2035 lines
54 KiB
JSON
2035 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.06697699340276615,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691980457305908,
|
|
"epoch": 0.00033488496701383075,
|
|
"grad_norm": 11.875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8068,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9336.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691903591156006,
|
|
"epoch": 0.0006697699340276615,
|
|
"grad_norm": 11.0625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7633,
|
|
"mean_token_accuracy": 8.298755274154246e-05,
|
|
"num_tokens": 18837.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690729331970214,
|
|
"epoch": 0.0010046549010414923,
|
|
"grad_norm": 8.1875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5956,
|
|
"mean_token_accuracy": 0.024797544209286572,
|
|
"num_tokens": 28150.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.671813011169434,
|
|
"epoch": 0.001339539868055323,
|
|
"grad_norm": 5.9375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.3354,
|
|
"mean_token_accuracy": 0.03901095166802406,
|
|
"num_tokens": 37469.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.546046257019043,
|
|
"epoch": 0.0016744248350691537,
|
|
"grad_norm": 4.25,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.1607,
|
|
"mean_token_accuracy": 0.04424795918166637,
|
|
"num_tokens": 47955.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.492891216278077,
|
|
"epoch": 0.0020093098020829846,
|
|
"grad_norm": 3.03125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.0269,
|
|
"mean_token_accuracy": 0.042472783103585245,
|
|
"num_tokens": 58421.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.515302562713623,
|
|
"epoch": 0.0023441947690968153,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.9959,
|
|
"mean_token_accuracy": 0.043728823959827426,
|
|
"num_tokens": 69338.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.499641704559327,
|
|
"epoch": 0.002679079736110646,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.9476,
|
|
"mean_token_accuracy": 0.038607921078801156,
|
|
"num_tokens": 78964.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.51818790435791,
|
|
"epoch": 0.0030139647031244767,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8996,
|
|
"mean_token_accuracy": 0.04694473557174206,
|
|
"num_tokens": 90312.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.432500553131103,
|
|
"epoch": 0.0033488496701383074,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.8796,
|
|
"mean_token_accuracy": 0.04102238342165947,
|
|
"num_tokens": 99706.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.511144351959228,
|
|
"epoch": 0.003683734637152138,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.7999,
|
|
"mean_token_accuracy": 0.05329355746507645,
|
|
"num_tokens": 109737.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.39623498916626,
|
|
"epoch": 0.004018619604165969,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.7913,
|
|
"mean_token_accuracy": 0.05464463010430336,
|
|
"num_tokens": 119471.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.326496887207032,
|
|
"epoch": 0.0043535045711798,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.7391,
|
|
"mean_token_accuracy": 0.05522452406585217,
|
|
"num_tokens": 130844.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.346483325958252,
|
|
"epoch": 0.0046883895381936306,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.6734,
|
|
"mean_token_accuracy": 0.05130590014159679,
|
|
"num_tokens": 141266.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.326040744781494,
|
|
"epoch": 0.005023274505207461,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.5759,
|
|
"mean_token_accuracy": 0.053099123388528825,
|
|
"num_tokens": 152222.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.252498531341553,
|
|
"epoch": 0.005358159472221292,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.4974,
|
|
"mean_token_accuracy": 0.057894356176257135,
|
|
"num_tokens": 162931.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.303020477294922,
|
|
"epoch": 0.005693044439235123,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.459,
|
|
"mean_token_accuracy": 0.0599411066621542,
|
|
"num_tokens": 172587.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.248745536804199,
|
|
"epoch": 0.006027929406248953,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.3844,
|
|
"mean_token_accuracy": 0.06057738587260246,
|
|
"num_tokens": 181877.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.166729736328126,
|
|
"epoch": 0.006362814373262784,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3398,
|
|
"mean_token_accuracy": 0.06227475889027119,
|
|
"num_tokens": 192196.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.17726526260376,
|
|
"epoch": 0.006697699340276615,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2079,
|
|
"mean_token_accuracy": 0.06091195158660412,
|
|
"num_tokens": 201009.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.157612991333007,
|
|
"epoch": 0.007032584307290445,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1724,
|
|
"mean_token_accuracy": 0.06171322390437126,
|
|
"num_tokens": 209066.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.070836448669434,
|
|
"epoch": 0.007367469274304276,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.066,
|
|
"mean_token_accuracy": 0.0612255796790123,
|
|
"num_tokens": 218721.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.060692882537841,
|
|
"epoch": 0.007702354241318108,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0073,
|
|
"mean_token_accuracy": 0.06116213873028755,
|
|
"num_tokens": 228511.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.899711894989014,
|
|
"epoch": 0.008037239208331938,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9327,
|
|
"mean_token_accuracy": 0.06402782052755356,
|
|
"num_tokens": 237097.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.886706161499024,
