1254,
|
|
"mean_token_accuracy": 0.14367727637290956,
|
|
"num_tokens": 4630608.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.374285411834717,
|
|
"epoch": 0.12712607399614237,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 6.0613,
|
|
"mean_token_accuracy": 0.14458212554454802,
|
|
"num_tokens": 4640880.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.355184412002563,
|
|
"epoch": 0.12741831784440938,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 6.0759,
|
|
"mean_token_accuracy": 0.14326890408992768,
|
|
"num_tokens": 4651789.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.270130443572998,
|
|
"epoch": 0.12771056169267636,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 6.1599,
|
|
"mean_token_accuracy": 0.1395838215947151,
|
|
"num_tokens": 4663332.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.427866268157959,
|
|
"epoch": 0.12800280554094337,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 6.1089,
|
|
"mean_token_accuracy": 0.14353024065494538,
|
|
"num_tokens": 4674354.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.280450820922852,
|
|
"epoch": 0.12829504938921035,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 6.0498,
|
|
"mean_token_accuracy": 0.14545923769474028,
|
|
"num_tokens": 4684892.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.3894530773162845,
|
|
"epoch": 0.12858729323747736,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 6.117,
|
|
"mean_token_accuracy": 0.14214996322989465,
|
|
"num_tokens": 4695346.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.316702604293823,
|
|
"epoch": 0.12887953708574434,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 6.0321,
|
|
"mean_token_accuracy": 0.147607421875,
|
|
"num_tokens": 4706033.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.334753227233887,
|
|
"epoch": 0.12917178093401133,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 6.1575,
|
|
"mean_token_accuracy": 0.140997451543808,
|
|
"num_tokens": 4717731.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.351246976852417,
|
|
"epoch": 0.12946402478227834,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 6.0196,
|
|
"mean_token_accuracy": 0.15844787210226058,
|
|
"num_tokens": 4728017.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.340083742141724,
|
|
"epoch": 0.12975626863054532,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 6.0713,
|
|
"mean_token_accuracy": 0.1446048215031624,
|
|
"num_tokens": 4738389.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.309916210174561,
|
|
"epoch": 0.13004851247881233,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 6.0374,
|
|
"mean_token_accuracy": 0.14775751829147338,
|
|
"num_tokens": 4749139.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.323849725723266,
|
|
"epoch": 0.1303407563270793,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 6.0786,
|
|
"mean_token_accuracy": 0.15062154829502106,
|
|
"num_tokens": 4759795.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.3192380428314205,
|
|
"epoch": 0.13063300017534632,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 6.168,
|
|
"mean_token_accuracy": 0.13482984006404877,
|
|
"num_tokens": 4771405.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.37792649269104,
|
|
"epoch": 0.1309252440236133,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 6.0199,
|
|
"mean_token_accuracy": 0.15233568400144576,
|
|
"num_tokens": 4782053.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.356834936141968,
|
|
"epoch": 0.13121748787188028,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 6.0445,
|
|
"mean_token_accuracy": 0.1477915331721306,
|
|
"num_tokens": 4792027.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.320478439331055,
|
|
"epoch": 0.1315097317201473,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 6.094,
|
|
"mean_token_accuracy": 0.14143309071660043,
|
|
"num_tokens": 4801960.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.34202766418457,
|
|
"epoch": 0.13180197556841428,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 6.0362,
|
|
"mean_token_accuracy": 0.14499389231204987,
|
|
"num_tokens": 4813086.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.347697448730469,
|
|
"epoch": 0.13209421941668129,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 6.0563,
|
|
"mean_token_accuracy": 0.13959380388259887,
|
|
"num_tokens": 4823048.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.341627979278565,
|
|
"epoch": 0.13238646326494827,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 6.073,
|
|
"mean_token_accuracy": 0.13953104987740517,
|
|
"num_tokens": 4833780.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.374620962142944,
|
|
"epoch": 0.13267870711321528,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 6.0613,
|
|
"mean_token_accuracy": 0.1489867702126503,
|
|
"num_tokens": 4843610.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.369307899475098,
|
|
"epoch": 0.13297095096148226,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 6.0823,
|
|
"mean_token_accuracy": 0.14383584856987,
|
|
"num_tokens": 4855190.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.288772964477539,
|
|
"epoch": 0.13326319480974924,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 6.109,
|
|
"mean_token_accuracy": 0.1422196790575981,
|
|
