6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.2881844380403458,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742348670959473,
|
|
"epoch": 0.0004803073967339097,
|
|
"grad_norm": 4.875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7861,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10821.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742372608184814,
|
|
"epoch": 0.0009606147934678194,
|
|
"grad_norm": 4.5625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7936,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 21838.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742314338684082,
|
|
"epoch": 0.001440922190201729,
|
|
"grad_norm": 4.90625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7771,
|
|
"mean_token_accuracy": 6.802721181884409e-05,
|
|
"num_tokens": 33483.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742245483398438,
|
|
"epoch": 0.0019212295869356388,
|
|
"grad_norm": 4.5625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.728,
|
|
"mean_token_accuracy": 7.44047632906586e-05,
|
|
"num_tokens": 45150.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.74207124710083,
|
|
"epoch": 0.0024015369836695487,
|
|
"grad_norm": 4.5,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6308,
|
|
"mean_token_accuracy": 0.0004121576203033328,
|
|
"num_tokens": 55862.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741769409179687,
|
|
"epoch": 0.002881844380403458,
|
|
"grad_norm": 3.546875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.5443,
|
|
"mean_token_accuracy": 0.002983335469616577,
|
|
"num_tokens": 66766.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741297435760497,
|
|
"epoch": 0.0033621517771373678,
|
|
"grad_norm": 3.34375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3781,
|
|
"mean_token_accuracy": 0.02464218223467469,
|
|
"num_tokens": 77012.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.740566635131836,
|
|
"epoch": 0.0038424591738712775,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2814,
|
|
"mean_token_accuracy": 0.03235357943922281,
|
|
"num_tokens": 89034.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.739375877380372,
|
|
"epoch": 0.004322766570605188,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.1514,
|
|
"mean_token_accuracy": 0.038061602786183354,
|
|
"num_tokens": 100341.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.738532257080077,
|
|
"epoch": 0.004803073967339097,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.0584,
|
|
"mean_token_accuracy": 0.038722694478929044,
|
|
"num_tokens": 111997.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.738390827178955,
|
|
"epoch": 0.005283381364073006,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.0148,
|
|
"mean_token_accuracy": 0.03572444599121809,
|
|
"num_tokens": 123705.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.737564182281494,
|
|
"epoch": 0.005763688760806916,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.9071,
|
|
"mean_token_accuracy": 0.03962940648198128,
|
|
"num_tokens": 134091.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.7362642288208,
|
|
"epoch": 0.006243996157540826,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8911,
|
|
"mean_token_accuracy": 0.04431229718029499,
|
|
"num_tokens": 145731.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.734374618530273,
|
|
"epoch": 0.0067243035542747355,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8077,
|
|
"mean_token_accuracy": 0.042512409016489984,
|
|
"num_tokens": 156979.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.732056808471679,
|
|
"epoch": 0.007204610951008645,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7359,
|
|
"mean_token_accuracy": 0.04284955617040396,
|
|
"num_tokens": 167021.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.728988742828369,
|
|
"epoch": 0.007684918347742555,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6987,
|
|
"mean_token_accuracy": 0.044812921434640884,
|
|
"num_tokens": 179269.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.725158405303954,
|
|
"epoch": 0.008165225744476465,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.6221,
|
|
"mean_token_accuracy": 0.0440741615369916,
|
|
"num_tokens": 190328.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.719916248321534,
|
|
"epoch": 0.008645533141210375,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4954,
|
|
"mean_token_accuracy": 0.04290707334876061,
|
|
"num_tokens": 201522.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.711879539489747,
|
|
"epoch": 0.009125840537944284,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4379,
|
|
"mean_token_accuracy": 0.046910974010825154,
|
|
"num_tokens": 212963.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.700801849365234,
|
|
"epoch": 0.009606147934678195,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.4243,
|
|
"mean_token_accuracy": 0.04375532791018486,
|
|
"num_tokens": 224446.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.685577392578125,
|
|
"epoch": 0.010086455331412104,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2512,
|
|
"mean_token_accuracy": 0.05137390121817589,
|
|
"num_tokens": 235334.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.665137195587159,
|
|
"epoch": 0.010566762728146013,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.173,
|
|
"mean_token_accuracy": 0.051672637835144995,
|
|
"num_tokens": 247024.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.635279273986816,
|
|
"epoch": 0.011047070124879923,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0046,
|
|
"mean_token_accuracy": 0.05748001225292683,
|
|
"num_tokens": 258354.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.598550987243652,
|
|
"epoch": 0.011527377521613832,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9106,
|
|
"mean_token_accuracy": 0.05730769820511341,
|
|
"num_tokens": 271202.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.54614782333374,
|
|
"epoch": 0.012007684918347743,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7763,
|
|
"mean_token_accuracy": 0.05949556417763233,
|
|
"num_tokens": 282405.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.473622226715088,
|
|
"epoch": 0.012487992315081652,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6177,
|
|
"mean_token_accuracy": 0.067269791290164,
|
|
"num_tokens": 294211.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.406041431427003,
|
|
"epoch": 0.012968299711815562,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.5544,
|
|
"mean_token_accuracy": 0.06200371570885181,
|
|
"num_tokens": 305420.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.324399662017822,
|
|
"epoch": 0.013448607108549471,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.3911,
|
|
"mean_token_accuracy": 0.060704247653484346,
|
|
"num_tokens": 316430.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.210215950012207,
|
|
"epoch": 0.013928914505283382,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.2694,
|
|
"mean_token_accuracy": 0.06927172876894475,
|
|
"num_tokens": 328406.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.086783313751221,
|
|
"epoch": 0.01440922190201729,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.1457,
|
|
"mean_token_accuracy": 0.06873320862650871,
|
|
"num_tokens": 339051.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.933918571472168,
|
|
"epoch": 0.014889529298751201,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.0403,
|
|
"mean_token_accuracy": 0.07153546996414661,
|
|
"num_tokens": 350308.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.77935724258423,