|
|
"epoch": 0.008372124175345769,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.8186,
|
|
"mean_token_accuracy": 0.06528197042644024,
|
|
"num_tokens": 246013.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.725101757049561,
|
|
"epoch": 0.0087070091423596,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.7975,
|
|
"mean_token_accuracy": 0.06460081338882447,
|
|
"num_tokens": 255741.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.658132648468017,
|
|
"epoch": 0.00904189410937343,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.6777,
|
|
"mean_token_accuracy": 0.06698540337383747,
|
|
"num_tokens": 265877.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.571129512786865,
|
|
"epoch": 0.009376779076387261,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.6745,
|
|
"mean_token_accuracy": 0.06440977044403554,
|
|
"num_tokens": 275631.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.40587911605835,
|
|
"epoch": 0.009711664043401092,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.6249,
|
|
"mean_token_accuracy": 0.06725125461816787,
|
|
"num_tokens": 286584.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.319342231750488,
|
|
"epoch": 0.010046549010414923,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.6003,
|
|
"mean_token_accuracy": 0.06583596840500831,
|
|
"num_tokens": 296903.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.266637802124023,
|
|
"epoch": 0.010381433977428753,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.5182,
|
|
"mean_token_accuracy": 0.06669672690331936,
|
|
"num_tokens": 306940.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.121322727203369,
|
|
"epoch": 0.010716318944442584,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.5713,
|
|
"mean_token_accuracy": 0.05945994444191456,
|
|
"num_tokens": 317344.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.144170761108398,
|
|
"epoch": 0.011051203911456415,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.4318,
|
|
"mean_token_accuracy": 0.06840684190392494,
|
|
"num_tokens": 326076.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.001359462738037,
|
|
"epoch": 0.011386088878470245,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.4173,
|
|
"mean_token_accuracy": 0.06733019910752773,
|
|
"num_tokens": 334645.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.87405948638916,
|
|
"epoch": 0.011720973845484076,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.457,
|
|
"mean_token_accuracy": 0.06887451484799385,
|
|
"num_tokens": 343784.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.911943817138672,
|
|
"epoch": 0.012055858812497907,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.4102,
|
|
"mean_token_accuracy": 0.06883096322417259,
|
|
"num_tokens": 353230.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.82119836807251,
|
|
"epoch": 0.012390743779511737,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.3994,
|
|
"mean_token_accuracy": 0.072719157487154,
|
|
"num_tokens": 362406.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.854674816131592,
|
|
"epoch": 0.012725628746525568,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.3534,
|
|
"mean_token_accuracy": 0.07635012455284595,
|
|
"num_tokens": 372418.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.726993560791016,
|
|
"epoch": 0.013060513713539399,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.3505,
|
|
"mean_token_accuracy": 0.06566916257143021,
|
|
"num_tokens": 381089.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.8147479057312,
|
|
"epoch": 0.01339539868055323,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.4281,
|
|
"mean_token_accuracy": 0.06762638352811337,
|
|
"num_tokens": 391801.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.720731449127197,
|
|
"epoch": 0.01373028364756706,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.3811,
|
|
"mean_token_accuracy": 0.06919934712350369,
|
|
"num_tokens": 402234.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.717157459259033,
|
|
"epoch": 0.01406516861458089,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.3737,
|
|
"mean_token_accuracy": 0.06955731771886349,
|
|
"num_tokens": 411665.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.639013862609863,
|
|
"epoch": 0.014400053581594721,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.3918,
|
|
"mean_token_accuracy": 0.07283912301063537,
|
|
"num_tokens": 421989.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.687447452545166,
|
|
"epoch": 0.014734938548608552,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.26,
|
|
"mean_token_accuracy": 0.07393333688378334,
|
|
"num_tokens": 431258.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.666478729248047,
|
|
"epoch": 0.015069823515622383,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.3854,
|
|
"mean_token_accuracy": 0.06810639537870884,
|
|
"num_tokens": 441920.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.616051387786865,
|
|
"epoch": 0.015404708482636215,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.2877,
|
|
"mean_token_accuracy": 0.07677756659686566,
|
|
"num_tokens": 451460.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.660403823852539,
|
|
"epoch": 0.015739593449650046,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.2816,
|
|
"mean_token_accuracy": 0.06862423159182071,
|
|
"num_tokens": 461612.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.5113130569458,
|
|
"epoch": 0.016074478416663877,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.3412,
|
|
"mean_token_accuracy": 0.073118532076478,
|
|
"num_tokens": 472706.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.678703594207764,
|
|
"epoch": 0.016409363383677707,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.3613,
|
|
"mean_token_accuracy": 0.06715637855231762,
|
|
"num_tokens": 482709.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.611370754241943,
|
|
