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": 10.742561435699463,
|
|
"epoch": 0.00033488496701383075,
|
|
"grad_norm": 4.5,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7498,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9336.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742577743530273,
|
|
"epoch": 0.0006697699340276615,
|
|
"grad_norm": 4.3125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7384,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 18837.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742557716369628,
|
|
"epoch": 0.0010046549010414923,
|
|
"grad_norm": 4.125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7272,
|
|
"mean_token_accuracy": 9.718172950670123e-05,
|
|
"num_tokens": 28150.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742551231384278,
|
|
"epoch": 0.001339539868055323,
|
|
"grad_norm": 4.40625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6837,
|
|
"mean_token_accuracy": 0.0009352716442663223,
|
|
"num_tokens": 37469.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742449760437012,
|
|
"epoch": 0.0016744248350691537,
|
|
"grad_norm": 4.125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6409,
|
|
"mean_token_accuracy": 0.008969881373923272,
|
|
"num_tokens": 47955.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742036151885987,
|
|
"epoch": 0.0020093098020829846,
|
|
"grad_norm": 3.71875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.5406,
|
|
"mean_token_accuracy": 0.03436212912201882,
|
|
"num_tokens": 58421.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740588188171387,
|
|
"epoch": 0.0023441947690968153,
|
|
"grad_norm": 3.0,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.4482,
|
|
"mean_token_accuracy": 0.04009146671742201,
|
|
"num_tokens": 69338.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.735538864135743,
|
|
"epoch": 0.002679079736110646,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.3231,
|
|
"mean_token_accuracy": 0.0391630582511425,
|
|
"num_tokens": 78964.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.725551223754882,
|
|
"epoch": 0.0030139647031244767,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.2374,
|
|
"mean_token_accuracy": 0.046432715654373166,
|
|
"num_tokens": 90312.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.711534595489502,
|
|
"epoch": 0.0033488496701383074,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.1706,
|
|
"mean_token_accuracy": 0.040825183317065236,
|
|
"num_tokens": 99706.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.701226234436035,
|
|
"epoch": 0.003683734637152138,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.1097,
|
|
"mean_token_accuracy": 0.04384128861129284,
|
|
"num_tokens": 109737.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.695288467407227,
|
|
"epoch": 0.004018619604165969,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.0916,
|
|
"mean_token_accuracy": 0.04214213602244854,
|
|
"num_tokens": 119471.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.689713382720948,
|
|
"epoch": 0.0043535045711798,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.0375,
|
|
"mean_token_accuracy": 0.041708166152238844,
|
|
"num_tokens": 130844.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.684234142303467,
|
|
"epoch": 0.0046883895381936306,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.9974,
|
|
"mean_token_accuracy": 0.04208777397871018,
|
|
"num_tokens": 141266.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.676689434051514,
|
|
"epoch": 0.005023274505207461,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.898,
|
|
"mean_token_accuracy": 0.04250998180359602,
|
|
"num_tokens": 152222.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.664115810394287,
|
|
"epoch": 0.005358159472221292,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.8609,
|
|
"mean_token_accuracy": 0.04332525990903378,
|
|
"num_tokens": 162931.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.650795078277588,
|
|
"epoch": 0.005693044439235123,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.8387,
|
|
"mean_token_accuracy": 0.044832929782569406,
|
|
"num_tokens": 172587.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.638714599609376,
|
|
"epoch": 0.006027929406248953,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.768,
|
|
"mean_token_accuracy": 0.04449521526694298,
|
|
"num_tokens": 181877.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.623728466033935,
|
|
"epoch": 0.006362814373262784,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.6987,
|
|
"mean_token_accuracy": 0.047365859150886536,
|
|
"num_tokens": 192196.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.59886417388916,
|
|
"epoch": 0.006697699340276615,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.6205,
|
|
"mean_token_accuracy": 0.04793250001966953,
|
|
"num_tokens": 201009.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.578139305114746,
|
|
"epoch": 0.007032584307290445,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.5589,
|
|
"mean_token_accuracy": 0.05401066094636917,
|
|
"num_tokens": 209066.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.551801109313965,
|
|
"epoch": 0.007367469274304276,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.4901,
|
|
"mean_token_accuracy": 0.05924240723252296,
|
|
"num_tokens": 218721.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.502099704742431,
|
|
"epoch": 0.007702354241318108,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.4194,
|
|
"mean_token_accuracy": 0.05640989877283573,
|
|
"num_tokens": 228511.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.450156116485596,
|
|
"epoch": 0.008037239208331938,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.3421,
|
|
"mean_token_accuracy": 0.060460330545902254,
|
|
"num_tokens": 237097.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.362788772583007,
|
|
"epoch": 0.008372124175345769,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.2305,