"num_tokens": 4866712.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.3134016513824465,
|
|
"epoch": 0.13355543865801625,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 6.0559,
|
|
"mean_token_accuracy": 0.150993312895298,
|
|
"num_tokens": 4877477.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.394559049606324,
|
|
"epoch": 0.13384768250628323,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 6.0979,
|
|
"mean_token_accuracy": 0.14567883014678956,
|
|
"num_tokens": 4889146.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.385619020462036,
|
|
"epoch": 0.13413992635455024,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 6.0707,
|
|
"mean_token_accuracy": 0.13998873606324197,
|
|
"num_tokens": 4899564.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.345492696762085,
|
|
"epoch": 0.13443217020281722,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 6.0711,
|
|
"mean_token_accuracy": 0.14661325067281722,
|
|
"num_tokens": 4910172.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.300006341934204,
|
|
"epoch": 0.13472441405108423,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 6.113,
|
|
"mean_token_accuracy": 0.14074116125702857,
|
|
"num_tokens": 4920965.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.389887809753418,
|
|
"epoch": 0.13501665789935122,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 6.087,
|
|
"mean_token_accuracy": 0.14458852931857108,
|
|
"num_tokens": 4931595.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.26234393119812,
|
|
"epoch": 0.1353089017476182,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 6.0745,
|
|
"mean_token_accuracy": 0.1497304245829582,
|
|
"num_tokens": 4942705.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.2813232898712155,
|
|
"epoch": 0.1356011455958852,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 5.9695,
|
|
"mean_token_accuracy": 0.14979308024048804,
|
|
"num_tokens": 4952904.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.265338993072509,
|
|
"epoch": 0.1358933894441522,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 6.0539,
|
|
"mean_token_accuracy": 0.14731836020946504,
|
|
"num_tokens": 4964016.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.366107177734375,
|
|
"epoch": 0.1361856332924192,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 5.9783,
|
|
"mean_token_accuracy": 0.1532459706068039,
|
|
"num_tokens": 4974641.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.311573219299317,
|
|
"epoch": 0.13647787714068618,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 6.0779,
|
|
"mean_token_accuracy": 0.14657149612903594,
|
|
"num_tokens": 4985630.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.295196580886841,
|
|
"epoch": 0.1367701209889532,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 5.9574,
|
|
"mean_token_accuracy": 0.15096406787633895,
|
|
"num_tokens": 4996732.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.277723550796509,
|
|
"epoch": 0.13706236483722017,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 6.0122,
|
|
"mean_token_accuracy": 0.152355620265007,
|
|
"num_tokens": 5006309.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.2717853546142575,
|
|
"epoch": 0.13735460868548718,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 6.0904,
|
|
"mean_token_accuracy": 0.13694389089941977,
|
|
"num_tokens": 5018051.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.42239031791687,
|
|
"epoch": 0.13764685253375417,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 6.0987,
|
|
"mean_token_accuracy": 0.14377557635307311,
|
|
"num_tokens": 5028439.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.3302037715911865,
|
|
"epoch": 0.13793909638202115,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 6.1282,
|
|
"mean_token_accuracy": 0.1435047522187233,
|
|
"num_tokens": 5038779.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.355064821243286,
|
|
"epoch": 0.13823134023028816,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 6.1844,
|
|
"mean_token_accuracy": 0.13278792276978493,
|
|
"num_tokens": 5050662.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.376042032241822,
|
|
"epoch": 0.13852358407855514,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 6.019,
|
|
"mean_token_accuracy": 0.14877774715423583,
|
|
"num_tokens": 5061886.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.330750370025635,
|
|
"epoch": 0.13881582792682215,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 6.1781,
|
|
"mean_token_accuracy": 0.138632183521986,
|
|
"num_tokens": 5073243.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.362486124038696,
|
|
"epoch": 0.13910807177508913,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 6.0967,
|
|
"mean_token_accuracy": 0.14613902121782302,
|
|
"num_tokens": 5084913.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.349033451080322,
|
|
"epoch": 0.13940031562335614,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 6.0496,
|
|
"mean_token_accuracy": 0.14588934034109116,
|
|
"num_tokens": 5096204.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.35503921508789,
|
|
"epoch": 0.13969255947162312,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 6.059,
|
|
"mean_token_accuracy": 0.1500083699822426,
|
|
"num_tokens": 5107326.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.249629116058349,
|
|