"epoch": 0.016744248350691538,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.3567,
|
|
"mean_token_accuracy": 0.06981554329395294,
|
|
"num_tokens": 492566.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.615898895263673,
|
|
"epoch": 0.01707913331770537,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.2729,
|
|
"mean_token_accuracy": 0.07023949660360813,
|
|
"num_tokens": 502912.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.566937446594238,
|
|
"epoch": 0.0174140182847192,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.3078,
|
|
"mean_token_accuracy": 0.06855836510658264,
|
|
"num_tokens": 513000.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.519447898864746,
|
|
"epoch": 0.01774890325173303,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.2477,
|
|
"mean_token_accuracy": 0.07444192729890346,
|
|
"num_tokens": 523345.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.572467136383057,
|
|
"epoch": 0.01808378821874686,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.2611,
|
|
"mean_token_accuracy": 0.0756999634206295,
|
|
"num_tokens": 532121.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.594667053222656,
|
|
"epoch": 0.01841867318576069,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.3316,
|
|
"mean_token_accuracy": 0.0701010324060917,
|
|
"num_tokens": 541895.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.535953617095947,
|
|
"epoch": 0.018753558152774522,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.2591,
|
|
"mean_token_accuracy": 0.07193546742200851,
|
|
"num_tokens": 552144.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.536889743804931,
|
|
"epoch": 0.019088443119788353,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.2719,
|
|
"mean_token_accuracy": 0.0705334760248661,
|
|
"num_tokens": 562336.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.450227165222168,
|
|
"epoch": 0.019423328086802184,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.218,
|
|
"mean_token_accuracy": 0.08047289326786995,
|
|
"num_tokens": 572020.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.459360504150391,
|
|
"epoch": 0.019758213053816014,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.2205,
|
|
"mean_token_accuracy": 0.07739253155887127,
|
|
"num_tokens": 581418.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.509122371673584,
|
|
"epoch": 0.020093098020829845,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.2136,
|
|
"mean_token_accuracy": 0.07784642353653907,
|
|
"num_tokens": 590716.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.435087871551513,
|
|
"epoch": 0.020427982987843676,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.0997,
|
|
"mean_token_accuracy": 0.08416117876768112,
|
|
"num_tokens": 599337.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.536314296722413,
|
|
"epoch": 0.020762867954857506,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.2036,
|
|
"mean_token_accuracy": 0.07563756629824639,
|
|
"num_tokens": 609600.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.392087364196778,
|
|
"epoch": 0.021097752921871337,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.0995,
|
|
"mean_token_accuracy": 0.08121243491768837,
|
|
"num_tokens": 619121.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.443190097808838,
|
|
"epoch": 0.021432637888885168,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.1407,
|
|
"mean_token_accuracy": 0.07933203130960464,
|
|
"num_tokens": 628741.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.41283540725708,
|
|
"epoch": 0.021767522855899,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.2179,
|
|
"mean_token_accuracy": 0.07958511859178544,
|
|
"num_tokens": 637422.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.518020153045654,
|
|
"epoch": 0.02210240782291283,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.241,
|
|
"mean_token_accuracy": 0.07295509129762649,
|
|
"num_tokens": 646541.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.41762990951538,
|
|
"epoch": 0.02243729278992666,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.1165,
|
|
"mean_token_accuracy": 0.07536539658904076,
|
|
"num_tokens": 656771.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.371623420715332,
|
|
"epoch": 0.02277217775694049,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.1849,
|
|
"mean_token_accuracy": 0.08415180742740631,
|
|
"num_tokens": 666273.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.425055980682373,
|
|
"epoch": 0.02310706272395432,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.0979,
|
|
"mean_token_accuracy": 0.08333219587802887,
|
|
"num_tokens": 676473.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.404365062713623,
|
|
"epoch": 0.023441947690968152,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.1858,
|
|
"mean_token_accuracy": 0.07680471651256085,
|
|
"num_tokens": 686180.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.4719557762146,
|
|
"epoch": 0.023776832657981983,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.1134,
|
|
"mean_token_accuracy": 0.08005921840667725,
|
|
"num_tokens": 695372.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.450627994537353,
|
|
"epoch": 0.024111717624995813,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.202,
|
|
"mean_token_accuracy": 0.07519823238253594,
|
|
"num_tokens": 706182.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.343936538696289,
|
|
"epoch": 0.024446602592009644,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.0952,
|
|
"mean_token_accuracy": 0.08658334016799926,
|
|
"num_tokens": 715601.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.429388046264648,
|
|
"epoch": 0.024781487559023475,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.1968,
|
|
"mean_token_accuracy": 0.0763387732207775,
|
|
"num_tokens": 724919.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.473710441589356,
|
|
"epoch": 0.025116372526037305,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.0876,
|
|