|
|
"mean_token_accuracy": 0.06126848310232162,
|
|
"num_tokens": 246013.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.28974199295044,
|
|
"epoch": 0.0087070091423596,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.2132,
|
|
"mean_token_accuracy": 0.0567143976688385,
|
|
"num_tokens": 255741.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.277145481109619,
|
|
"epoch": 0.00904189410937343,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 9.0746,
|
|
"mean_token_accuracy": 0.060845992714166644,
|
|
"num_tokens": 265877.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.188594436645507,
|
|
"epoch": 0.009376779076387261,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 9.0594,
|
|
"mean_token_accuracy": 0.05822984836995602,
|
|
"num_tokens": 275631.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.113962650299072,
|
|
"epoch": 0.009711664043401092,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.9686,
|
|
"mean_token_accuracy": 0.06056818813085556,
|
|
"num_tokens": 286584.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.048769855499268,
|
|
"epoch": 0.010046549010414923,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.9176,
|
|
"mean_token_accuracy": 0.05582499206066131,
|
|
"num_tokens": 296903.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.948976993560791,
|
|
"epoch": 0.010381433977428753,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.8091,
|
|
"mean_token_accuracy": 0.058087332919239995,
|
|
"num_tokens": 306940.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.834512996673585,
|
|
"epoch": 0.010716318944442584,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.821,
|
|
"mean_token_accuracy": 0.05299450382590294,
|
|
"num_tokens": 317344.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.730674839019775,
|
|
"epoch": 0.011051203911456415,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.6657,
|
|
"mean_token_accuracy": 0.06257540844380856,
|
|
"num_tokens": 326076.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.602271366119385,
|
|
"epoch": 0.011386088878470245,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.6223,
|
|
"mean_token_accuracy": 0.05909014530479908,
|
|
"num_tokens": 334645.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.467283821105957,
|
|
"epoch": 0.011720973845484076,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.6422,
|
|
"mean_token_accuracy": 0.06433657817542553,
|
|
"num_tokens": 343784.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.435225105285644,
|
|
"epoch": 0.012055858812497907,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.5783,
|
|
"mean_token_accuracy": 0.05760964490473271,
|
|
"num_tokens": 353230.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.283469009399415,
|
|
"epoch": 0.012390743779511737,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.5425,
|
|
"mean_token_accuracy": 0.06422759890556336,
|
|
"num_tokens": 362406.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 9.118220233917237,
|
|
"epoch": 0.012725628746525568,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.4932,
|
|
"mean_token_accuracy": 0.06339454613626003,
|
|
"num_tokens": 372418.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 9.104973220825196,
|
|
"epoch": 0.013060513713539399,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.484,
|
|
"mean_token_accuracy": 0.06246692761778831,
|
|
"num_tokens": 381089.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 9.07259292602539,
|
|
"epoch": 0.01339539868055323,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.5434,
|
|
"mean_token_accuracy": 0.05604027174413204,
|
|
"num_tokens": 391801.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 9.01559944152832,
|
|
"epoch": 0.01373028364756706,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.5002,
|
|
"mean_token_accuracy": 0.06068759709596634,
|
|
"num_tokens": 402234.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.936877250671387,
|
|
"epoch": 0.01406516861458089,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.479,
|
|
"mean_token_accuracy": 0.06238379329442978,
|
|
"num_tokens": 411665.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.94469394683838,
|
|
"epoch": 0.014400053581594721,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.4931,
|
|
"mean_token_accuracy": 0.0650034338235855,
|
|
"num_tokens": 421989.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.900172710418701,
|
|
"epoch": 0.014734938548608552,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.3895,
|
|
"mean_token_accuracy": 0.06547467894852162,
|
|
"num_tokens": 431258.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.874738311767578,
|
|
"epoch": 0.015069823515622383,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.4852,
|
|
"mean_token_accuracy": 0.06038682572543621,
|
|
"num_tokens": 441920.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.890042400360107,
|
|
"epoch": 0.015404708482636215,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.3921,
|
|
"mean_token_accuracy": 0.06986854746937751,
|
|
"num_tokens": 451460.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.85048770904541,
|
|
"epoch": 0.015739593449650046,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.3998,
|
|
"mean_token_accuracy": 0.06450125053524972,
|
|
"num_tokens": 461612.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.77711887359619,
|
|
"epoch": 0.016074478416663877,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.4376,
|
|
"mean_token_accuracy": 0.06631115414202213,
|
|
"num_tokens": 472706.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.85504970550537,
|
|
"epoch": 0.016409363383677707,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.4614,
|
|
"mean_token_accuracy": 0.061580969393253325,
|
|
"num_tokens": 482709.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.814317607879639,
|
|
"epoch": 0.016744248350691538,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.4475,
