4f896eeaf6
Model: fpadovani/nld-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.45433893684688775,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 6.768713426589966,
|
|
"epoch": 0.004543389368468878,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.7222,
|
|
"mean_token_accuracy": 0.1095728911459446,
|
|
"num_tokens": 10131.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 6.720575475692749,
|
|
"epoch": 0.009086778736937756,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.6561,
|
|
"mean_token_accuracy": 0.10461715683341026,
|
|
"num_tokens": 20042.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 6.794066381454468,
|
|
"epoch": 0.013630168105406633,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.6777,
|
|
"mean_token_accuracy": 0.10963529199361802,
|
|
"num_tokens": 30596.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 6.780745935440064,
|
|
"epoch": 0.01817355747387551,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.6242,
|
|
"mean_token_accuracy": 0.11041242256760597,
|
|
"num_tokens": 40667.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 6.800935125350952,
|
|
"epoch": 0.02271694684234439,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.6384,
|
|
"mean_token_accuracy": 0.10581032261252403,
|
|
"num_tokens": 51027.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.829052972793579,
|
|
"epoch": 0.027260336210813266,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.7264,
|
|
"mean_token_accuracy": 0.10166022032499314,
|
|
"num_tokens": 61860.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 6.852202463150024,
|
|
"epoch": 0.03180372557928214,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.5345,
|
|
"mean_token_accuracy": 0.1121172659099102,
|
|
"num_tokens": 72136.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 6.821422100067139,
|
|
"epoch": 0.03634711494775102,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.624,
|
|
"mean_token_accuracy": 0.10981205776333809,
|
|
"num_tokens": 82611.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 6.870565414428711,
|
|
"epoch": 0.0408905043162199,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.6909,
|
|
"mean_token_accuracy": 0.10643857344985008,
|
|
"num_tokens": 92722.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 6.841821718215942,
|
|
"epoch": 0.04543389368468878,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.6865,
|
|
"mean_token_accuracy": 0.10719444304704666,
|
|
"num_tokens": 103258.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 6.887676668167114,
|
|
"epoch": 0.049977283053157656,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.7029,
|
|
"mean_token_accuracy": 0.10421370416879654,
|
|
"num_tokens": 114277.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 6.867907333374023,
|
|
"epoch": 0.05452067242162653,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.6028,
|
|
"mean_token_accuracy": 0.10567226111888886,
|
|
"num_tokens": 124098.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 6.858053731918335,
|
|
"epoch": 0.059064061790095414,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.6126,
|
|
"mean_token_accuracy": 0.11433581411838531,
|
|
"num_tokens": 133775.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 6.852071142196655,
|
|
"epoch": 0.06360745115856428,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.6068,
|
|
"mean_token_accuracy": 0.11076592430472373,
|
|
"num_tokens": 144765.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 6.78433780670166,
|
|
"epoch": 0.06815084052703317,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.6368,
|
|
"mean_token_accuracy": 0.1169683001935482,
|
|
"num_tokens": 154641.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 6.829514408111573,
|
|
"epoch": 0.07269422989550205,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.5928,
|
|
"mean_token_accuracy": 0.11736959144473076,
|
|
"num_tokens": 164728.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 6.903993606567383,
|
|
"epoch": 0.07723761926397092,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.5341,
|
|
"mean_token_accuracy": 0.10781509801745415,
|
|
"num_tokens": 174739.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 6.8236547946929935,
|
|
"epoch": 0.0817810086324398,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.5946,
|
|
"mean_token_accuracy": 0.1138243854045868,
|
|
"num_tokens": 184580.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 6.850541257858277,
|
|
"epoch": 0.08632439800090867,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.6544,
|
|
"mean_token_accuracy": 0.1113918036222458,
|
|
"num_tokens": 195243.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 6.854983806610107,
|
|
"epoch": 0.09086778736937756,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.6038,
|
|
"mean_token_accuracy": 0.11426160708069802,
|
|