"mean_token_accuracy": 0.0873330645263195,
|
|
"num_tokens": 733214.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.316863822937012,
|
|
"epoch": 0.025451257493051136,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.1652,
|
|
"mean_token_accuracy": 0.08139564543962478,
|
|
"num_tokens": 743230.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.357980537414551,
|
|
"epoch": 0.025786142460064967,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.1065,
|
|
"mean_token_accuracy": 0.08524948060512542,
|
|
"num_tokens": 753175.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.391164398193359,
|
|
"epoch": 0.026121027427078797,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.0377,
|
|
"mean_token_accuracy": 0.08165767267346383,
|
|
"num_tokens": 763168.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.216160011291503,
|
|
"epoch": 0.026455912394092628,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.1343,
|
|
"mean_token_accuracy": 0.078636434674263,
|
|
"num_tokens": 773477.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.393694496154785,
|
|
"epoch": 0.02679079736110646,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.1224,
|
|
"mean_token_accuracy": 0.07863845936954021,
|
|
"num_tokens": 782351.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.31768217086792,
|
|
"epoch": 0.02712568232812029,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.0673,
|
|
"mean_token_accuracy": 0.08583122119307518,
|
|
"num_tokens": 792179.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.328267860412598,
|
|
"epoch": 0.02746056729513412,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.0654,
|
|
"mean_token_accuracy": 0.08307353034615517,
|
|
"num_tokens": 802243.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.319038486480713,
|
|
"epoch": 0.02779545226214795,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.0445,
|
|
"mean_token_accuracy": 0.08092730417847634,
|
|
"num_tokens": 812069.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.23815851211548,
|
|
"epoch": 0.02813033722916178,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.033,
|
|
"mean_token_accuracy": 0.08340432792901993,
|
|
"num_tokens": 821608.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.37398271560669,
|
|
"epoch": 0.028465222196175612,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.1558,
|
|
"mean_token_accuracy": 0.08059651404619217,
|
|
"num_tokens": 831677.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.293180561065673,
|
|
"epoch": 0.028800107163189443,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.01,
|
|
"mean_token_accuracy": 0.07991714328527451,
|
|
"num_tokens": 840087.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.316405105590821,
|
|
"epoch": 0.029134992130203274,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.9978,
|
|
"mean_token_accuracy": 0.0796763051301241,
|
|
"num_tokens": 850516.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.243999099731445,
|
|
"epoch": 0.029469877097217104,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.0956,
|
|
"mean_token_accuracy": 0.0780600018799305,
|
|
"num_tokens": 860921.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.28464241027832,
|
|
"epoch": 0.029804762064230935,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0153,
|
|
"mean_token_accuracy": 0.07812976092100143,
|
|
"num_tokens": 871450.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.157562828063964,
|
|
"epoch": 0.030139647031244766,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.9852,
|
|
"mean_token_accuracy": 0.08450187891721725,
|
|
"num_tokens": 880241.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.207323741912841,
|
|
"epoch": 0.030474531998258596,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.0256,
|
|
"mean_token_accuracy": 0.08647001013159752,
|
|
"num_tokens": 890156.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.360474586486816,
|
|
"epoch": 0.03080941696527243,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.0575,
|
|
"mean_token_accuracy": 0.08064594492316246,
|
|
"num_tokens": 899718.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.222134017944336,
|
|
"epoch": 0.03114430193228626,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.9727,
|
|
"mean_token_accuracy": 0.08729927316308021,
|
|
"num_tokens": 908461.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.155136108398438,
|
|
"epoch": 0.03147918689930009,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.9164,
|
|
"mean_token_accuracy": 0.09267583414912224,
|
|
"num_tokens": 917418.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.265091705322266,
|
|
"epoch": 0.03181407186631392,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 8.005,
|
|
"mean_token_accuracy": 0.08021655455231666,
|
|
"num_tokens": 927495.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.160477352142333,
|
|
"epoch": 0.03214895683332775,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.026,
|
|
"mean_token_accuracy": 0.08863752149045467,
|
|
"num_tokens": 937510.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.250221538543702,
|
|
"epoch": 0.032483841800341584,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.9761,
|
|
"mean_token_accuracy": 0.08346135094761849,
|
|
"num_tokens": 948171.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.188724803924561,
|
|
"epoch": 0.032818726767355415,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.9508,
|
|
"mean_token_accuracy": 0.08198098838329315,
|
|
"num_tokens": 956506.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.181706428527832,
|
|
"epoch": 0.033153611734369245,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.9687,
|
|
"mean_token_accuracy": 0.08284949623048306,
|
|
"num_tokens": 966063.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.172396850585937,
|
|
"epoch": 0.033488496701383076,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9673,
|
|
"mean_token_accuracy": 0.08704572841525078,
|
|