|
|
"mean_token_accuracy": 0.06584244705736637,
|
|
"num_tokens": 492566.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.831695365905762,
|
|
"epoch": 0.01707913331770537,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.4019,
|
|
"mean_token_accuracy": 0.06290778964757919,
|
|
"num_tokens": 502912.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.831443119049073,
|
|
"epoch": 0.0174140182847192,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.4192,
|
|
"mean_token_accuracy": 0.06170829311013222,
|
|
"num_tokens": 513000.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.78964204788208,
|
|
"epoch": 0.01774890325173303,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.3577,
|
|
"mean_token_accuracy": 0.06793267726898193,
|
|
"num_tokens": 523345.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.735497283935548,
|
|
"epoch": 0.01808378821874686,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.3634,
|
|
"mean_token_accuracy": 0.07245749980211258,
|
|
"num_tokens": 532121.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.812878513336182,
|
|
"epoch": 0.01841867318576069,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.4324,
|
|
"mean_token_accuracy": 0.06445505432784557,
|
|
"num_tokens": 541895.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.757740497589111,
|
|
"epoch": 0.018753558152774522,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.3649,
|
|
"mean_token_accuracy": 0.06517385765910148,
|
|
"num_tokens": 552144.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.80039710998535,
|
|
"epoch": 0.019088443119788353,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.3675,
|
|
"mean_token_accuracy": 0.06585304848849774,
|
|
"num_tokens": 562336.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.644570541381835,
|
|
"epoch": 0.019423328086802184,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.3217,
|
|
"mean_token_accuracy": 0.07459635213017464,
|
|
"num_tokens": 572020.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.7322265625,
|
|
"epoch": 0.019758213053816014,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.3229,
|
|
"mean_token_accuracy": 0.07014562897384166,
|
|
"num_tokens": 581418.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.696172523498536,
|
|
"epoch": 0.020093098020829845,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.304,
|
|
"mean_token_accuracy": 0.07059184238314628,
|
|
"num_tokens": 590716.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.63413028717041,
|
|
"epoch": 0.020427982987843676,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.2026,
|
|
"mean_token_accuracy": 0.07757920697331429,
|
|
"num_tokens": 599337.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.697781085968018,
|
|
"epoch": 0.020762867954857506,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.3118,
|
|
"mean_token_accuracy": 0.06973933763802051,
|
|
"num_tokens": 609600.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.631188106536865,
|
|
"epoch": 0.021097752921871337,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.2052,
|
|
"mean_token_accuracy": 0.0789581760764122,
|
|
"num_tokens": 619121.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.685782337188721,
|
|
"epoch": 0.021432637888885168,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.2304,
|
|
"mean_token_accuracy": 0.07488703206181527,
|
|
"num_tokens": 628741.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.586778736114502,
|
|
"epoch": 0.021767522855899,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.3125,
|
|
"mean_token_accuracy": 0.07511252909898758,
|
|
"num_tokens": 637422.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.691543483734131,
|
|
"epoch": 0.02210240782291283,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.3278,
|
|
"mean_token_accuracy": 0.06828603111207485,
|
|
"num_tokens": 646541.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.629717159271241,
|
|
"epoch": 0.02243729278992666,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.2181,
|
|
"mean_token_accuracy": 0.07183332405984402,
|
|
"num_tokens": 656771.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.613917541503906,
|
|
"epoch": 0.02277217775694049,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.2728,
|
|
"mean_token_accuracy": 0.08168347477912903,
|
|
"num_tokens": 666273.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.618037605285645,
|
|
"epoch": 0.02310706272395432,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.2002,
|
|
"mean_token_accuracy": 0.07752258330583572,
|
|
"num_tokens": 676473.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.643811130523682,
|
|
"epoch": 0.023441947690968152,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.273,
|
|
"mean_token_accuracy": 0.07563822567462922,
|
|
"num_tokens": 686180.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.547807312011718,
|
|
"epoch": 0.023776832657981983,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.2201,
|
|
"mean_token_accuracy": 0.07755913883447647,
|
|
"num_tokens": 695372.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.731934547424316,
|
|
"epoch": 0.024111717624995813,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.3056,
|
|
"mean_token_accuracy": 0.07374768368899823,
|
|
"num_tokens": 706182.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.546996784210204,
|
|
"epoch": 0.024446602592009644,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.1937,
|
|
"mean_token_accuracy": 0.08396602459251881,
|
|
"num_tokens": 715601.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.672658920288086,
|
|
"epoch": 0.024781487559023475,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.281,
|
|
"mean_token_accuracy": 0.07340774051845074,
|
|
"num_tokens": 724919.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.621439933776855,
|
|
"epoch": 0.025116372526037305,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.1873,
|
|
"mean_token_accuracy": 0.08298620432615281,
|
|