"num_tokens": 205751.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 6.885841703414917,
|
|
"epoch": 0.09541117673784644,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.5497,
|
|
"mean_token_accuracy": 0.11042018011212348,
|
|
"num_tokens": 217283.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 6.744452333450317,
|
|
"epoch": 0.09995456610631531,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.51,
|
|
"mean_token_accuracy": 0.11327873021364213,
|
|
"num_tokens": 226909.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 6.87985577583313,
|
|
"epoch": 0.10449795547478419,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.5855,
|
|
"mean_token_accuracy": 0.11328148767352104,
|
|
"num_tokens": 237343.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 6.860165071487427,
|
|
"epoch": 0.10904134484325306,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.5179,
|
|
"mean_token_accuracy": 0.1166656605899334,
|
|
"num_tokens": 246601.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 6.857304906845092,
|
|
"epoch": 0.11358473421172194,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.5727,
|
|
"mean_token_accuracy": 0.11443846747279167,
|
|
"num_tokens": 256818.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 6.8257914066314695,
|
|
"epoch": 0.11812812358019083,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.4876,
|
|
"mean_token_accuracy": 0.11914753541350365,
|
|
"num_tokens": 267157.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 6.724686956405639,
|
|
"epoch": 0.1226715129486597,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.4988,
|
|
"mean_token_accuracy": 0.11270106509327889,
|
|
"num_tokens": 277370.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 6.806635904312134,
|
|
"epoch": 0.12721490231712856,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.5163,
|
|
"mean_token_accuracy": 0.11316018253564834,
|
|
"num_tokens": 287851.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 6.965766048431396,
|
|
"epoch": 0.13175829168559747,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.6036,
|
|
"mean_token_accuracy": 0.11733311712741852,
|
|
"num_tokens": 299043.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 6.744367074966431,
|
|
"epoch": 0.13630168105406634,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.5794,
|
|
"mean_token_accuracy": 0.11460703909397126,
|
|
"num_tokens": 308525.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 6.8237261295318605,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.5493,
|
|
"mean_token_accuracy": 0.11844029426574706,
|
|
"num_tokens": 318301.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 6.790290117263794,
|
|
"epoch": 0.1453884597910041,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.6009,
|
|
"mean_token_accuracy": 0.11239130273461342,
|
|
"num_tokens": 329599.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 6.843774080276489,
|
|
"epoch": 0.14993184915947297,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.4774,
|
|
"mean_token_accuracy": 0.11909914687275887,
|
|
"num_tokens": 339446.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 6.9367317199707035,
|
|
"epoch": 0.15447523852794184,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.6333,
|
|
"mean_token_accuracy": 0.1179355539381504,
|
|
"num_tokens": 349579.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 6.8179216384887695,
|
|
"epoch": 0.15901862789641072,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.5631,
|
|
"mean_token_accuracy": 0.11152750179171562,
|
|
"num_tokens": 359452.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 6.830770826339721,
|
|
"epoch": 0.1635620172648796,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.5702,
|
|
"mean_token_accuracy": 0.11352085247635842,
|
|
"num_tokens": 369558.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 6.878524255752564,
|
|
"epoch": 0.16810540663334847,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.5577,
|
|
"mean_token_accuracy": 0.11322271078824997,
|
|
"num_tokens": 380458.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 6.9294568538665775,
|
|
"epoch": 0.17264879600181735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.5669,
|
|
"mean_token_accuracy": 0.1112134151160717,
|
|
"num_tokens": 390051.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 6.777473735809326,
|
|
"epoch": 0.17719218537028622,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.5164,
|
|
"mean_token_accuracy": 0.11806451380252839,
|
|
"num_tokens": 400343.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 6.889895868301392,
|
|
"epoch": 0.18173557473875512,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.5619,
|
|
"mean_token_accuracy": 0.11565203443169594,
|
|
"num_tokens": 410004.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 6.7963909149169925,
|
|