"num_tokens": 976793.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.157568168640136,
|
|
"epoch": 0.03382338166839691,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.9731,
|
|
"mean_token_accuracy": 0.08012475073337555,
|
|
"num_tokens": 986453.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.20995044708252,
|
|
"epoch": 0.03415826663541074,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.9692,
|
|
"mean_token_accuracy": 0.0834043450653553,
|
|
"num_tokens": 996220.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.131259965896607,
|
|
"epoch": 0.03449315160242457,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.8932,
|
|
"mean_token_accuracy": 0.08495024554431438,
|
|
"num_tokens": 1005602.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.127531290054321,
|
|
"epoch": 0.0348280365694384,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.9133,
|
|
"mean_token_accuracy": 0.08695907667279243,
|
|
"num_tokens": 1015219.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.21120090484619,
|
|
"epoch": 0.03516292153645223,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.9165,
|
|
"mean_token_accuracy": 0.08631600588560104,
|
|
"num_tokens": 1023939.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.061803770065307,
|
|
"epoch": 0.03549780650346606,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.8585,
|
|
"mean_token_accuracy": 0.08982987627387047,
|
|
"num_tokens": 1033585.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.100368022918701,
|
|
"epoch": 0.03583269147047989,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.9565,
|
|
"mean_token_accuracy": 0.08503142818808555,
|
|
"num_tokens": 1044766.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.082072877883911,
|
|
"epoch": 0.03616757643749372,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.8732,
|
|
"mean_token_accuracy": 0.08982019945979118,
|
|
"num_tokens": 1054126.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.120585727691651,
|
|
"epoch": 0.03650246140450755,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.8768,
|
|
"mean_token_accuracy": 0.09157968685030937,
|
|
"num_tokens": 1063650.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.008113908767701,
|
|
"epoch": 0.03683734637152138,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.8189,
|
|
"mean_token_accuracy": 0.09066137000918388,
|
|
"num_tokens": 1072543.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.09608597755432,
|
|
"epoch": 0.037172231338535214,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.8198,
|
|
"mean_token_accuracy": 0.08501054644584656,
|
|
"num_tokens": 1082444.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.06026873588562,
|
|
"epoch": 0.037507116305549044,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8317,
|
|
"mean_token_accuracy": 0.08541221767663956,
|
|
"num_tokens": 1092055.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.042710876464843,
|
|
"epoch": 0.037842001272562875,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.887,
|
|
"mean_token_accuracy": 0.08810561299324035,
|
|
"num_tokens": 1101890.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.018218183517456,
|
|
"epoch": 0.038176886239576706,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8152,
|
|
"mean_token_accuracy": 0.08934114873409271,
|
|
"num_tokens": 1110735.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.101987218856811,
|
|
"epoch": 0.03851177120659054,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.9603,
|
|
"mean_token_accuracy": 0.08381157740950584,
|
|
"num_tokens": 1120557.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.086980247497559,
|
|
"epoch": 0.03884665617360437,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.8561,
|
|
"mean_token_accuracy": 0.07995514757931232,
|
|
"num_tokens": 1130622.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.9681366920471195,
|
|
"epoch": 0.0391815411406182,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.8259,
|
|
"mean_token_accuracy": 0.0891093835234642,
|
|
"num_tokens": 1141581.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.046728277206421,
|
|
"epoch": 0.03951642610763203,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.8651,
|
|
"mean_token_accuracy": 0.09294448494911194,
|
|
"num_tokens": 1151617.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.984439277648926,
|
|
"epoch": 0.03985131107464586,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.8352,
|
|
"mean_token_accuracy": 0.09366164356470108,
|
|
"num_tokens": 1162892.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.067850637435914,
|
|
"epoch": 0.04018619604165969,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.9426,
|
|
"mean_token_accuracy": 0.0844522412866354,
|
|
"num_tokens": 1173612.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.005484294891357,
|
|
"epoch": 0.04052108100867352,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.807,
|
|
"mean_token_accuracy": 0.08863909542560577,
|
|
"num_tokens": 1182808.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.054452276229858,
|
|
"epoch": 0.04085596597568735,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.8751,
|
|
"mean_token_accuracy": 0.08839099630713462,
|
|
"num_tokens": 1192361.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.001619625091553,
|
|
"epoch": 0.04119085094270118,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.851,
|
|
"mean_token_accuracy": 0.08427068591117859,
|
|
"num_tokens": 1202631.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.034376001358032,
|
|
"epoch": 0.04152573590971501,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.7912,
|
|
"mean_token_accuracy": 0.08779133036732674,
|
|
"num_tokens": 1211768.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.979119968414307,
|
|
"epoch": 0.04186062087672884,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.8158,
|
|
"mean_token_accuracy": 0.09106965214014054,
|
|
"num_tokens": 1221728.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.9051981449127195,