|
|
"epoch": 0.01536983669548511,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9367,
|
|
"mean_token_accuracy": 0.06920299977064133,
|
|
"num_tokens": 362159.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.58675937652588,
|
|
"epoch": 0.01585014409221902,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.8746,
|
|
"mean_token_accuracy": 0.06771804988384247,
|
|
"num_tokens": 375157.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.392772483825684,
|
|
"epoch": 0.01633045148895293,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7082,
|
|
"mean_token_accuracy": 0.07377725951373577,
|
|
"num_tokens": 387318.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.153958225250244,
|
|
"epoch": 0.01681075888568684,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.662,
|
|
"mean_token_accuracy": 0.08222539648413658,
|
|
"num_tokens": 398933.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.963881874084473,
|
|
"epoch": 0.01729106628242075,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.5815,
|
|
"mean_token_accuracy": 0.0776383075863123,
|
|
"num_tokens": 410460.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.766354751586913,
|
|
"epoch": 0.01777137367915466,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5298,
|
|
"mean_token_accuracy": 0.07729721069335938,
|
|
"num_tokens": 421894.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.61376428604126,
|
|
"epoch": 0.01825168107588857,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.5055,
|
|
"mean_token_accuracy": 0.07357447482645511,
|
|
"num_tokens": 432976.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.49586353302002,
|
|
"epoch": 0.018731988472622477,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.5487,
|
|
"mean_token_accuracy": 0.07971205189824104,
|
|
"num_tokens": 444859.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.427717113494873,
|
|
"epoch": 0.01921229586935639,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4231,
|
|
"mean_token_accuracy": 0.08752711713314057,
|
|
"num_tokens": 456508.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.312041282653809,
|
|
"epoch": 0.0196926032660903,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4263,
|
|
"mean_token_accuracy": 0.08335452601313591,
|
|
"num_tokens": 469064.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.230023956298828,
|
|
"epoch": 0.020172910662824207,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.347,
|
|
"mean_token_accuracy": 0.08570738211274147,
|
|
"num_tokens": 480743.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.206699085235595,
|
|
"epoch": 0.020653218059558116,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3673,
|
|
"mean_token_accuracy": 0.08695473670959472,
|
|
"num_tokens": 493378.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.153348922729492,
|
|
"epoch": 0.021133525456292025,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.3107,
|
|
"mean_token_accuracy": 0.08663721978664399,
|
|
"num_tokens": 505641.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.109747982025146,
|
|
"epoch": 0.021613832853025938,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3616,
|
|
"mean_token_accuracy": 0.08820611983537674,
|
|
"num_tokens": 517783.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.069980669021607,
|
|
"epoch": 0.022094140249759846,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2154,
|
|
"mean_token_accuracy": 0.09702874794602394,
|
|
"num_tokens": 528407.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.088402652740479,
|
|
"epoch": 0.022574447646493755,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.3335,
|
|
"mean_token_accuracy": 0.0919241763651371,
|
|
"num_tokens": 539974.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.991567897796631,
|
|
"epoch": 0.023054755043227664,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1789,
|
|
"mean_token_accuracy": 0.09620562493801117,
|
|
"num_tokens": 551127.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.010430145263673,
|
|
"epoch": 0.023535062439961577,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1818,
|
|
"mean_token_accuracy": 0.09887873977422715,
|
|
"num_tokens": 561980.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.9398411273956295,
|
|
"epoch": 0.024015369836695485,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2171,
|
|
"mean_token_accuracy": 0.09803635850548745,
|
|
"num_tokens": 573840.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.9045103073120115,
|
|
"epoch": 0.024495677233429394,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1223,
|
|
"mean_token_accuracy": 0.10269570201635361,
|
|
"num_tokens": 585110.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.831203842163086,
|
|
"epoch": 0.024975984630163303,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0168,
|
|
"mean_token_accuracy": 0.10523020774126053,
|
|
"num_tokens": 596902.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.8273866176605225,
|
|
"epoch": 0.025456292026897216,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1703,
|
|
"mean_token_accuracy": 0.1047072984278202,
|
|
"num_tokens": 608738.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.822991180419922,
|
|
"epoch": 0.025936599423631124,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0956,
|
|
"mean_token_accuracy": 0.10134880691766739,
|
|
"num_tokens": 621078.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.900579166412354,
|
|
"epoch": 0.026416906820365033,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1582,
|
|
"mean_token_accuracy": 0.10307180657982826,
|
|
"num_tokens": 632333.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.694871425628662,
|
|
"epoch": 0.026897214217098942,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.9781,
|
|
"mean_token_accuracy": 0.10547222793102265,
|
|
"num_tokens": 644474.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.7671226978302,
|
|
"epoch": 0.027377521613832854,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.05,
|
|
"mean_token_accuracy": 0.10735271498560905,
|
|
"num_tokens": 655608.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.7116927146911625,
|
|
"epoch": 0.027857829010566763,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0705,
|
|
"mean_token_accuracy": 0.10069631710648537,
|
|
"num_tokens": 667371.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.697999811172485,
|
|
"epoch": 0.028338136407300672,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0716,
|
|
"mean_token_accuracy": 0.10165307149291039,
|
|
"num_tokens": 679152.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.703333139419556,
|
|
"epoch": 0.02881844380403458,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.976,
|
|
"mean_token_accuracy": 0.11086443439126015,
|
|
"num_tokens": 690359.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.646430587768554,
|
|
"epoch": 0.029298751200768493,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9582,
|
|
"mean_token_accuracy": 0.10852264314889908,
|
|
"num_tokens": 701819.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.659921550750733,
|
|
"epoch": 0.029779058597502402,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0023,
|
|
"mean_token_accuracy": 0.10957548543810844,
|
|
"num_tokens": 711048.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.590796375274659,
|
|
"epoch": 0.03025936599423631,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8928,
|
|
"mean_token_accuracy": 0.11393210589885712,
|
|
"num_tokens": 721399.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.65145468711853,
|
|