"epoch": 0.186278964107224,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.5851,
|
|
"mean_token_accuracy": 0.10973369553685189,
|
|
"num_tokens": 421286.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 6.9212535381317135,
|
|
"epoch": 0.19082235347569287,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.6359,
|
|
"mean_token_accuracy": 0.10912902057170867,
|
|
"num_tokens": 432351.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 6.842854261398315,
|
|
"epoch": 0.19536574284416175,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.627,
|
|
"mean_token_accuracy": 0.11367111280560493,
|
|
"num_tokens": 442843.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 6.870680522918701,
|
|
"epoch": 0.19990913221263062,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.4973,
|
|
"mean_token_accuracy": 0.11255754232406616,
|
|
"num_tokens": 453370.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 6.800969171524048,
|
|
"epoch": 0.2044525215810995,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.6865,
|
|
"mean_token_accuracy": 0.10846318081021308,
|
|
"num_tokens": 464715.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 6.875910711288452,
|
|
"epoch": 0.20899591094956838,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.4927,
|
|
"mean_token_accuracy": 0.11165988445281982,
|
|
"num_tokens": 475086.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 6.743642044067383,
|
|
"epoch": 0.21353930031803725,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.6147,
|
|
"mean_token_accuracy": 0.12273732423782349,
|
|
"num_tokens": 486088.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 6.893033027648926,
|
|
"epoch": 0.21808268968650613,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.563,
|
|
"mean_token_accuracy": 0.10623423308134079,
|
|
"num_tokens": 496978.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 6.730010509490967,
|
|
"epoch": 0.222626079054975,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.3713,
|
|
"mean_token_accuracy": 0.12526726275682448,
|
|
"num_tokens": 507794.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 6.739193630218506,
|
|
"epoch": 0.22716946842344388,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.5377,
|
|
"mean_token_accuracy": 0.10970785096287727,
|
|
"num_tokens": 517763.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 6.85530915260315,
|
|
"epoch": 0.23171285779191278,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.4664,
|
|
"mean_token_accuracy": 0.11921982616186141,
|
|
"num_tokens": 527374.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 6.741254997253418,
|
|
"epoch": 0.23625624716038165,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.5256,
|
|
"mean_token_accuracy": 0.11517209336161613,
|
|
"num_tokens": 537435.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 6.810067796707154,
|
|
"epoch": 0.24079963652885053,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.4963,
|
|
"mean_token_accuracy": 0.11652736142277717,
|
|
"num_tokens": 547236.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 6.788457632064819,
|
|
"epoch": 0.2453430258973194,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.4943,
|
|
"mean_token_accuracy": 0.11792855635285378,
|
|
"num_tokens": 557130.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 6.698097038269043,
|
|
"epoch": 0.24988641526578828,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.5301,
|
|
"mean_token_accuracy": 0.11694961413741112,
|
|
"num_tokens": 567482.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 6.759066724777222,
|
|
"epoch": 0.25442980463425713,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.3925,
|
|
"mean_token_accuracy": 0.11411905512213708,
|
|
"num_tokens": 577084.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 6.76326265335083,
|
|
"epoch": 0.25897319400272606,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.5115,
|
|
"mean_token_accuracy": 0.11677431538701058,
|
|
"num_tokens": 587980.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 6.667721605300903,
|
|
"epoch": 0.26351658337119493,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.4219,
|
|
"mean_token_accuracy": 0.12288179770112037,
|
|
"num_tokens": 598989.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 6.985043001174927,
|
|
"epoch": 0.2680599727396638,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.6076,
|
|
"mean_token_accuracy": 0.11418735831975937,
|
|
"num_tokens": 608599.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 6.601436614990234,
|
|
"epoch": 0.2726033621081327,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.4418,
|
|
"mean_token_accuracy": 0.11823201850056649,
|
|
"num_tokens": 619267.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 6.909968709945678,
|
|
"epoch": 0.27714675147660156,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.6386,
|
|