"num_tokens": 733214.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.571719074249268,
|
|
"epoch": 0.025451257493051136,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.2474,
|
|
"mean_token_accuracy": 0.07782503142952919,
|
|
"num_tokens": 743230.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.560567378997803,
|
|
"epoch": 0.025786142460064967,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.1869,
|
|
"mean_token_accuracy": 0.08309442102909088,
|
|
"num_tokens": 753175.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.578225898742676,
|
|
"epoch": 0.026121027427078797,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.1224,
|
|
"mean_token_accuracy": 0.0769454549998045,
|
|
"num_tokens": 763168.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.538741397857667,
|
|
"epoch": 0.026455912394092628,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.2106,
|
|
"mean_token_accuracy": 0.0778761513531208,
|
|
"num_tokens": 773477.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.53778133392334,
|
|
"epoch": 0.02679079736110646,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.2055,
|
|
"mean_token_accuracy": 0.07703198343515397,
|
|
"num_tokens": 782351.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.543316268920899,
|
|
"epoch": 0.02712568232812029,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.1518,
|
|
"mean_token_accuracy": 0.08180877156555652,
|
|
"num_tokens": 792179.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.528862571716308,
|
|
"epoch": 0.02746056729513412,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.1353,
|
|
"mean_token_accuracy": 0.07953180633485317,
|
|
"num_tokens": 802243.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.574699401855469,
|
|
"epoch": 0.02779545226214795,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.1209,
|
|
"mean_token_accuracy": 0.08045852780342103,
|
|
"num_tokens": 812069.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.437831401824951,
|
|
"epoch": 0.02813033722916178,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.1081,
|
|
"mean_token_accuracy": 0.08233479075133801,
|
|
"num_tokens": 821608.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.531076431274414,
|
|
"epoch": 0.028465222196175612,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.2246,
|
|
"mean_token_accuracy": 0.07980309575796127,
|
|
"num_tokens": 831677.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.510465621948242,
|
|
"epoch": 0.028800107163189443,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.0808,
|
|
"mean_token_accuracy": 0.08042851686477662,
|
|
"num_tokens": 840087.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.538240432739258,
|
|
"epoch": 0.029134992130203274,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.0842,
|
|
"mean_token_accuracy": 0.08349445015192032,
|
|
"num_tokens": 850516.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.433059692382812,
|
|
"epoch": 0.029469877097217104,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.1666,
|
|
"mean_token_accuracy": 0.07715897038578987,
|
|
"num_tokens": 860921.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.488123321533203,
|
|
"epoch": 0.029804762064230935,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0755,
|
|
"mean_token_accuracy": 0.07828016802668572,
|
|
"num_tokens": 871450.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.392322254180907,
|
|
"epoch": 0.030139647031244766,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.0472,
|
|
"mean_token_accuracy": 0.08468902185559272,
|
|
"num_tokens": 880241.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.428736782073974,
|
|
"epoch": 0.030474531998258596,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.0976,
|
|
"mean_token_accuracy": 0.08420571461319923,
|
|
"num_tokens": 890156.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.475104236602784,
|
|
"epoch": 0.03080941696527243,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.1217,
|
|
"mean_token_accuracy": 0.08020563796162605,
|
|
"num_tokens": 899718.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.438492584228516,
|
|
"epoch": 0.03114430193228626,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 8.0464,
|
|
"mean_token_accuracy": 0.08637708500027656,
|
|
"num_tokens": 908461.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.375166416168213,
|
|
"epoch": 0.03147918689930009,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.9991,
|
|
"mean_token_accuracy": 0.0926103100180626,
|
|
"num_tokens": 917418.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.406765747070313,
|
|
"epoch": 0.03181407186631392,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 8.079,
|
|
"mean_token_accuracy": 0.08318312540650367,
|
|
"num_tokens": 927495.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.399221515655517,
|
|
"epoch": 0.03214895683332775,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.0856,
|
|
"mean_token_accuracy": 0.08131228312849999,
|
|
"num_tokens": 937510.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.47674627304077,
|
|
"epoch": 0.032483841800341584,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.0468,
|
|
"mean_token_accuracy": 0.08284490816295147,
|
|
"num_tokens": 948171.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.403480339050294,
|
|
"epoch": 0.032818726767355415,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 8.0069,
|
|
"mean_token_accuracy": 0.08470370024442672,
|
|
"num_tokens": 956506.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.396288394927979,
|
|
"epoch": 0.033153611734369245,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 8.0398,
|
|
"mean_token_accuracy": 0.08274327740073203,
|
|
"num_tokens": 966063.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.39925765991211,
|
|
"epoch": 0.033488496701383076,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 8.0258,
|
|
"mean_token_accuracy": 0.08637790307402611,
|
|