"epoch": 0.03073967339097022,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9061,
|
|
"mean_token_accuracy": 0.11852319762110711,
|
|
"num_tokens": 732144.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.4970308303833,
|
|
"epoch": 0.03121998078770413,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8658,
|
|
"mean_token_accuracy": 0.11263575032353401,
|
|
"num_tokens": 743962.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.605427598953247,
|
|
"epoch": 0.03170028818443804,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.0196,
|
|
"mean_token_accuracy": 0.11022250354290009,
|
|
"num_tokens": 755641.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.507912826538086,
|
|
"epoch": 0.03218059558117195,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8509,
|
|
"mean_token_accuracy": 0.11853420287370682,
|
|
"num_tokens": 766798.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.568133211135864,
|
|
"epoch": 0.03266090297790586,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9301,
|
|
"mean_token_accuracy": 0.11699827089905739,
|
|
"num_tokens": 778097.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.431187677383423,
|
|
"epoch": 0.03314121037463977,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8572,
|
|
"mean_token_accuracy": 0.12058786600828171,
|
|
"num_tokens": 789775.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.508008241653442,
|
|
"epoch": 0.03362151777137368,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9158,
|
|
"mean_token_accuracy": 0.11246205344796181,
|
|
"num_tokens": 800155.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.496634769439697,
|
|
"epoch": 0.034101825168107586,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9179,
|
|
"mean_token_accuracy": 0.11470787152647972,
|
|
"num_tokens": 811213.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.542068147659302,
|
|
"epoch": 0.0345821325648415,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9034,
|
|
"mean_token_accuracy": 0.10915727838873863,
|
|
"num_tokens": 822832.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.420533800125122,
|
|
"epoch": 0.03506243996157541,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8752,
|
|
"mean_token_accuracy": 0.11799765825271606,
|
|
"num_tokens": 834824.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.471309661865234,
|
|
"epoch": 0.03554274735830932,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9009,
|
|
"mean_token_accuracy": 0.11093073859810829,
|
|
"num_tokens": 846969.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.535832786560059,
|
|
"epoch": 0.03602305475504323,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.9809,
|
|
"mean_token_accuracy": 0.10712321475148201,
|
|
"num_tokens": 859271.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.452226400375366,
|
|
"epoch": 0.03650336215177714,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8774,
|
|
"mean_token_accuracy": 0.11831732466816902,
|
|
"num_tokens": 869754.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.40360279083252,
|
|
"epoch": 0.036983669548511046,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8011,
|
|
"mean_token_accuracy": 0.1226475939154625,
|
|
"num_tokens": 880519.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.318375492095948,
|
|
"epoch": 0.037463976945244955,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7696,
|
|
"mean_token_accuracy": 0.11847401708364487,
|
|
"num_tokens": 890765.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.387319278717041,
|
|
"epoch": 0.037944284341978864,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7752,
|
|
"mean_token_accuracy": 0.11672125980257989,
|
|
"num_tokens": 901717.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.397585868835449,
|
|
"epoch": 0.03842459173871278,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8494,
|
|
"mean_token_accuracy": 0.11776698976755143,
|
|
"num_tokens": 913334.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.4451488018035885,
|
|
"epoch": 0.03890489913544669,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8579,
|
|
"mean_token_accuracy": 0.11650485098361969,
|
|
"num_tokens": 924753.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.340163087844848,
|
|
"epoch": 0.0393852065321806,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7891,
|
|
"mean_token_accuracy": 0.12223831489682198,
|
|
"num_tokens": 935692.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.3431860446929935,
|
|
"epoch": 0.039865513928914506,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8737,
|
|
"mean_token_accuracy": 0.11860141828656197,
|
|
"num_tokens": 947902.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.3516919136047365,
|
|
"epoch": 0.040345821325648415,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8063,
|
|
"mean_token_accuracy": 0.12171181738376617,
|
|
"num_tokens": 959437.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.20117769241333,
|
|
"epoch": 0.040826128722382324,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6245,
|
|
"mean_token_accuracy": 0.12898706570267676,
|
|
"num_tokens": 970191.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.262586164474487,
|
|
"epoch": 0.04130643611911623,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7321,
|
|
"mean_token_accuracy": 0.12053022459149361,
|
|
"num_tokens": 981811.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.3112424373626705,
|
|
"epoch": 0.04178674351585014,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8546,
|
|
"mean_token_accuracy": 0.11418978720903397,
|
|
"num_tokens": 994034.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.285151576995849,
|
|
"epoch": 0.04226705091258405,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7999,
|
|
"mean_token_accuracy": 0.11736572235822677,
|
|
"num_tokens": 1005506.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.276934432983398,
|
|
"epoch": 0.042747358309317966,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7517,
|
|
"mean_token_accuracy": 0.11421283110976219,
|
|
"num_tokens": 1017103.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.244399690628052,
|
|
"epoch": 0.043227665706051875,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.84,
|
|
"mean_token_accuracy": 0.11778046488761902,
|
|
"num_tokens": 1029606.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.335804605484009,
|
|
"epoch": 0.043707973102785784,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.789,
|
|
"mean_token_accuracy": 0.11802555918693543,
|
|
"num_tokens": 1041899.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.2514319896698,
|
|
"epoch": 0.04418828049951969,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7092,
|
|
"mean_token_accuracy": 0.12920955047011376,
|
|
"num_tokens": 1053577.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.230760812759399,
|
|
"epoch": 0.0446685878962536,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7625,
|
|
"mean_token_accuracy": 0.12477129399776458,
|
|
"num_tokens": 1065541.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.258627557754517,
|
|
"epoch": 0.04514889529298751,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7034,
|
|
"mean_token_accuracy": 0.121210016310215,
|
|
"num_tokens": 1076138.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.243265295028687,
|
|
"epoch": 0.04562920268972142,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7561,
|
|
"mean_token_accuracy": 0.12767059430480004,
|
|
"num_tokens": 1087620.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.192060947418213,
|
|
"epoch": 0.04610951008645533,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6443,