"mean_token_accuracy": 0.12054958641529083,
|
|
"num_tokens": 629991.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 6.833962726593017,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.5693,
|
|
"mean_token_accuracy": 0.11267199367284775,
|
|
"num_tokens": 641311.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 6.812391519546509,
|
|
"epoch": 0.2862335302135393,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.5406,
|
|
"mean_token_accuracy": 0.11235396489501,
|
|
"num_tokens": 651812.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 6.799654388427735,
|
|
"epoch": 0.2907769195820082,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.5081,
|
|
"mean_token_accuracy": 0.11410669162869454,
|
|
"num_tokens": 662387.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 6.780321025848389,
|
|
"epoch": 0.29532030895047706,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.4997,
|
|
"mean_token_accuracy": 0.11494562402367592,
|
|
"num_tokens": 673018.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 6.777340126037598,
|
|
"epoch": 0.29986369831894594,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.434,
|
|
"mean_token_accuracy": 0.12099529802799225,
|
|
"num_tokens": 683192.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 6.75422306060791,
|
|
"epoch": 0.3044070876874148,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.5675,
|
|
"mean_token_accuracy": 0.1253136858344078,
|
|
"num_tokens": 692747.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 6.676179361343384,
|
|
"epoch": 0.3089504770558837,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.5389,
|
|
"mean_token_accuracy": 0.11221015229821205,
|
|
"num_tokens": 705109.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 6.835248422622681,
|
|
"epoch": 0.31349386642435256,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.4544,
|
|
"mean_token_accuracy": 0.11584834307432175,
|
|
"num_tokens": 715658.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 6.698165464401245,
|
|
"epoch": 0.31803725579282144,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.4818,
|
|
"mean_token_accuracy": 0.1224296160042286,
|
|
"num_tokens": 724961.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 6.6860706329345705,
|
|
"epoch": 0.3225806451612903,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.4358,
|
|
"mean_token_accuracy": 0.1249639131128788,
|
|
"num_tokens": 735416.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 6.823048305511475,
|
|
"epoch": 0.3271240345297592,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.5408,
|
|
"mean_token_accuracy": 0.12204954698681832,
|
|
"num_tokens": 745005.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.874798774719238,
|
|
"epoch": 0.33166742389822806,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.5836,
|
|
"mean_token_accuracy": 0.11828166246414185,
|
|
"num_tokens": 754743.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 6.583785772323608,
|
|
"epoch": 0.33621081326669694,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.437,
|
|
"mean_token_accuracy": 0.1200392097234726,
|
|
"num_tokens": 764897.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 6.818075704574585,
|
|
"epoch": 0.3407542026351658,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.511,
|
|
"mean_token_accuracy": 0.11428396552801132,
|
|
"num_tokens": 775302.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.651220989227295,
|
|
"epoch": 0.3452975920036347,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.4495,
|
|
"mean_token_accuracy": 0.12070777788758277,
|
|
"num_tokens": 785694.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.732499074935913,
|
|
"epoch": 0.34984098137210357,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.3957,
|
|
"mean_token_accuracy": 0.12520513609051703,
|
|
"num_tokens": 795045.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.696088409423828,
|
|
"epoch": 0.35438437074057244,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.4921,
|
|
"mean_token_accuracy": 0.1158549316227436,
|
|
"num_tokens": 805579.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.654592990875244,
|
|
"epoch": 0.35892776010904137,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.3914,
|
|
"mean_token_accuracy": 0.12176291272044182,
|
|
"num_tokens": 816743.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.768208026885986,
|
|
"epoch": 0.36347114947751025,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.5313,
|
|
"mean_token_accuracy": 0.11511081382632256,
|
|
"num_tokens": 827217.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.615100955963134,
|
|
"epoch": 0.3680145388459791,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.3835,
|
|
"mean_token_accuracy": 0.1261222928762436,
|
|
"num_tokens": 837555.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.711054515838623,
|
|