"epoch": 0.1399848033198901,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 6.0051,
|
|
"mean_token_accuracy": 0.1472495175898075,
|
|
"num_tokens": 5118091.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.361003971099853,
|
|
"epoch": 0.14027704716815712,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 6.1266,
|
|
"mean_token_accuracy": 0.13576674535870553,
|
|
"num_tokens": 5128676.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.3102632522583,
|
|
"epoch": 0.1405692910164241,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 5.9663,
|
|
"mean_token_accuracy": 0.14985124617815018,
|
|
"num_tokens": 5139013.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.260567092895508,
|
|
"epoch": 0.1408615348646911,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 6.0188,
|
|
"mean_token_accuracy": 0.14859278574585916,
|
|
"num_tokens": 5149315.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.377109670639038,
|
|
"epoch": 0.1411537787129581,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 6.0113,
|
|
"mean_token_accuracy": 0.14948303997516632,
|
|
"num_tokens": 5160065.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.292223167419434,
|
|
"epoch": 0.1414460225612251,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 6.0482,
|
|
"mean_token_accuracy": 0.14009236469864844,
|
|
"num_tokens": 5171391.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.260541915893555,
|
|
"epoch": 0.14173826640949208,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 6.0294,
|
|
"mean_token_accuracy": 0.14464596211910247,
|
|
"num_tokens": 5183493.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.41909704208374,
|
|
"epoch": 0.14203051025775906,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 6.0594,
|
|
"mean_token_accuracy": 0.1522395893931389,
|
|
"num_tokens": 5195200.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.261653280258178,
|
|
"epoch": 0.14232275410602607,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 6.0382,
|
|
"mean_token_accuracy": 0.14992009550333024,
|
|
"num_tokens": 5206557.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.275675058364868,
|
|
"epoch": 0.14261499795429305,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 5.951,
|
|
"mean_token_accuracy": 0.15225704610347748,
|
|
"num_tokens": 5216091.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.355823040008545,
|
|
"epoch": 0.14290724180256006,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 6.0124,
|
|
"mean_token_accuracy": 0.1506971910595894,
|
|
"num_tokens": 5226294.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.2695698738098145,
|
|
"epoch": 0.14319948565082705,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 5.9299,
|
|
"mean_token_accuracy": 0.14692416712641715,
|
|
"num_tokens": 5236806.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.268785667419434,
|
|
"epoch": 0.14349172949909406,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 5.9965,
|
|
"mean_token_accuracy": 0.15370989143848418,
|
|
"num_tokens": 5247018.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.303906202316284,
|
|
"epoch": 0.14378397334736104,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 6.0195,
|
|
"mean_token_accuracy": 0.147578976303339,
|
|
"num_tokens": 5257655.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.313438129425049,
|
|
"epoch": 0.14407621719562802,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 6.0365,
|
|
"mean_token_accuracy": 0.1458984911441803,
|
|
"num_tokens": 5267642.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.31391224861145,
|
|
"epoch": 0.14436846104389503,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 6.0389,
|
|
"mean_token_accuracy": 0.15489369928836821,
|
|
"num_tokens": 5278254.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.30980076789856,
|
|
"epoch": 0.144660704892162,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 6.0574,
|
|
"mean_token_accuracy": 0.1470210149884224,
|
|
"num_tokens": 5289614.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.327652454376221,
|
|
"epoch": 0.14495294874042902,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 5.9535,
|
|
"mean_token_accuracy": 0.15733135789632796,
|
|
"num_tokens": 5300031.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.262141752243042,
|
|
"epoch": 0.145245192588696,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 6.005,
|
|
"mean_token_accuracy": 0.14667644649744033,
|
|
"num_tokens": 5310921.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.296258068084716,
|
|
"epoch": 0.145537436436963,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 5.9811,
|
|
"mean_token_accuracy": 0.15191537663340568,
|
|
"num_tokens": 5321022.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.2787957191467285,
|
|
"epoch": 0.14582968028523,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 6.0729,
|
|
"mean_token_accuracy": 0.1448964238166809,
|
|
"num_tokens": 5332066.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.259270477294922,
|
|
"epoch": 0.14612192413349698,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 6.014,
|
|
"mean_token_accuracy": 0.15154204070568084,
|
|
"num_tokens": 5342679.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.317260456085205,
|
|
"epoch": 0.146414167981764,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002740575249600342,
|
|
"loss": 6.0077,
|
|