|
|
"mean_token_accuracy": 0.12917531430721282,
|
|
"num_tokens": 1099493.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.143154096603394,
|
|
"epoch": 0.046589817483189244,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.696,
|
|
"mean_token_accuracy": 0.12751599624753,
|
|
"num_tokens": 1110559.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.090892791748047,
|
|
"epoch": 0.04707012487992315,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6684,
|
|
"mean_token_accuracy": 0.1201030507683754,
|
|
"num_tokens": 1121051.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.2134480476379395,
|
|
"epoch": 0.04755043227665706,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6299,
|
|
"mean_token_accuracy": 0.12791844457387924,
|
|
"num_tokens": 1132305.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.121208190917969,
|
|
"epoch": 0.04803073967339097,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6354,
|
|
"mean_token_accuracy": 0.13059915378689765,
|
|
"num_tokens": 1142979.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.124688339233399,
|
|
"epoch": 0.04851104707012488,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.6429,
|
|
"mean_token_accuracy": 0.13205073550343513,
|
|
"num_tokens": 1153889.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.173402786254883,
|
|
"epoch": 0.04899135446685879,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.6555,
|
|
"mean_token_accuracy": 0.12510655000805854,
|
|
"num_tokens": 1166030.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.213367700576782,
|
|
"epoch": 0.0494716618635927,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.8218,
|
|
"mean_token_accuracy": 0.11716388240456581,
|
|
"num_tokens": 1177545.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.142013311386108,
|
|
"epoch": 0.049951969260326606,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.63,
|
|
"mean_token_accuracy": 0.13155306726694108,
|
|
"num_tokens": 1188944.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.023855352401734,
|
|
"epoch": 0.05043227665706052,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6974,
|
|
"mean_token_accuracy": 0.12238200083374977,
|
|
"num_tokens": 1199517.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.1075825691223145,
|
|
"epoch": 0.05091258405379443,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6283,
|
|
"mean_token_accuracy": 0.12261475175619126,
|
|
"num_tokens": 1211536.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.080765342712402,
|
|
"epoch": 0.05139289145052834,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5351,
|
|
"mean_token_accuracy": 0.13503727167844773,
|
|
"num_tokens": 1224159.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.064063882827758,
|
|
"epoch": 0.05187319884726225,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6023,
|
|
"mean_token_accuracy": 0.13505509868264198,
|
|
"num_tokens": 1235226.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.969502115249634,
|
|
"epoch": 0.05235350624399616,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5304,
|
|
"mean_token_accuracy": 0.1370991587638855,
|
|
"num_tokens": 1245862.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.037106370925903,
|
|
"epoch": 0.052833813640730067,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5373,
|
|
"mean_token_accuracy": 0.1353856362402439,
|
|
"num_tokens": 1257512.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.028940439224243,
|
|
"epoch": 0.053314121037463975,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6458,
|
|
"mean_token_accuracy": 0.1271481417119503,
|
|
"num_tokens": 1269317.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.974031114578247,
|
|
"epoch": 0.053794428434197884,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.5319,
|
|
"mean_token_accuracy": 0.12299542650580406,
|
|
"num_tokens": 1281878.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.952550315856934,
|
|
"epoch": 0.05427473583093179,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.5502,
|
|
"mean_token_accuracy": 0.13390207588672637,
|
|
"num_tokens": 1293178.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.94937572479248,
|
|
"epoch": 0.05475504322766571,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.533,
|
|
"mean_token_accuracy": 0.13220486268401146,
|
|
"num_tokens": 1304969.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.081587600708008,
|
|
"epoch": 0.05523535062439962,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5722,
|
|
"mean_token_accuracy": 0.13268239945173263,
|
|
"num_tokens": 1317301.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.960885334014892,
|
|
"epoch": 0.05571565802113353,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5506,
|
|
"mean_token_accuracy": 0.13222013115882875,
|
|
"num_tokens": 1327688.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.99568247795105,
|
|
"epoch": 0.056195965417867436,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5001,
|
|
"mean_token_accuracy": 0.13380020037293433,
|
|
"num_tokens": 1338750.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.9811159610748295,
|
|
"epoch": 0.056676272814601344,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6382,
|
|
"mean_token_accuracy": 0.12341757640242576,
|
|
"num_tokens": 1350899.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.057421350479126,
|
|
"epoch": 0.05715658021133525,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5033,
|
|
"mean_token_accuracy": 0.13546376153826714,
|
|
"num_tokens": 1363184.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.953424024581909,
|
|
"epoch": 0.05763688760806916,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4861,
|
|
"mean_token_accuracy": 0.1358728751540184,
|
|
"num_tokens": 1374192.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.766305589675904,
|
|
"epoch": 0.05811719500480307,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4998,
|
|
"mean_token_accuracy": 0.13279664665460586,
|
|
"num_tokens": 1384888.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.100843906402588,
|
|
"epoch": 0.05859750240153699,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.5908,
|
|
"mean_token_accuracy": 0.13604374155402182,
|
|
"num_tokens": 1395139.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.812396669387818,
|
|
"epoch": 0.059077809798270896,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.3873,
|
|
"mean_token_accuracy": 0.13559869676828384,
|
|
"num_tokens": 1407219.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.959701156616211,
|
|
"epoch": 0.059558117195004805,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4986,
|
|
"mean_token_accuracy": 0.13413800969719886,
|
|
"num_tokens": 1418917.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.8605883598327635,
|
|
"epoch": 0.060038424591738714,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4729,
|
|
"mean_token_accuracy": 0.13215331062674524,
|
|
"num_tokens": 1430792.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.870712947845459,
|
|
"epoch": 0.06051873198847262,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.4879,
|
|
"mean_token_accuracy": 0.1390850931406021,
|
|
"num_tokens": 1442825.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.9546905040740965,
|
|
"epoch": 0.06099903938520653,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5848,
|
|
"mean_token_accuracy": 0.1320830576121807,
|
|
"num_tokens": 1454078.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.909980916976929,
|
|
"epoch": 0.06147934678194044,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5095,
|
|