|
|
"epoch": 0.042195505843742674,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.7732,
|
|
"mean_token_accuracy": 0.08973816186189651,
|
|
"num_tokens": 1232231.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.067471170425415,
|
|
"epoch": 0.042530390810756505,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.8144,
|
|
"mean_token_accuracy": 0.08753226101398467,
|
|
"num_tokens": 1241226.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 8.007232856750488,
|
|
"epoch": 0.042865275777770336,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.8573,
|
|
"mean_token_accuracy": 0.08177237510681153,
|
|
"num_tokens": 1250749.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.910326814651489,
|
|
"epoch": 0.043200160744784166,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.7954,
|
|
"mean_token_accuracy": 0.08533023521304131,
|
|
"num_tokens": 1261082.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.903780174255371,
|
|
"epoch": 0.043535045711798,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7432,
|
|
"mean_token_accuracy": 0.09238319396972657,
|
|
"num_tokens": 1270351.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.963475847244263,
|
|
"epoch": 0.04386993067881183,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.7291,
|
|
"mean_token_accuracy": 0.09666994661092758,
|
|
"num_tokens": 1281811.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.799669075012207,
|
|
"epoch": 0.04420481564582566,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.7548,
|
|
"mean_token_accuracy": 0.08764155507087708,
|
|
"num_tokens": 1292202.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.957435178756714,
|
|
"epoch": 0.04453970061283949,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.7427,
|
|
"mean_token_accuracy": 0.09530294463038444,
|
|
"num_tokens": 1301439.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.809108877182007,
|
|
"epoch": 0.04487458557985332,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.7337,
|
|
"mean_token_accuracy": 0.09918036758899688,
|
|
"num_tokens": 1311604.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 8.025202703475951,
|
|
"epoch": 0.04520947054686715,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.7786,
|
|
"mean_token_accuracy": 0.09218334034085274,
|
|
"num_tokens": 1322518.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.847118234634399,
|
|
"epoch": 0.04554435551388098,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.7374,
|
|
"mean_token_accuracy": 0.09252650514245034,
|
|
"num_tokens": 1333581.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.792127895355224,
|
|
"epoch": 0.04587924048089481,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.6902,
|
|
"mean_token_accuracy": 0.09097213670611382,
|
|
"num_tokens": 1344249.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.907757902145386,
|
|
"epoch": 0.04621412544790864,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.6788,
|
|
"mean_token_accuracy": 0.09092599153518677,
|
|
"num_tokens": 1353761.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.737379646301269,
|
|
"epoch": 0.04654901041492247,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.722,
|
|
"mean_token_accuracy": 0.08851858228445053,
|
|
"num_tokens": 1362835.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.9839959144592285,
|
|
"epoch": 0.046883895381936304,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.7869,
|
|
"mean_token_accuracy": 0.09413292780518531,
|
|
"num_tokens": 1372847.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.847360467910766,
|
|
"epoch": 0.047218780348950135,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.6977,
|
|
"mean_token_accuracy": 0.09609557688236237,
|
|
"num_tokens": 1381698.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.798676586151123,
|
|
"epoch": 0.047553665315963965,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.7114,
|
|
"mean_token_accuracy": 0.09114584550261498,
|
|
"num_tokens": 1391759.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.8620161533355715,
|
|
"epoch": 0.047888550282977796,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.6854,
|
|
"mean_token_accuracy": 0.09425407946109772,
|
|
"num_tokens": 1400586.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.909142541885376,
|
|
"epoch": 0.04822343524999163,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.695,
|
|
"mean_token_accuracy": 0.09497113153338432,
|
|
"num_tokens": 1410677.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.711387968063354,
|
|
"epoch": 0.04855832021700546,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.6768,
|
|
"mean_token_accuracy": 0.09457632526755333,
|
|
"num_tokens": 1419818.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.831688690185547,
|
|
"epoch": 0.04889320518401929,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.6158,
|
|
"mean_token_accuracy": 0.10325995236635208,
|
|
"num_tokens": 1428681.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.851517391204834,
|
|
"epoch": 0.04922809015103312,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.6949,
|
|
"mean_token_accuracy": 0.09031575918197632,
|
|
"num_tokens": 1438008.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.784875440597534,
|
|
"epoch": 0.04956297511804695,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.7544,
|
|
"mean_token_accuracy": 0.09106084704399109,
|
|
"num_tokens": 1447091.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.801356983184815,
|
|
"epoch": 0.04989786008506078,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.656,
|
|
"mean_token_accuracy": 0.0895721461623907,
|
|
"num_tokens": 1458027.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.824974298477173,
|
|
"epoch": 0.05023274505207461,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.6719,
|
|
"mean_token_accuracy": 0.09690412506461143,
|
|
"num_tokens": 1468156.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.877511739730835,
|
|