"mean_token_accuracy": 0.15051330029964446,
|
|
"num_tokens": 5353326.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.2807210922241214,
|
|
"epoch": 0.14670641183003097,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00027287945635254263,
|
|
"loss": 6.0099,
|
|
"mean_token_accuracy": 0.15015970021486283,
|
|
"num_tokens": 5363390.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.2885383605957035,
|
|
"epoch": 0.14699865567829798,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00027170144588082635,
|
|
"loss": 5.9957,
|
|
"mean_token_accuracy": 0.1560601845383644,
|
|
"num_tokens": 5375049.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.292453050613403,
|
|
"epoch": 0.14729089952656496,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00027052352584063763,
|
|
"loss": 6.0015,
|
|
"mean_token_accuracy": 0.1479575976729393,
|
|
"num_tokens": 5385059.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.297848415374756,
|
|
"epoch": 0.14758314337483197,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 6.0885,
|
|
"mean_token_accuracy": 0.14717495888471605,
|
|
"num_tokens": 5396150.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.3423357009887695,
|
|
"epoch": 0.14787538722309895,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026816808622456937,
|
|
"loss": 5.9861,
|
|
"mean_token_accuracy": 0.15002903938293458,
|
|
"num_tokens": 5406861.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.322475433349609,
|
|
"epoch": 0.14816763107136594,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002669906312242569,
|
|
"loss": 5.9466,
|
|
"mean_token_accuracy": 0.1505460575222969,
|
|
"num_tokens": 5416484.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.265355634689331,
|
|
"epoch": 0.14845987491963294,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00026581339580483525,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.15213914066553116,
|
|
"num_tokens": 5427324.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.301798009872437,
|
|
"epoch": 0.14875211876789993,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002646364122408082,
|
|
"loss": 6.0276,
|
|
"mean_token_accuracy": 0.14176551550626754,
|
|
"num_tokens": 5438163.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.322329807281494,
|
|
"epoch": 0.14904436261616694,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002634597127997749,
|
|
"loss": 6.0182,
|
|
"mean_token_accuracy": 0.151769657433033,
|
|
"num_tokens": 5449501.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.334671449661255,
|
|
"epoch": 0.14933660646443392,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002622833297415445,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.13977302014827728,
|
|
"num_tokens": 5460662.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.293178367614746,
|
|
"epoch": 0.14962885031270093,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002611072953172531,
|
|
"loss": 5.9184,
|
|
"mean_token_accuracy": 0.1580108404159546,
|
|
"num_tokens": 5470626.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.27024393081665,
|
|
"epoch": 0.1499210941609679,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00025993164176847845,
|
|
"loss": 6.0356,
|
|
"mean_token_accuracy": 0.14866323620080948,
|
|
"num_tokens": 5481276.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.257556676864624,
|
|
"epoch": 0.1502133380092349,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002587564013263564,
|
|
"loss": 6.02,
|
|
"mean_token_accuracy": 0.1493818372488022,
|
|
"num_tokens": 5493304.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.301083326339722,
|
|
"epoch": 0.1505055818575019,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002575816062106974,
|
|
"loss": 5.9748,
|
|
"mean_token_accuracy": 0.1513707622885704,
|
|
"num_tokens": 5504324.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.3232574462890625,
|
|
"epoch": 0.15079782570576888,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00025640728862910293,
|
|
"loss": 5.956,
|
|
"mean_token_accuracy": 0.15466781854629516,
|
|
"num_tokens": 5513831.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.2814758777618405,
|
|
"epoch": 0.1510900695540359,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 5.9387,
|
|
"mean_token_accuracy": 0.15324688255786895,
|
|
"num_tokens": 5524540.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.252959537506103,
|
|
"epoch": 0.15138231340230288,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00025406021483217225,
|
|
"loss": 5.9262,
|
|
"mean_token_accuracy": 0.1604474425315857,
|
|
"num_tokens": 5535662.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.271982145309448,
|
|
"epoch": 0.15167455725056989,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00025288752296304963,
|
|
"loss": 6.0198,
|
|
"mean_token_accuracy": 0.14998863637447357,
|
|
"num_tokens": 5546028.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.2861700534820555,
|
|
"epoch": 0.15196680109883687,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000251715437318655,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.15974197536706924,
|
|
"num_tokens": 5556233.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.292358255386352,
|
|
"epoch": 0.15225904494710385,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0002505439900323084,
|
|
"loss": 5.938,
|
|