"num_tokens": 976793.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.331637382507324,
|
|
"epoch": 0.03382338166839691,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 8.0365,
|
|
"mean_token_accuracy": 0.08065105900168419,
|
|
"num_tokens": 986453.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.44508924484253,
|
|
"epoch": 0.03415826663541074,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 8.0303,
|
|
"mean_token_accuracy": 0.08101510927081108,
|
|
"num_tokens": 996220.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.363963508605957,
|
|
"epoch": 0.03449315160242457,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.9742,
|
|
"mean_token_accuracy": 0.08482848778367043,
|
|
"num_tokens": 1005602.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.352523231506348,
|
|
"epoch": 0.0348280365694384,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.9811,
|
|
"mean_token_accuracy": 0.08442948013544083,
|
|
"num_tokens": 1015219.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.40156135559082,
|
|
"epoch": 0.03516292153645223,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.9737,
|
|
"mean_token_accuracy": 0.0867108203470707,
|
|
"num_tokens": 1023939.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.279272842407227,
|
|
"epoch": 0.03549780650346606,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.9333,
|
|
"mean_token_accuracy": 0.0901190310716629,
|
|
"num_tokens": 1033585.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.41298599243164,
|
|
"epoch": 0.03583269147047989,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 8.0043,
|
|
"mean_token_accuracy": 0.08536179438233375,
|
|
"num_tokens": 1044766.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.206547260284424,
|
|
"epoch": 0.03616757643749372,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.9332,
|
|
"mean_token_accuracy": 0.09000402987003327,
|
|
"num_tokens": 1054126.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.345331573486328,
|
|
"epoch": 0.03650246140450755,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.9442,
|
|
"mean_token_accuracy": 0.09181151911616325,
|
|
"num_tokens": 1063650.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.213069820404053,
|
|
"epoch": 0.03683734637152138,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.8779,
|
|
"mean_token_accuracy": 0.08874976187944413,
|
|
"num_tokens": 1072543.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.327767181396485,
|
|
"epoch": 0.037172231338535214,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.8762,
|
|
"mean_token_accuracy": 0.08635893240571021,
|
|
"num_tokens": 1082444.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.307810878753662,
|
|
"epoch": 0.037507116305549044,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8906,
|
|
"mean_token_accuracy": 0.0864050842821598,
|
|
"num_tokens": 1092055.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.254592514038086,
|
|
"epoch": 0.037842001272562875,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.9288,
|
|
"mean_token_accuracy": 0.0862724743783474,
|
|
"num_tokens": 1101890.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.24144401550293,
|
|
"epoch": 0.038176886239576706,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8582,
|
|
"mean_token_accuracy": 0.09320853427052497,
|
|
"num_tokens": 1110735.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.267009735107422,
|
|
"epoch": 0.03851177120659054,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.9943,
|
|
"mean_token_accuracy": 0.08730715587735176,
|
|
"num_tokens": 1120557.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.319797897338868,
|
|
"epoch": 0.03884665617360437,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.9042,
|
|
"mean_token_accuracy": 0.08325572535395623,
|
|
"num_tokens": 1130622.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.201663589477539,
|
|
"epoch": 0.0391815411406182,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.8789,
|
|
"mean_token_accuracy": 0.08977498188614845,
|
|
"num_tokens": 1141581.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.268255710601807,
|
|
"epoch": 0.03951642610763203,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.923,
|
|
"mean_token_accuracy": 0.09211237207055092,
|
|
"num_tokens": 1151617.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.204073715209962,
|
|
"epoch": 0.03985131107464586,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.8934,
|
|
"mean_token_accuracy": 0.09280748590826988,
|
|
"num_tokens": 1162892.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.284354782104492,
|
|
"epoch": 0.04018619604165969,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.9914,
|
|
"mean_token_accuracy": 0.08592328876256942,
|
|
"num_tokens": 1173612.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.187160301208497,
|
|
"epoch": 0.04052108100867352,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.8707,
|
|
"mean_token_accuracy": 0.08741234391927719,
|
|
"num_tokens": 1182808.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.288273811340332,
|
|
"epoch": 0.04085596597568735,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.9246,
|
|
"mean_token_accuracy": 0.09135790020227433,
|
|
"num_tokens": 1192361.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.222489738464356,
|
|
"epoch": 0.04119085094270118,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.9113,
|
|
"mean_token_accuracy": 0.08726226650178433,
|
|
"num_tokens": 1202631.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.24560251235962,
|
|
"epoch": 0.04152573590971501,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.8665,
|
|
"mean_token_accuracy": 0.08927992060780525,
|
|
"num_tokens": 1211768.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.212625122070312,
|
|
"epoch": 0.04186062087672884,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.8706,
|
|
"mean_token_accuracy": 0.09207477122545242,
|
|
"num_tokens": 1221728.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.09905309677124,
