3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.0876731544800982,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742142391204833,
|
|
"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
|
|
}
|
|
],
|
|
"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": 702218932715520.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|