"mean_token_accuracy": 0.15526107028126718,
|
|
"num_tokens": 5568084.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.2493288040161135,
|
|
"epoch": 0.15255128879537086,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00024937321321982894,
|
|
"loss": 6.0073,
|
|
"mean_token_accuracy": 0.1541647344827652,
|
|
"num_tokens": 5578655.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.323029518127441,
|
|
"epoch": 0.15284353264363784,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00024820313897865433,
|
|
"loss": 6.0417,
|
|
"mean_token_accuracy": 0.14907857701182364,
|
|
"num_tokens": 5589358.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.300899314880371,
|
|
"epoch": 0.15313577649190485,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00024703379938696105,
|
|
"loss": 6.0246,
|
|
"mean_token_accuracy": 0.14676530808210372,
|
|
"num_tokens": 5600090.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.281002330780029,
|
|
"epoch": 0.15342802034017183,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00024586522650278447,
|
|
"loss": 6.0045,
|
|
"mean_token_accuracy": 0.1517194151878357,
|
|
"num_tokens": 5610960.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.264290618896484,
|
|
"epoch": 0.15372026418843884,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00024469745236314064,
|
|
"loss": 5.985,
|
|
"mean_token_accuracy": 0.14969370514154434,
|
|
"num_tokens": 5621513.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.30010085105896,
|
|
"epoch": 0.15401250803670583,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00024353050898314767,
|
|
"loss": 5.9837,
|
|
"mean_token_accuracy": 0.15460643470287322,
|
|
"num_tokens": 5631872.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.305936288833618,
|
|
"epoch": 0.15430475188497284,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00024236442835514743,
|
|
"loss": 5.9839,
|
|
"mean_token_accuracy": 0.15173401087522506,
|
|
"num_tokens": 5642350.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.251981496810913,
|
|
"epoch": 0.15459699573323982,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 5.9868,
|
|
"mean_token_accuracy": 0.15278189331293107,
|
|
"num_tokens": 5654040.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.2975434303283695,
|
|
"epoch": 0.1548892395815068,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00024003498320535462,
|
|
"loss": 6.007,
|
|
"mean_token_accuracy": 0.15547289401292802,
|
|
"num_tokens": 5665573.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.307774353027344,
|
|
"epoch": 0.1551814834297738,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00023887168254647727,
|
|
"loss": 5.979,
|
|
"mean_token_accuracy": 0.1505805715918541,
|
|
"num_tokens": 5676133.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.279534816741943,
|
|
"epoch": 0.1554737272780408,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00023770937236367308,
|
|
"loss": 5.9441,
|
|
"mean_token_accuracy": 0.15719100683927537,
|
|
"num_tokens": 5686647.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.252136325836181,
|
|
"epoch": 0.1557659711263078,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00023654808452226278,
|
|
"loss": 5.9825,
|
|
"mean_token_accuracy": 0.15101202726364135,
|
|
"num_tokens": 5697344.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.269639921188355,
|
|
"epoch": 0.15605821497457478,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00023538785085953912,
|
|
"loss": 5.9909,
|
|
"mean_token_accuracy": 0.15395767539739608,
|
|
"num_tokens": 5707742.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.316746234893799,
|
|
"epoch": 0.1563504588228418,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00023422870318389404,
|
|
"loss": 5.9525,
|
|
"mean_token_accuracy": 0.15365390479564667,
|
|
"num_tokens": 5718311.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.269629859924317,
|
|
"epoch": 0.15664270267110877,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002330706732739468,
|
|
"loss": 5.978,
|
|
"mean_token_accuracy": 0.1512306362390518,
|
|
"num_tokens": 5729696.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.27536826133728,
|
|
"epoch": 0.15693494651937576,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00023191379287767211,
|
|
"loss": 6.0173,
|
|
"mean_token_accuracy": 0.14431617707014083,
|
|
"num_tokens": 5740110.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.2984823226928714,
|
|
"epoch": 0.15722719036764277,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002307580937115305,
|
|
"loss": 5.9895,
|
|
"mean_token_accuracy": 0.14950032085180281,
|
|
"num_tokens": 5751621.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.1781737327575685,
|
|
"epoch": 0.15751943421590975,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00022960360745959846,
|
|
"loss": 5.9204,
|
|
"mean_token_accuracy": 0.15968104153871537,
|
|
"num_tokens": 5761709.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.324066495895385,
|
|
"epoch": 0.15781167806417676,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00022845036577269972,
|
|
"loss": 6.0165,
|
|
"mean_token_accuracy": 0.14895785972476006,
|
|
"num_tokens": 5772815.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.318694734573365,
|
|
"epoch": 0.15810392191244374,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 5.9737,