|
|
"epoch": 0.042195505843742674,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.8402,
|
|
"mean_token_accuracy": 0.09309252053499222,
|
|
"num_tokens": 1232231.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.238802337646485,
|
|
"epoch": 0.042530390810756505,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.8703,
|
|
"mean_token_accuracy": 0.08867279961705207,
|
|
"num_tokens": 1241226.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 8.307090473175048,
|
|
"epoch": 0.042865275777770336,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.9115,
|
|
"mean_token_accuracy": 0.08364064507186413,
|
|
"num_tokens": 1250749.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.128699350357056,
|
|
"epoch": 0.043200160744784166,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.8504,
|
|
"mean_token_accuracy": 0.08777497857809066,
|
|
"num_tokens": 1261082.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 8.1730393409729,
|
|
"epoch": 0.043535045711798,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.8066,
|
|
"mean_token_accuracy": 0.09291494712233543,
|
|
"num_tokens": 1270351.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.146085834503173,
|
|
"epoch": 0.04386993067881183,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.7899,
|
|
"mean_token_accuracy": 0.09739234745502472,
|
|
"num_tokens": 1281811.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 8.08869490623474,
|
|
"epoch": 0.04420481564582566,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.7993,
|
|
"mean_token_accuracy": 0.08824670910835267,
|
|
"num_tokens": 1292202.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.057586479187012,
|
|
"epoch": 0.04453970061283949,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.807,
|
|
"mean_token_accuracy": 0.09224808439612389,
|
|
"num_tokens": 1301439.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 8.100923252105712,
|
|
"epoch": 0.04487458557985332,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.7846,
|
|
"mean_token_accuracy": 0.09779466465115547,
|
|
"num_tokens": 1311604.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 8.290678691864013,
|
|
"epoch": 0.04520947054686715,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.8302,
|
|
"mean_token_accuracy": 0.09104525521397591,
|
|
"num_tokens": 1322518.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 8.035249519348145,
|
|
"epoch": 0.04554435551388098,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.7964,
|
|
"mean_token_accuracy": 0.09436259418725967,
|
|
"num_tokens": 1333581.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 8.07200093269348,
|
|
"epoch": 0.04587924048089481,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.7399,
|
|
"mean_token_accuracy": 0.09273072853684425,
|
|
"num_tokens": 1344249.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 8.145339584350586,
|
|
"epoch": 0.04621412544790864,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.7256,
|
|
"mean_token_accuracy": 0.09456809088587761,
|
|
"num_tokens": 1353761.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.990784692764282,
|
|
"epoch": 0.04654901041492247,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.785,
|
|
"mean_token_accuracy": 0.08893016204237938,
|
|
"num_tokens": 1362835.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 8.228912353515625,
|
|
"epoch": 0.046883895381936304,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.8368,
|
|
"mean_token_accuracy": 0.09362595304846763,
|
|
"num_tokens": 1372847.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 8.148432970046997,
|
|
"epoch": 0.047218780348950135,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.7399,
|
|
"mean_token_accuracy": 0.10006837248802185,
|
|
"num_tokens": 1381698.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.946338415145874,
|
|
"epoch": 0.047553665315963965,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.762,
|
|
"mean_token_accuracy": 0.09292992800474167,
|
|
"num_tokens": 1391759.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 8.14890651702881,
|
|
"epoch": 0.047888550282977796,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.731,
|
|
"mean_token_accuracy": 0.09815617725253105,
|
|
"num_tokens": 1400586.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 8.16442232131958,
|
|
"epoch": 0.04822343524999163,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.7465,
|
|
"mean_token_accuracy": 0.09635337144136429,
|
|
"num_tokens": 1410677.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.930854368209839,
|
|
"epoch": 0.04855832021700546,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.7272,
|
|
"mean_token_accuracy": 0.0943112052977085,
|
|
"num_tokens": 1419818.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 8.03712043762207,
|
|
"epoch": 0.04889320518401929,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.654,
|
|
"mean_token_accuracy": 0.10538181439042091,
|
|
"num_tokens": 1428681.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 8.100661420822144,
|
|
"epoch": 0.04922809015103312,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.7416,
|
|
"mean_token_accuracy": 0.09413149803876877,
|
|
"num_tokens": 1438008.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 8.020353555679321,
|
|
"epoch": 0.04956297511804695,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.7914,
|
|
"mean_token_accuracy": 0.0905725158751011,
|
|
"num_tokens": 1447091.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 8.02063798904419,
|
|
"epoch": 0.04989786008506078,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.698,
|
|
"mean_token_accuracy": 0.09145140424370765,
|
|
"num_tokens": 1458027.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 8.062599182128906,
|
|
"epoch": 0.05023274505207461,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.7247,
|
|
"mean_token_accuracy": 0.09785099253058434,
|
|
"num_tokens": 1468156.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 8.08269281387329,