"epoch": 0.05056763001908844,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.6835,
|
|
"mean_token_accuracy": 0.08788845539093018,
|
|
"num_tokens": 1477830.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.862154722213745,
|
|
"epoch": 0.05090251498610227,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.7345,
|
|
"mean_token_accuracy": 0.08901140540838241,
|
|
"num_tokens": 1486705.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.808433771133423,
|
|
"epoch": 0.0512373999531161,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6362,
|
|
"mean_token_accuracy": 0.09365730583667756,
|
|
"num_tokens": 1496568.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.737856769561768,
|
|
"epoch": 0.051572284920129934,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.7076,
|
|
"mean_token_accuracy": 0.0917304813861847,
|
|
"num_tokens": 1506425.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.765826082229614,
|
|
"epoch": 0.051907169887143764,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.5994,
|
|
"mean_token_accuracy": 0.1009859286248684,
|
|
"num_tokens": 1515568.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.728000831604004,
|
|
"epoch": 0.052242054854157595,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.5968,
|
|
"mean_token_accuracy": 0.10046595484018325,
|
|
"num_tokens": 1523858.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.7207512855529785,
|
|
"epoch": 0.052576939821171426,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.6111,
|
|
"mean_token_accuracy": 0.09502439796924592,
|
|
"num_tokens": 1533509.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.673251533508301,
|
|
"epoch": 0.052911824788185256,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.6744,
|
|
"mean_token_accuracy": 0.09320516288280487,
|
|
"num_tokens": 1543192.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.777507066726685,
|
|
"epoch": 0.05324670975519909,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.6586,
|
|
"mean_token_accuracy": 0.0901578888297081,
|
|
"num_tokens": 1553045.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.7652387619018555,
|
|
"epoch": 0.05358159472221292,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.6073,
|
|
"mean_token_accuracy": 0.09814728274941445,
|
|
"num_tokens": 1562446.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.633108186721802,
|
|
"epoch": 0.05391647968922675,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5497,
|
|
"mean_token_accuracy": 0.10263112857937813,
|
|
"num_tokens": 1571461.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.752575826644898,
|
|
"epoch": 0.05425136465624058,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.5798,
|
|
"mean_token_accuracy": 0.1005148634314537,
|
|
"num_tokens": 1582071.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.670538425445557,
|
|
"epoch": 0.05458624962325441,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.6808,
|
|
"mean_token_accuracy": 0.09083895236253739,
|
|
"num_tokens": 1591691.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.858022546768188,
|
|
"epoch": 0.05492113459026824,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.6396,
|
|
"mean_token_accuracy": 0.09683400839567184,
|
|
"num_tokens": 1600244.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.612673759460449,
|
|
"epoch": 0.05525601955728207,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.619,
|
|
"mean_token_accuracy": 0.09343247562646866,
|
|
"num_tokens": 1609779.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.683722066879272,
|
|
"epoch": 0.0555909045242959,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.5412,
|
|
"mean_token_accuracy": 0.09939340800046921,
|
|
"num_tokens": 1619135.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.664820003509521,
|
|
"epoch": 0.05592578949130973,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.586,
|
|
"mean_token_accuracy": 0.09291699826717377,
|
|
"num_tokens": 1628381.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.673054552078247,
|
|
"epoch": 0.05626067445832356,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.6192,
|
|
"mean_token_accuracy": 0.09085285998880863,
|
|
"num_tokens": 1639386.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.608000946044922,
|
|
"epoch": 0.056595559425337394,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.5164,
|
|
"mean_token_accuracy": 0.10141795799136162,
|
|
"num_tokens": 1647807.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.665489768981933,
|
|
"epoch": 0.056930444392351225,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.5468,
|
|
"mean_token_accuracy": 0.09349812343716621,
|
|
"num_tokens": 1657635.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.703561162948608,
|
|
"epoch": 0.057265329359365055,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.6531,
|
|
"mean_token_accuracy": 0.08841223120689393,
|
|
"num_tokens": 1667823.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.5995354652404785,
|
|
"epoch": 0.057600214326378886,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.5959,
|
|
"mean_token_accuracy": 0.09260657280683518,
|
|
"num_tokens": 1676925.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.700877571105957,
|
|
"epoch": 0.05793509929339272,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.6418,
|
|
"mean_token_accuracy": 0.08991135507822037,
|
|
"num_tokens": 1687397.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.677203750610351,
|
|
"epoch": 0.05826998426040655,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.5981,
|
|
"mean_token_accuracy": 0.09751952067017555,
|
|
"num_tokens": 1697230.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.6891765117645265,
|
|
"epoch": 0.05860486922742038,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.611,
|
|
"mean_token_accuracy": 0.09383615255355834,
|
|
"num_tokens": 1706229.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.641365814208984,
|
|