|
|
"mean_token_accuracy": 0.14964210242033005,
|
|
"num_tokens": 5783075.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.265916347503662,
|
|
"epoch": 0.15839616576071075,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002261477425258287,
|
|
"loss": 5.976,
|
|
"mean_token_accuracy": 0.1537642687559128,
|
|
"num_tokens": 5794032.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.2307812690734865,
|
|
"epoch": 0.15868840960897773,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002249984240934358,
|
|
"loss": 5.9045,
|
|
"mean_token_accuracy": 0.15973859503865243,
|
|
"num_tokens": 5804329.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.30174880027771,
|
|
"epoch": 0.1589806534572447,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00022385047647950464,
|
|
"loss": 5.9994,
|
|
"mean_token_accuracy": 0.15198224782943726,
|
|
"num_tokens": 5815155.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.3249983310699465,
|
|
"epoch": 0.15927289730551172,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002227039311555986,
|
|
"loss": 6.1159,
|
|
"mean_token_accuracy": 0.14026172012090682,
|
|
"num_tokens": 5828451.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.371728610992432,
|
|
"epoch": 0.1595651411537787,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0002215588195548372,
|
|
"loss": 6.0352,
|
|
"mean_token_accuracy": 0.15274893939495088,
|
|
"num_tokens": 5838925.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.263821315765381,
|
|
"epoch": 0.15985738500204572,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00022041517307103337,
|
|
"loss": 5.9746,
|
|
"mean_token_accuracy": 0.1520010381937027,
|
|
"num_tokens": 5850086.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.332649803161621,
|
|
"epoch": 0.1601496288503127,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0002192730230578331,
|
|
"loss": 6.0075,
|
|
"mean_token_accuracy": 0.148774591088295,
|
|
"num_tokens": 5861718.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.266178560256958,
|
|
"epoch": 0.1604418726985797,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0002181324008278559,
|
|
"loss": 5.9925,
|
|
"mean_token_accuracy": 0.15184382870793342,
|
|
"num_tokens": 5873893.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.322199106216431,
|
|
"epoch": 0.1607341165468467,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021699333765183655,
|
|
"loss": 5.9198,
|
|
"mean_token_accuracy": 0.15252243727445602,
|
|
"num_tokens": 5884191.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.2973285675048825,
|
|
"epoch": 0.16102636039511367,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002158558647577673,
|
|
"loss": 6.0063,
|
|
"mean_token_accuracy": 0.1471739687025547,
|
|
"num_tokens": 5896458.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.268417930603027,
|
|
"epoch": 0.16131860424338068,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00021472001333004215,
|
|
"loss": 5.9305,
|
|
"mean_token_accuracy": 0.15274266600608827,
|
|
"num_tokens": 5907356.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.361458253860474,
|
|
"epoch": 0.16161084809164766,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 6.0666,
|
|
"mean_token_accuracy": 0.1383233055472374,
|
|
"num_tokens": 5920130.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.228791332244873,
|
|
"epoch": 0.16190309193991467,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002124532993880799,
|
|
"loss": 5.9537,
|
|
"mean_token_accuracy": 0.16121236383914947,
|
|
"num_tokens": 5931782.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.31856484413147,
|
|
"epoch": 0.16219533578818165,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00021132249901695044,
|
|
"loss": 6.0046,
|
|
"mean_token_accuracy": 0.15023706257343292,
|
|
"num_tokens": 5942650.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.253255891799927,
|
|
"epoch": 0.16248757963644866,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00021019344439667705,
|
|
"loss": 5.8906,
|
|
"mean_token_accuracy": 0.16185409277677537,
|
|
"num_tokens": 5953660.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.299812316894531,
|
|
"epoch": 0.16277982348471565,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00020906616648086213,
|
|
"loss": 5.9545,
|
|
"mean_token_accuracy": 0.15241442620754242,
|
|
"num_tokens": 5963955.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.273017454147339,
|
|
"epoch": 0.16307206733298263,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00020794069617439942,
|
|
"loss": 5.8862,
|
|
"mean_token_accuracy": 0.16038259118795395,
|
|
"num_tokens": 5974590.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.225327730178833,
|
|
"epoch": 0.16336431118124964,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00020681706433262593,
|
|
"loss": 5.8901,
|
|
"mean_token_accuracy": 0.1646873489022255,
|
|
"num_tokens": 5985749.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.262038278579712,
|
|
"epoch": 0.16365655502951662,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00020569530176047602,
|
|
"loss": 6.0507,
|
|
"mean_token_accuracy": 0.14661407619714736,
|
|
"num_tokens": 5996314.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.3573963165283205,
|
|
"epoch": 0.16394879887778363,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002045754392116374,