"mean_token_accuracy": 0.13304588049650193,
|
|
"num_tokens": 1465164.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.899385166168213,
|
|
"epoch": 0.06195965417867435,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4964,
|
|
"mean_token_accuracy": 0.13703863248229026,
|
|
"num_tokens": 1476813.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.867896032333374,
|
|
"epoch": 0.06243996157540826,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5169,
|
|
"mean_token_accuracy": 0.13063866794109344,
|
|
"num_tokens": 1488547.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.025978374481201,
|
|
"epoch": 0.06292026897214217,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.522,
|
|
"mean_token_accuracy": 0.13043418899178505,
|
|
"num_tokens": 1500714.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.821457815170288,
|
|
"epoch": 0.06340057636887608,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5044,
|
|
"mean_token_accuracy": 0.13662170022726058,
|
|
"num_tokens": 1512212.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.883290338516235,
|
|
"epoch": 0.06388088376560999,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.471,
|
|
"mean_token_accuracy": 0.13708159923553467,
|
|
"num_tokens": 1523610.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.790850353240967,
|
|
"epoch": 0.0643611911623439,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4375,
|
|
"mean_token_accuracy": 0.13195957690477372,
|
|
"num_tokens": 1535543.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.966132020950317,
|
|
"epoch": 0.06484149855907781,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6641,
|
|
"mean_token_accuracy": 0.12745087668299676,
|
|
"num_tokens": 1547183.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.799446535110474,
|
|
"epoch": 0.06532180595581172,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4042,
|
|
"mean_token_accuracy": 0.14067018032073975,
|
|
"num_tokens": 1559265.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.883531522750855,
|
|
"epoch": 0.06580211335254563,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.435,
|
|
"mean_token_accuracy": 0.13348256796598434,
|
|
"num_tokens": 1569774.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.758243894577026,
|
|
"epoch": 0.06628242074927954,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3664,
|
|
"mean_token_accuracy": 0.14002081379294395,
|
|
"num_tokens": 1579408.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.737592172622681,
|
|
"epoch": 0.06676272814601344,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.3952,
|
|
"mean_token_accuracy": 0.14295617043972014,
|
|
"num_tokens": 1591849.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.807192468643189,
|
|
"epoch": 0.06724303554274735,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4707,
|
|
"mean_token_accuracy": 0.13754646629095077,
|
|
"num_tokens": 1603269.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.934098720550537,
|
|
"epoch": 0.06772334293948126,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5128,
|
|
"mean_token_accuracy": 0.12959487438201905,
|
|
"num_tokens": 1613639.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.877666711807251,
|
|
"epoch": 0.06820365033621517,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5304,
|
|
"mean_token_accuracy": 0.13218488544225693,
|
|
"num_tokens": 1625393.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.748217916488647,
|
|
"epoch": 0.0686839577329491,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.3657,
|
|
"mean_token_accuracy": 0.13785189315676688,
|
|
"num_tokens": 1636355.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.8460822105407715,
|
|
"epoch": 0.069164265129683,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4009,
|
|
"mean_token_accuracy": 0.13792139366269113,
|
|
"num_tokens": 1646157.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.716140365600586,
|
|
"epoch": 0.06964457252641691,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3961,
|
|
"mean_token_accuracy": 0.13261248767375947,
|
|
"num_tokens": 1657827.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.787322759628296,
|
|
"epoch": 0.07012487992315082,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4788,
|
|
"mean_token_accuracy": 0.14132225662469863,
|
|
"num_tokens": 1669023.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.8325389385223385,
|
|
"epoch": 0.07060518731988473,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.4273,
|
|
"mean_token_accuracy": 0.13686123490333557,
|
|
"num_tokens": 1680048.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.897904634475708,
|
|
"epoch": 0.07108549471661864,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.6169,
|
|
"mean_token_accuracy": 0.12690431475639344,
|
|
"num_tokens": 1690907.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.893639278411865,
|
|
"epoch": 0.07156580211335255,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4363,
|
|
"mean_token_accuracy": 0.13882233425974846,
|
|
"num_tokens": 1702337.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.728409624099731,
|
|
"epoch": 0.07204610951008646,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3874,
|
|
"mean_token_accuracy": 0.13424424231052398,
|
|
"num_tokens": 1713242.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.744375705718994,
|
|
"epoch": 0.07252641690682037,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.3574,
|
|
"mean_token_accuracy": 0.13707793802022933,
|
|
"num_tokens": 1724939.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.712884521484375,
|
|
"epoch": 0.07300672430355427,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3049,
|
|
"mean_token_accuracy": 0.14935051500797272,
|
|
"num_tokens": 1737311.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.824639749526978,
|
|
"epoch": 0.07348703170028818,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4514,
|
|
"mean_token_accuracy": 0.13682929798960686,
|
|
"num_tokens": 1749528.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.68354549407959,
|
|
"epoch": 0.07396733909702209,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3275,
|
|
"mean_token_accuracy": 0.14307046681642532,
|
|
"num_tokens": 1760304.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.660215139389038,
|
|
"epoch": 0.074447646493756,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3171,
|
|
"mean_token_accuracy": 0.14092384576797484,
|
|
"num_tokens": 1772014.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.7041209697723385,
|
|
"epoch": 0.07492795389048991,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3679,
|
|
"mean_token_accuracy": 0.1367984175682068,
|
|
"num_tokens": 1784585.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.728722429275512,
|
|
"epoch": 0.07540826128722382,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3004,
|
|
"mean_token_accuracy": 0.146743543446064,
|
|
"num_tokens": 1795246.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.596188020706177,
|
|
"epoch": 0.07588856868395773,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.2942,
|
|
"mean_token_accuracy": 0.14472177475690842,
|
|
"num_tokens": 1806826.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.763730049133301,
|
|
"epoch": 0.07636887608069164,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.4189,
|
|
"mean_token_accuracy": 0.14111077263951302,
|
|
"num_tokens": 1818766.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.678760385513305,
|
|
"epoch": 0.07684918347742556,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.2816,
|
|
"mean_token_accuracy": 0.15023117810487746,