|
|
"epoch": 0.05056763001908844,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.7357,
|
|
"mean_token_accuracy": 0.08801156133413315,
|
|
"num_tokens": 1477830.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 8.07049012184143,
|
|
"epoch": 0.05090251498610227,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.7971,
|
|
"mean_token_accuracy": 0.08908197171986103,
|
|
"num_tokens": 1486705.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 8.055652761459351,
|
|
"epoch": 0.0512373999531161,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6818,
|
|
"mean_token_accuracy": 0.09971900805830955,
|
|
"num_tokens": 1496568.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 8.030837202072144,
|
|
"epoch": 0.051572284920129934,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.7451,
|
|
"mean_token_accuracy": 0.09271370768547058,
|
|
"num_tokens": 1506425.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.966542053222656,
|
|
"epoch": 0.051907169887143764,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.646,
|
|
"mean_token_accuracy": 0.10024412348866463,
|
|
"num_tokens": 1515568.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 8.031002712249755,
|
|
"epoch": 0.052242054854157595,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.6379,
|
|
"mean_token_accuracy": 0.10324002727866173,
|
|
"num_tokens": 1523858.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.983342504501342,
|
|
"epoch": 0.052576939821171426,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.6499,
|
|
"mean_token_accuracy": 0.09914737194776535,
|
|
"num_tokens": 1533509.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.9228753566741945,
|
|
"epoch": 0.052911824788185256,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.7059,
|
|
"mean_token_accuracy": 0.09512233585119248,
|
|
"num_tokens": 1543192.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 8.076800632476807,
|
|
"epoch": 0.05324670975519909,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.6917,
|
|
"mean_token_accuracy": 0.09451689720153808,
|
|
"num_tokens": 1553045.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.996227884292603,
|
|
"epoch": 0.05358159472221292,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.648,
|
|
"mean_token_accuracy": 0.09735450223088264,
|
|
"num_tokens": 1562446.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.856033229827881,
|
|
"epoch": 0.05391647968922675,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5841,
|
|
"mean_token_accuracy": 0.10082900375127793,
|
|
"num_tokens": 1571461.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.977444648742676,
|
|
"epoch": 0.05425136465624058,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.6284,
|
|
"mean_token_accuracy": 0.10238974690437316,
|
|
"num_tokens": 1582071.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 8.023001098632813,
|
|
"epoch": 0.05458624962325441,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.7136,
|
|
"mean_token_accuracy": 0.0924065351486206,
|
|
"num_tokens": 1591691.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.991066980361938,
|
|
"epoch": 0.05492113459026824,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.6841,
|
|
"mean_token_accuracy": 0.09669945389032364,
|
|
"num_tokens": 1600244.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.979223203659058,
|
|
"epoch": 0.05525601955728207,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.6563,
|
|
"mean_token_accuracy": 0.09318544045090675,
|
|
"num_tokens": 1609779.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.925570821762085,
|
|
"epoch": 0.0555909045242959,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.5836,
|
|
"mean_token_accuracy": 0.10019738301634788,
|
|
"num_tokens": 1619135.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.887572002410889,
|
|
"epoch": 0.05592578949130973,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.6105,
|
|
"mean_token_accuracy": 0.09635785594582558,
|
|
"num_tokens": 1628381.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.935930490493774,
|
|
"epoch": 0.05626067445832356,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.6531,
|
|
"mean_token_accuracy": 0.09439398795366287,
|
|
"num_tokens": 1639386.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.879091262817383,
|
|
"epoch": 0.056595559425337394,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.5529,
|
|
"mean_token_accuracy": 0.10650224983692169,
|
|
"num_tokens": 1647807.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.901015996932983,
|
|
"epoch": 0.056930444392351225,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.584,
|
|
"mean_token_accuracy": 0.09403106421232224,
|
|
"num_tokens": 1657635.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.964585685729981,
|
|
"epoch": 0.057265329359365055,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.6859,
|
|
"mean_token_accuracy": 0.09096038043498993,
|
|
"num_tokens": 1667823.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.879450464248658,
|
|
"epoch": 0.057600214326378886,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.6386,
|
|
"mean_token_accuracy": 0.09283353984355927,
|
|
"num_tokens": 1676925.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.94977560043335,
|
|
"epoch": 0.05793509929339272,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.6586,
|
|
"mean_token_accuracy": 0.09341929107904434,
|
|
"num_tokens": 1687397.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.8930269241333,
|
|
"epoch": 0.05826998426040655,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.6242,
|
|
"mean_token_accuracy": 0.09667704999446869,
|
|
"num_tokens": 1697230.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.992204046249389,
|
|
"epoch": 0.05860486922742038,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.6393,
|
|
"mean_token_accuracy": 0.09668843373656273,
|
|
"num_tokens": 1706229.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.849767351150513,
|