"epoch": 0.05893975419443421,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.5046,
|
|
"mean_token_accuracy": 0.09413399584591389,
|
|
"num_tokens": 1715305.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.612051963806152,
|
|
"epoch": 0.05927463916144804,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.6119,
|
|
"mean_token_accuracy": 0.08910732716321945,
|
|
"num_tokens": 1724892.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.594716787338257,
|
|
"epoch": 0.05960952412846187,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.4677,
|
|
"mean_token_accuracy": 0.10197951793670654,
|
|
"num_tokens": 1733775.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.6107720851898195,
|
|
"epoch": 0.0599444090954757,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.5022,
|
|
"mean_token_accuracy": 0.10666033327579498,
|
|
"num_tokens": 1742501.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.536665630340576,
|
|
"epoch": 0.06027929406248953,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.5668,
|
|
"mean_token_accuracy": 0.09865522757172585,
|
|
"num_tokens": 1752780.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.698166084289551,
|
|
"epoch": 0.06061417902950336,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.6061,
|
|
"mean_token_accuracy": 0.08976189568638801,
|
|
"num_tokens": 1763457.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.537037992477417,
|
|
"epoch": 0.06094906399651719,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.5807,
|
|
"mean_token_accuracy": 0.09197548180818557,
|
|
"num_tokens": 1773069.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.7022778511047365,
|
|
"epoch": 0.06128394896353103,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.5185,
|
|
"mean_token_accuracy": 0.09935994818806648,
|
|
"num_tokens": 1781994.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.542873430252075,
|
|
"epoch": 0.06161883393054486,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.5044,
|
|
"mean_token_accuracy": 0.0956446073949337,
|
|
"num_tokens": 1791591.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.510850143432617,
|
|
"epoch": 0.06195371889755869,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.4682,
|
|
"mean_token_accuracy": 0.10109519436955453,
|
|
"num_tokens": 1801317.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.581069231033325,
|
|
"epoch": 0.06228860386457252,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.4803,
|
|
"mean_token_accuracy": 0.10234349593520164,
|
|
"num_tokens": 1811325.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.482662963867187,
|
|
"epoch": 0.06262348883158635,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4983,
|
|
"mean_token_accuracy": 0.09694207534193992,
|
|
"num_tokens": 1820504.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.551458549499512,
|
|
"epoch": 0.06295837379860018,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.4648,
|
|
"mean_token_accuracy": 0.09739739596843719,
|
|
"num_tokens": 1830823.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.533585739135742,
|
|
"epoch": 0.06329325876561401,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.4656,
|
|
"mean_token_accuracy": 0.0926261655986309,
|
|
"num_tokens": 1840379.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.447070169448852,
|
|
"epoch": 0.06362814373262785,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.4576,
|
|
"mean_token_accuracy": 0.09828432761132717,
|
|
"num_tokens": 1850699.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.504718685150147,
|
|
"epoch": 0.06396302869964167,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.4994,
|
|
"mean_token_accuracy": 0.09886055812239647,
|
|
"num_tokens": 1861251.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.470757913589478,
|
|
"epoch": 0.0642979136666555,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.3063,
|
|
"mean_token_accuracy": 0.10472059026360511,
|
|
"num_tokens": 1870868.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.452810573577881,
|
|
"epoch": 0.06463279863366933,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.4641,
|
|
"mean_token_accuracy": 0.09811603054404258,
|
|
"num_tokens": 1882353.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.483161878585816,
|
|
"epoch": 0.06496768360068317,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.4347,
|
|
"mean_token_accuracy": 0.10043499097228051,
|
|
"num_tokens": 1892823.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.4483558177948,
|
|
"epoch": 0.06530256856769699,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.485,
|
|
"mean_token_accuracy": 0.09796798676252365,
|
|
"num_tokens": 1903301.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.49862232208252,
|
|
"epoch": 0.06563745353471083,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.481,
|
|
"mean_token_accuracy": 0.0938406951725483,
|
|
"num_tokens": 1914291.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.550891542434693,
|
|
"epoch": 0.06597233850172465,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.4464,
|
|
"mean_token_accuracy": 0.09930765777826309,
|
|
"num_tokens": 1923717.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.47271728515625,
|
|
"epoch": 0.06630722346873849,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.5217,
|
|
"mean_token_accuracy": 0.09654433950781822,
|
|
"num_tokens": 1932816.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.521736001968383,
|
|
"epoch": 0.06664210843575231,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.4366,
|
|
"mean_token_accuracy": 0.10356902927160264,
|
|
"num_tokens": 1942181.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.441972732543945,
|
|
"epoch": 0.06697699340276615,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.4381,
|
|
"mean_token_accuracy": 0.101007379591465,
|
|
"num_tokens": 1951086.0,
|
|
"step": 1000
|
|
}
|
|
],
|
|
"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": 3067576621056000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|