|
|
"num_tokens": 1830824.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.609415769577026,
|
|
"epoch": 0.07732949087415947,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3689,
|
|
"mean_token_accuracy": 0.13778330758213997,
|
|
"num_tokens": 1841792.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.652772665023804,
|
|
"epoch": 0.07780979827089338,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3008,
|
|
"mean_token_accuracy": 0.14102251529693605,
|
|
"num_tokens": 1854683.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.727771043777466,
|
|
"epoch": 0.07829010566762729,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3995,
|
|
"mean_token_accuracy": 0.13676784187555313,
|
|
"num_tokens": 1865826.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.702198505401611,
|
|
"epoch": 0.0787704130643612,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3485,
|
|
"mean_token_accuracy": 0.14424023181200027,
|
|
"num_tokens": 1876759.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.572908020019531,
|
|
"epoch": 0.0792507204610951,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3879,
|
|
"mean_token_accuracy": 0.1390812113881111,
|
|
"num_tokens": 1888013.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.834191608428955,
|
|
"epoch": 0.07973102785782901,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3987,
|
|
"mean_token_accuracy": 0.1389234647154808,
|
|
"num_tokens": 1898964.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.619566154479981,
|
|
"epoch": 0.08021133525456292,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3096,
|
|
"mean_token_accuracy": 0.14486038088798522,
|
|
"num_tokens": 1909728.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.573022413253784,
|
|
"epoch": 0.08069164265129683,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.2396,
|
|
"mean_token_accuracy": 0.14724621176719666,
|
|
"num_tokens": 1919895.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.622575473785401,
|
|
"epoch": 0.08117195004803074,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.1686,
|
|
"mean_token_accuracy": 0.1482459105551243,
|
|
"num_tokens": 1930839.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.6008491039276125,
|
|
"epoch": 0.08165225744476465,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3622,
|
|
"mean_token_accuracy": 0.1424940586090088,
|
|
"num_tokens": 1942605.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.617079734802246,
|
|
"epoch": 0.08213256484149856,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3529,
|
|
"mean_token_accuracy": 0.13537877276539803,
|
|
"num_tokens": 1954863.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.722651815414428,
|
|
"epoch": 0.08261287223823247,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4252,
|
|
"mean_token_accuracy": 0.1417405441403389,
|
|
"num_tokens": 1966088.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.604581260681153,
|
|
"epoch": 0.08309317963496637,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.2567,
|
|
"mean_token_accuracy": 0.14236779659986495,
|
|
"num_tokens": 1978081.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.656841897964478,
|
|
"epoch": 0.08357348703170028,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3249,
|
|
"mean_token_accuracy": 0.1358143836259842,
|
|
"num_tokens": 1991212.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.694861984252929,
|
|
"epoch": 0.08405379442843419,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.311,
|
|
"mean_token_accuracy": 0.14360752776265145,
|
|
"num_tokens": 2002074.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.539075565338135,
|
|
"epoch": 0.0845341018251681,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2363,
|
|
"mean_token_accuracy": 0.14888798743486403,
|
|
"num_tokens": 2013039.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.617077302932739,
|
|
"epoch": 0.08501440922190202,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.3741,
|
|
"mean_token_accuracy": 0.1400503121316433,
|
|
"num_tokens": 2024744.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.588595581054688,
|
|
"epoch": 0.08549471661863593,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2358,
|
|
"mean_token_accuracy": 0.1453889586031437,
|
|
"num_tokens": 2035587.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.616622066497802,
|
|
"epoch": 0.08597502401536984,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.3266,
|
|
"mean_token_accuracy": 0.13622945696115493,
|
|
"num_tokens": 2047924.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.539765977859497,
|
|
"epoch": 0.08645533141210375,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2513,
|
|
"mean_token_accuracy": 0.14483485221862794,
|
|
"num_tokens": 2060161.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.607950258255005,
|
|
"epoch": 0.08693563880883766,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2629,
|
|
"mean_token_accuracy": 0.14627112001180648,
|
|
"num_tokens": 2071674.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.590583610534668,
|
|
"epoch": 0.08741594620557157,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3995,
|
|
"mean_token_accuracy": 0.1397729016840458,
|
|
"num_tokens": 2084025.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.667711496353149,
|
|
"epoch": 0.08789625360230548,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3207,
|
|
"mean_token_accuracy": 0.14221117421984672,
|
|
"num_tokens": 2094501.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.599315881729126,
|
|
"epoch": 0.08837656099903939,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3056,
|
|
"mean_token_accuracy": 0.14709803313016892,
|
|
"num_tokens": 2106207.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.5739052295684814,
|
|
"epoch": 0.0888568683957733,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2205,
|
|
"mean_token_accuracy": 0.1523227147758007,
|
|
"num_tokens": 2116879.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.544489240646362,
|
|
"epoch": 0.0893371757925072,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2681,
|
|
"mean_token_accuracy": 0.1414512149989605,
|
|
"num_tokens": 2128884.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.482966566085816,
|
|
"epoch": 0.08981748318924111,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2153,
|
|
"mean_token_accuracy": 0.14321899712085723,
|
|
"num_tokens": 2140810.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.64809398651123,
|
|
"epoch": 0.09029779058597502,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2752,
|
|
"mean_token_accuracy": 0.13929763957858085,
|
|
"num_tokens": 2152312.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.420036125183105,
|
|
"epoch": 0.09077809798270893,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.2228,
|
|
"mean_token_accuracy": 0.14043607711791992,
|
|
"num_tokens": 2164408.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.556408071517945,
|
|
"epoch": 0.09125840537944284,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.2407,
|
|
"mean_token_accuracy": 0.14646506905555726,
|
|
"num_tokens": 2176050.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.423853635787964,
|
|
"epoch": 0.09173871277617675,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1228,
|
|
"mean_token_accuracy": 0.15012319162487983,
|
|
"num_tokens": 2187354.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.393097972869873,
|
|
"epoch": 0.09221902017291066,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1458,
|
|
"mean_token_accuracy": 0.14260236471891402,
|
|