|
"epoch": 0.05893975419443421,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.5511,
|
|
"mean_token_accuracy": 0.09463300779461861,
|
|
"num_tokens": 1715305.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.9153985500335695,
|
|
"epoch": 0.05927463916144804,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.6376,
|
|
"mean_token_accuracy": 0.09340493753552437,
|
|
"num_tokens": 1724892.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.810058546066284,
|
|
"epoch": 0.05960952412846187,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.5227,
|
|
"mean_token_accuracy": 0.10508455261588097,
|
|
"num_tokens": 1733775.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.914309883117676,
|
|
"epoch": 0.0599444090954757,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.5395,
|
|
"mean_token_accuracy": 0.10911466106772423,
|
|
"num_tokens": 1742501.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.772480010986328,
|
|
"epoch": 0.06027929406248953,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.5789,
|
|
"mean_token_accuracy": 0.09991946816444397,
|
|
"num_tokens": 1752780.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.951027917861938,
|
|
"epoch": 0.06061417902950336,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.6364,
|
|
"mean_token_accuracy": 0.09094425886869431,
|
|
"num_tokens": 1763457.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.835178327560425,
|
|
"epoch": 0.06094906399651719,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.5989,
|
|
"mean_token_accuracy": 0.0938416913151741,
|
|
"num_tokens": 1773069.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.887903165817261,
|
|
"epoch": 0.06128394896353103,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.5498,
|
|
"mean_token_accuracy": 0.09773901179432869,
|
|
"num_tokens": 1781994.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.855052757263183,
|
|
"epoch": 0.06161883393054486,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.5503,
|
|
"mean_token_accuracy": 0.09462841302156448,
|
|
"num_tokens": 1791591.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.815085697174072,
|
|
"epoch": 0.06195371889755869,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.5154,
|
|
"mean_token_accuracy": 0.10258796066045761,
|
|
"num_tokens": 1801317.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.781359052658081,
|
|
"epoch": 0.06228860386457252,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.513,
|
|
"mean_token_accuracy": 0.10400681644678116,
|
|
"num_tokens": 1811325.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.810634708404541,
|
|
"epoch": 0.06262348883158635,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.5364,
|
|
"mean_token_accuracy": 0.09745331034064293,
|
|
"num_tokens": 1820504.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.7501506328582765,
|
|
"epoch": 0.06295837379860018,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.4991,
|
|
"mean_token_accuracy": 0.10291953086853027,
|
|
"num_tokens": 1830823.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.8284999370574955,
|
|
"epoch": 0.06329325876561401,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.4896,
|
|
"mean_token_accuracy": 0.09139353856444359,
|
|
"num_tokens": 1840379.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.695800733566284,
|
|
"epoch": 0.06362814373262785,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.4678,
|
|
"mean_token_accuracy": 0.10147519782185555,
|
|
"num_tokens": 1850699.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.767111921310425,
|
|
"epoch": 0.06396302869964167,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.5196,
|
|
"mean_token_accuracy": 0.09947453439235687,
|
|
"num_tokens": 1861251.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.708375740051269,
|
|
"epoch": 0.0642979136666555,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.3227,
|
|
"mean_token_accuracy": 0.10750749334692955,
|
|
"num_tokens": 1870868.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.748810529708862,
|
|
"epoch": 0.06463279863366933,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.5017,
|
|
"mean_token_accuracy": 0.09858887121081353,
|
|
"num_tokens": 1882353.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.75818099975586,
|
|
"epoch": 0.06496768360068317,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.4573,
|
|
"mean_token_accuracy": 0.10296346917748452,
|
|
"num_tokens": 1892823.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.661552047729492,
|
|
"epoch": 0.06530256856769699,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.5001,
|
|
"mean_token_accuracy": 0.09801281914114952,
|
|
"num_tokens": 1903301.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.781193447113037,
|
|
"epoch": 0.06563745353471083,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.4987,
|
|
"mean_token_accuracy": 0.09569328427314758,
|
|
"num_tokens": 1914291.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.752705764770508,
|
|
"epoch": 0.06597233850172465,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.4646,
|
|
"mean_token_accuracy": 0.09803513586521148,
|
|
"num_tokens": 1923717.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.723828744888306,
|
|
"epoch": 0.06630722346873849,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.5419,
|
|
"mean_token_accuracy": 0.09752252623438835,
|
|
"num_tokens": 1932816.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.702096891403198,
|
|
"epoch": 0.06664210843575231,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.4429,
|
|
"mean_token_accuracy": 0.10366016700863838,
|
|
"num_tokens": 1942181.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.773722410202026,
|
|
"epoch": 0.06697699340276615,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.4597,
|
|
"mean_token_accuracy": 0.101081582903862,
|
|
"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": 454804587970560.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|