|
|
"loss": 5.965,
|
|
"mean_token_accuracy": 0.15137797743082046,
|
|
"num_tokens": 6007958.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.2721435070037845,
|
|
"epoch": 0.1642410427260506,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00020345750738770757,
|
|
"loss": 5.8847,
|
|
"mean_token_accuracy": 0.16028138399124145,
|
|
"num_tokens": 6018226.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.209703016281128,
|
|
"epoch": 0.16453328657431762,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00020234153693735214,
|
|
"loss": 5.8803,
|
|
"mean_token_accuracy": 0.1555054947733879,
|
|
"num_tokens": 6027857.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.2504278182983395,
|
|
"epoch": 0.1648255304225846,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002012275584554647,
|
|
"loss": 5.9387,
|
|
"mean_token_accuracy": 0.15750135034322738,
|
|
"num_tokens": 6038776.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.242521905899048,
|
|
"epoch": 0.16511777427085159,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.15975674986839294,
|
|
"num_tokens": 6049857.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.300086832046508,
|
|
"epoch": 0.1654100181191186,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00019900569950277692,
|
|
"loss": 5.9769,
|
|
"mean_token_accuracy": 0.15135741531848906,
|
|
"num_tokens": 6060964.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.2466583251953125,
|
|
"epoch": 0.16570226196738558,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00019789787994536228,
|
|
"loss": 5.9394,
|
|
"mean_token_accuracy": 0.152987851947546,
|
|
"num_tokens": 6071545.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.261944580078125,
|
|
"epoch": 0.1659945058156526,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00019679217418151667,
|
|
"loss": 5.9064,
|
|
"mean_token_accuracy": 0.16121156588196756,
|
|
"num_tokens": 6081989.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.288651037216186,
|
|
"epoch": 0.16628674966391957,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00019568861252472236,
|
|
"loss": 5.9044,
|
|
"mean_token_accuracy": 0.1509763315320015,
|
|
"num_tokens": 6092003.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.266188669204712,
|
|
"epoch": 0.16657899351218658,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00019458722522967952,
|
|
"loss": 5.8771,
|
|
"mean_token_accuracy": 0.16544461399316787,
|
|
"num_tokens": 6102424.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.222381591796875,
|
|
"epoch": 0.16687123736045356,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00019348804249147723,
|
|
"loss": 5.8471,
|
|
"mean_token_accuracy": 0.15314731746912003,
|
|
"num_tokens": 6114241.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.209283494949341,
|
|
"epoch": 0.16716348120872054,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001923910944447655,
|
|
"loss": 6.0144,
|
|
"mean_token_accuracy": 0.151371481269598,
|
|
"num_tokens": 6126554.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.316224002838135,
|
|
"epoch": 0.16745572505698755,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00019129641116292928,
|
|
"loss": 5.9869,
|
|
"mean_token_accuracy": 0.1558510109782219,
|
|
"num_tokens": 6137073.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.278422546386719,
|
|
"epoch": 0.16774796890525454,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00019020402265726343,
|
|
"loss": 5.9021,
|
|
"mean_token_accuracy": 0.16068637669086455,
|
|
"num_tokens": 6148237.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.257135486602783,
|
|
"epoch": 0.16804021275352155,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001891139588761509,
|
|
"loss": 5.8467,
|
|
"mean_token_accuracy": 0.16827304363250734,
|
|
"num_tokens": 6158454.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.215215158462525,
|
|
"epoch": 0.16833245660178853,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00018802624970424076,
|
|
"loss": 5.939,
|
|
"mean_token_accuracy": 0.15383752435445786,
|
|
"num_tokens": 6169535.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.270279455184936,
|
|
"epoch": 0.16862470045005554,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 5.8992,
|
|
"mean_token_accuracy": 0.16143866926431655,
|
|
"num_tokens": 6179590.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.260125064849854,
|
|
"epoch": 0.16891694429832252,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00018585801440304306,
|
|
"loss": 5.9296,
|
|
"mean_token_accuracy": 0.1546649530529976,
|
|
"num_tokens": 6190415.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.25762791633606,
|
|
"epoch": 0.1692091881465895,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00018477754771702165,
|
|
"loss": 5.9416,
|
|
"mean_token_accuracy": 0.15465565025806427,
|
|
"num_tokens": 6202655.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.2693687915802006,
|
|
"epoch": 0.1695014319948565,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00018369955452510506,
|
|
"loss": 5.9241,
|
|
"mean_token_accuracy": 0.1609617494046688,
|
|
"num_tokens": 6213489.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.277913331985474,
|
|
"epoch": 0.1697936758431235,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001826240643810212,