"num_tokens": 2198686.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.461620712280274,
|
|
"epoch": 0.09269932756964457,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.186,
|
|
"mean_token_accuracy": 0.14377033933997155,
|
|
"num_tokens": 2209780.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.521939277648926,
|
|
"epoch": 0.09317963496637849,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2754,
|
|
"mean_token_accuracy": 0.15148081481456757,
|
|
"num_tokens": 2221760.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.4953220844268795,
|
|
"epoch": 0.0936599423631124,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.2828,
|
|
"mean_token_accuracy": 0.1426544666290283,
|
|
"num_tokens": 2233767.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.553147888183593,
|
|
"epoch": 0.0941402497598463,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2132,
|
|
"mean_token_accuracy": 0.1449471965432167,
|
|
"num_tokens": 2244841.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.426078653335571,
|
|
"epoch": 0.09462055715658022,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.068,
|
|
"mean_token_accuracy": 0.15349577814340593,
|
|
"num_tokens": 2255620.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.492886877059936,
|
|
"epoch": 0.09510086455331412,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.2664,
|
|
"mean_token_accuracy": 0.15196479558944703,
|
|
"num_tokens": 2266345.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.406527471542359,
|
|
"epoch": 0.09558117195004803,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1952,
|
|
"mean_token_accuracy": 0.14961037039756775,
|
|
"num_tokens": 2277190.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.483725643157959,
|
|
"epoch": 0.09606147934678194,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0722,
|
|
"mean_token_accuracy": 0.15189518705010413,
|
|
"num_tokens": 2288241.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.433297443389892,
|
|
"epoch": 0.09654178674351585,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999999983283737,
|
|
"loss": 6.2272,
|
|
"mean_token_accuracy": 0.144133760035038,
|
|
"num_tokens": 2301092.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.4213059425354,
|
|
"epoch": 0.09702209414024976,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999999915373924,
|
|
"loss": 6.1347,
|
|
"mean_token_accuracy": 0.1475769981741905,
|
|
"num_tokens": 2313622.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.370494604110718,
|
|
"epoch": 0.09750240153698367,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999999795225793,
|
|
"loss": 6.0938,
|
|
"mean_token_accuracy": 0.15354661494493485,
|
|
"num_tokens": 2325699.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.43495192527771,
|
|
"epoch": 0.09798270893371758,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999999622839347,
|
|
"loss": 6.1704,
|
|
"mean_token_accuracy": 0.1460404135286808,
|
|
"num_tokens": 2337816.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.590832901000977,
|
|
"epoch": 0.09846301633045149,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999999398214593,
|
|
"loss": 6.2957,
|
|
"mean_token_accuracy": 0.14283925592899321,
|
|
"num_tokens": 2348793.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.46081771850586,
|
|
"epoch": 0.0989433237271854,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999999121351532,
|
|
"loss": 6.1991,
|
|
"mean_token_accuracy": 0.14770057797431946,
|
|
"num_tokens": 2360209.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.489052057266235,
|
|
"epoch": 0.0994236311239193,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999998792250173,
|
|
"loss": 6.2072,
|
|
"mean_token_accuracy": 0.14201200753450394,
|
|
"num_tokens": 2370597.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.489638328552246,
|
|
"epoch": 0.09990393852065321,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999998410910524,
|
|
"loss": 6.2478,
|
|
"mean_token_accuracy": 0.1417911544442177,
|
|
"num_tokens": 2381466.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.575153255462647,
|
|
"epoch": 0.10038424591738712,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999997977332592,
|
|
"loss": 6.3352,
|
|
"mean_token_accuracy": 0.138764651119709,
|
|
"num_tokens": 2394507.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.458651733398438,
|
|
"epoch": 0.10086455331412104,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999997491516389,
|
|
"loss": 6.1915,
|
|
"mean_token_accuracy": 0.14835959821939468,
|
|
"num_tokens": 2406367.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.432104206085205,
|
|
"epoch": 0.10134486071085495,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999996953461925,
|
|
"loss": 6.1627,
|
|
"mean_token_accuracy": 0.14652103781700135,
|
|
"num_tokens": 2416962.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.524100685119629,
|
|
"epoch": 0.10182516810758886,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999996363169212,
|
|
"loss": 6.2133,
|
|
"mean_token_accuracy": 0.13934570401906968,
|
|
"num_tokens": 2429441.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.394040298461914,
|
|
"epoch": 0.10230547550432277,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999995720638266,
|
|
"loss": 6.0984,
|
|
"mean_token_accuracy": 0.1505374975502491,
|
|
"num_tokens": 2441360.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.361066913604736,
|
|
"epoch": 0.10278578290105668,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00049999950258691,
|
|
"loss": 6.1226,
|
|
"mean_token_accuracy": 0.14372813031077386,
|
|
"num_tokens": 2452928.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.374180650711059,
|
|
"epoch": 0.10326609029779059,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999994278861731,
|
|
"loss": 6.0498,
|
|
"mean_token_accuracy": 0.15205594003200532,
|
|
"num_tokens": 2464218.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.407055807113648,
|
|
"epoch": 0.1037463976945245,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999993479616175,
|
|
"loss": 6.1292,
|
|
"mean_token_accuracy": 0.15480452179908752,
|
|
"num_tokens": 2475084.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.378732442855835,
|
|
"epoch": 0.1042267050912584,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004999992628132451,
|
|
"loss": 6.1464,
|
|
"mean_token_accuracy": 0.14829933345317842,
|
|
"num_tokens": 2486845.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.37392635345459,
|
|
"epoch": 0.10470701248799232,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999991724410582,
|
|
"loss": 6.2167,
|
|
"mean_token_accuracy": 0.14155052751302719,
|
|
"num_tokens": 2498049.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.389601182937622,
|
|
"epoch": 0.10518731988472622,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999990768450583,
|
|
"loss": 6.1558,
|
|
"mean_token_accuracy": 0.15035107880830764,
|
|
"num_tokens": 2509511.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.484518432617188,
|
|
"epoch": 0.10566762728146013,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999989760252482,
|
|
"loss": 6.109,
|
|
"mean_token_accuracy": 0.156458543241024,
|
|
"num_tokens": 2521518.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.210693788528443,
|
|
"epoch": 0.10614793467819404,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999988699816299,
|
|
"loss": 5.9549,
|
|
"mean_token_accuracy": 0.1586344212293625,
|
|
"num_tokens": 2533141.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.256671905517578,
|
|