"epoch": 0.372557928214448,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.4678,
|
|
"mean_token_accuracy": 0.11874150708317757,
|
|
"num_tokens": 846726.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.75947527885437,
|
|
"epoch": 0.3771013175829169,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.4571,
|
|
"mean_token_accuracy": 0.12268495261669159,
|
|
"num_tokens": 856535.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.683750581741333,
|
|
"epoch": 0.38164470695138575,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.531,
|
|
"mean_token_accuracy": 0.1186832346022129,
|
|
"num_tokens": 866973.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.770697355270386,
|
|
"epoch": 0.3861880963198546,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.4716,
|
|
"mean_token_accuracy": 0.11631686612963676,
|
|
"num_tokens": 877873.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.69672966003418,
|
|
"epoch": 0.3907314856883235,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.4657,
|
|
"mean_token_accuracy": 0.11843668594956398,
|
|
"num_tokens": 887135.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.835273313522339,
|
|
"epoch": 0.3952748750567924,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.4843,
|
|
"mean_token_accuracy": 0.1205623060464859,
|
|
"num_tokens": 897206.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.631858015060425,
|
|
"epoch": 0.39981826442526125,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.4857,
|
|
"mean_token_accuracy": 0.1176255464553833,
|
|
"num_tokens": 906808.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.630966377258301,
|
|
"epoch": 0.4043616537937301,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.3833,
|
|
"mean_token_accuracy": 0.1273773953318596,
|
|
"num_tokens": 917177.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.6929981231689455,
|
|
"epoch": 0.408905043162199,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.4283,
|
|
"mean_token_accuracy": 0.11764879748225213,
|
|
"num_tokens": 927229.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.704322004318238,
|
|
"epoch": 0.4134484325306679,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.4261,
|
|
"mean_token_accuracy": 0.12415409311652184,
|
|
"num_tokens": 937832.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.592918109893799,
|
|
"epoch": 0.41799182189913675,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.3477,
|
|
"mean_token_accuracy": 0.12181811258196831,
|
|
"num_tokens": 948038.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.663812637329102,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.4591,
|
|
"mean_token_accuracy": 0.11563793793320656,
|
|
"num_tokens": 957924.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.706633520126343,
|
|
"epoch": 0.4270786006360745,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.4292,
|
|
"mean_token_accuracy": 0.1167376309633255,
|
|
"num_tokens": 968497.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.624789667129517,
|
|
"epoch": 0.4316219900045434,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.3514,
|
|
"mean_token_accuracy": 0.12173967584967613,
|
|
"num_tokens": 978878.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.5878393173217775,
|
|
"epoch": 0.43616537937301225,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.4639,
|
|
"mean_token_accuracy": 0.12137211859226227,
|
|
"num_tokens": 989984.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.6825902462005615,
|
|
"epoch": 0.4407087687414811,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.3862,
|
|
"mean_token_accuracy": 0.12439708858728409,
|
|
"num_tokens": 1000254.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.580802011489868,
|
|
"epoch": 0.44525215810995,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.4602,
|
|
"mean_token_accuracy": 0.11807657480239868,
|
|
"num_tokens": 1010697.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.8235663890838625,
|
|
"epoch": 0.4497955474784189,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5558,
|
|
"mean_token_accuracy": 0.1186821848154068,
|
|
"num_tokens": 1020892.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.656906843185425,
|
|
"epoch": 0.45433893684688775,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.4358,
|
|
"mean_token_accuracy": 0.1271519921720028,
|
|
"num_tokens": 1030380.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.45433893684688775,
|
|
"eval_entropy": 6.70977929470151,
|
|
"eval_loss": 6.51187801361084,
|
|
"eval_mean_token_accuracy": 0.11979533588990222,
|
|
"eval_num_tokens": 1030380.0,
|
|
"eval_runtime": 2.6058,
|
|
"eval_samples_per_second": 1318.216,
|
|
"eval_steps_per_second": 165.017,
|
|
"step": 500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1490466908160000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|