|
|
"loss": 5.8743,
|
|
"mean_token_accuracy": 0.16472582519054413,
|
|
"num_tokens": 6223308.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.258041715621948,
|
|
"epoch": 0.1700859196913905,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001815511067698758,
|
|
"loss": 5.9134,
|
|
"mean_token_accuracy": 0.15096938610076904,
|
|
"num_tokens": 6232979.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.179304933547973,
|
|
"epoch": 0.17037816353965748,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001804807111073436,
|
|
"loss": 5.8924,
|
|
"mean_token_accuracy": 0.16016541868448259,
|
|
"num_tokens": 6243925.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.271915054321289,
|
|
"epoch": 0.1706704073879245,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001794129067388625,
|
|
"loss": 5.8819,
|
|
"mean_token_accuracy": 0.16427685022354127,
|
|
"num_tokens": 6254542.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.276421165466308,
|
|
"epoch": 0.17096265123619148,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00017834772293882868,
|
|
"loss": 5.8985,
|
|
"mean_token_accuracy": 0.14977855384349822,
|
|
"num_tokens": 6265763.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.269158172607422,
|
|
"epoch": 0.17125489508445846,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000177285188909794,
|
|
"loss": 5.9142,
|
|
"mean_token_accuracy": 0.15885187834501266,
|
|
"num_tokens": 6276480.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.261989879608154,
|
|
"epoch": 0.17154713893272547,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001762253337816656,
|
|
"loss": 5.9506,
|
|
"mean_token_accuracy": 0.15117736756801606,
|
|
"num_tokens": 6286202.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.273619747161865,
|
|
"epoch": 0.17183938278099245,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00017516818661090738,
|
|
"loss": 5.9504,
|
|
"mean_token_accuracy": 0.15589409917593003,
|
|
"num_tokens": 6298023.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.270647048950195,
|
|
"epoch": 0.17213162662925946,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 5.8714,
|
|
"mean_token_accuracy": 0.15770725160837173,
|
|
"num_tokens": 6309134.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.248269891738891,
|
|
"epoch": 0.17242387047752644,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00017306213199536115,
|
|
"loss": 5.8306,
|
|
"mean_token_accuracy": 0.16057109832763672,
|
|
"num_tokens": 6319254.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.228716802597046,
|
|
"epoch": 0.17271611432579345,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001720132822891243,
|
|
"loss": 5.9539,
|
|
"mean_token_accuracy": 0.16186094880104065,
|
|
"num_tokens": 6331053.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.291829681396484,
|
|
"epoch": 0.17300835817406043,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001709672560157769,
|
|
"loss": 5.9382,
|
|
"mean_token_accuracy": 0.1547654837369919,
|
|
"num_tokens": 6341803.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.229765462875366,
|
|
"epoch": 0.17330060202232744,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016992408185265758,
|
|
"loss": 5.8939,
|
|
"mean_token_accuracy": 0.1608334004878998,
|
|
"num_tokens": 6352389.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.2507017135620115,
|
|
"epoch": 0.17359284587059443,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00016888378839891298,
|
|
"loss": 6.027,
|
|
"mean_token_accuracy": 0.14867232590913773,
|
|
"num_tokens": 6363997.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.276239967346191,
|
|
"epoch": 0.1738850897188614,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001678464041747137,
|
|
"loss": 5.8903,
|
|
"mean_token_accuracy": 0.16226340234279632,
|
|
"num_tokens": 6374673.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.351266670227051,
|
|
"epoch": 0.17417733356712842,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00016681195762047223,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.1535469964146614,
|
|
"num_tokens": 6385743.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.219260358810425,
|
|
"epoch": 0.1744695774153954,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016578047709606337,
|
|
"loss": 5.9053,
|
|
"mean_token_accuracy": 0.15972462743520738,
|
|
"num_tokens": 6395742.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.239864826202393,
|
|
"epoch": 0.1747618212636624,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016475199088004678,
|
|
"loss": 5.8669,
|
|
"mean_token_accuracy": 0.15947192460298537,
|
|
"num_tokens": 6405917.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.29384331703186,
|
|
"epoch": 0.1750540651119294,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00016372652716889163,
|
|
"loss": 5.8966,
|
|
"mean_token_accuracy": 0.15279381796717645,
|
|
"num_tokens": 6417190.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.237878799438477,
|
|
"epoch": 0.1753463089601964,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001627041140762035,
|
|
"loss": 5.8984,
|
|
"mean_token_accuracy": 0.15591774731874466,
|
|
"num_tokens": 6426954.0,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1416156606627840.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|