"epoch": 0.10662824207492795,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999987587142058,
|
|
"loss": 5.949,
|
|
"mean_token_accuracy": 0.1536861389875412,
|
|
"num_tokens": 2542907.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.330707740783692,
|
|
"epoch": 0.10710854947166186,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999986422229789,
|
|
"loss": 6.1024,
|
|
"mean_token_accuracy": 0.14371281042695044,
|
|
"num_tokens": 2554497.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.479738664627075,
|
|
"epoch": 0.10758885686839577,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999985205079514,
|
|
"loss": 6.2156,
|
|
"mean_token_accuracy": 0.1489127516746521,
|
|
"num_tokens": 2567035.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.258924865722657,
|
|
"epoch": 0.10806916426512968,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999983935691265,
|
|
"loss": 5.9988,
|
|
"mean_token_accuracy": 0.15842699110507966,
|
|
"num_tokens": 2578003.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.524848222732544,
|
|
"epoch": 0.10854947166186359,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000499998261406507,
|
|
"loss": 6.305,
|
|
"mean_token_accuracy": 0.14301338717341422,
|
|
"num_tokens": 2590272.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.27402629852295,
|
|
"epoch": 0.10902977905859751,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999981240200958,
|
|
"loss": 6.1217,
|
|
"mean_token_accuracy": 0.14725035130977632,
|
|
"num_tokens": 2602874.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.340481901168824,
|
|
"epoch": 0.10951008645533142,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999979814098966,
|
|
"loss": 6.0716,
|
|
"mean_token_accuracy": 0.15673150569200517,
|
|
"num_tokens": 2613835.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.4138932704925535,
|
|
"epoch": 0.10999039385206533,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999978335759121,
|
|
"loss": 6.1849,
|
|
"mean_token_accuracy": 0.14377753883600236,
|
|
"num_tokens": 2624986.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.416733598709106,
|
|
"epoch": 0.11047070124879924,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999976805181461,
|
|
"loss": 6.1885,
|
|
"mean_token_accuracy": 0.13901318833231927,
|
|
"num_tokens": 2637654.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.371509313583374,
|
|
"epoch": 0.11095100864553314,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499997522236602,
|
|
"loss": 6.0905,
|
|
"mean_token_accuracy": 0.1510764852166176,
|
|
"num_tokens": 2648947.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.312252235412598,
|
|
"epoch": 0.11143131604226705,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999973587312837,
|
|
"loss": 6.087,
|
|
"mean_token_accuracy": 0.14612228646874428,
|
|
"num_tokens": 2659573.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.283674192428589,
|
|
"epoch": 0.11191162343900096,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999971900021947,
|
|
"loss": 6.0625,
|
|
"mean_token_accuracy": 0.15488046854734422,
|
|
"num_tokens": 2671491.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.3300315856933596,
|
|
"epoch": 0.11239193083573487,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999970160493391,
|
|
"loss": 6.0533,
|
|
"mean_token_accuracy": 0.1487934708595276,
|
|
"num_tokens": 2682705.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.23258752822876,
|
|
"epoch": 0.11287223823246878,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999968368727209,
|
|
"loss": 5.9906,
|
|
"mean_token_accuracy": 0.15133922249078752,
|
|
"num_tokens": 2693539.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.278180265426636,
|
|
"epoch": 0.11335254562920269,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999966524723442,
|
|
"loss": 6.0406,
|
|
"mean_token_accuracy": 0.1541109189391136,
|
|
"num_tokens": 2705403.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.453830718994141,
|
|
"epoch": 0.1138328530259366,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999964628482135,
|
|
"loss": 6.1408,
|
|
"mean_token_accuracy": 0.14809799268841745,
|
|
"num_tokens": 2717320.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.294747066497803,
|
|
"epoch": 0.1143131604226705,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999962680003328,
|
|
"loss": 6.1135,
|
|
"mean_token_accuracy": 0.14518905133008958,
|
|
"num_tokens": 2730056.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.351133012771607,
|
|
"epoch": 0.11479346781940442,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000499996067928707,
|
|
"loss": 5.9844,
|
|
"mean_token_accuracy": 0.15574622750282288,
|
|
"num_tokens": 2741399.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.388057708740234,
|
|
"epoch": 0.11527377521613832,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999958626333406,
|
|
"loss": 6.2151,
|
|
"mean_token_accuracy": 0.1420239381492138,
|
|
"num_tokens": 2752277.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.356492948532105,
|
|
"epoch": 0.11575408261287223,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999956521142383,
|
|
"loss": 6.1025,
|
|
"mean_token_accuracy": 0.14670252203941345,
|
|
"num_tokens": 2763172.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.38686170578003,
|
|
"epoch": 0.11623439000960614,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999954363714051,
|
|
"loss": 6.1064,
|
|
"mean_token_accuracy": 0.1515388995409012,
|
|
"num_tokens": 2775096.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.348104858398438,
|
|
"epoch": 0.11671469740634005,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999952154048459,
|
|
"loss": 6.192,
|
|
"mean_token_accuracy": 0.145056813955307,
|
|
"num_tokens": 2786007.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.295395755767823,
|
|
"epoch": 0.11719500480307397,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499994989214566,
|
|
"loss": 6.0018,
|
|
"mean_token_accuracy": 0.15835101902484894,
|
|
"num_tokens": 2797265.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.344145345687866,
|
|
"epoch": 0.11767531219980788,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999947578005705,
|
|
"loss": 6.1591,
|
|
"mean_token_accuracy": 0.15130333825945855,
|
|
"num_tokens": 2809427.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.2863054275512695,
|
|
"epoch": 0.11815561959654179,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999945211628648,
|
|
"loss": 6.0872,
|
|
"mean_token_accuracy": 0.15372449159622192,
|
|
"num_tokens": 2822880.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.415712070465088,
|
|
"epoch": 0.1186359269932757,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999942793014544,
|
|
"loss": 6.172,
|
|
"mean_token_accuracy": 0.14482916817069053,
|
|
"num_tokens": 2834760.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.345460033416748,
|
|
"epoch": 0.11911623439000961,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499994032216345,
|
|
"loss": 6.115,
|
|
"mean_token_accuracy": 0.14920846298336982,
|
|
"num_tokens": 2846517.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.268563079833984,
|
|
"epoch": 0.11959654178674352,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999937799075422,
|
|
"loss": 5.9979,
|
|
"mean_token_accuracy": 0.15922427102923392,
|
|
"num_tokens": 2857280.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.290250253677368,
|
|
"epoch": 0.12007684918347743,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000499993522375052,
|
|
"loss": 6.0479,
|
|
"mean_token_accuracy": 0.15346607118844985,
|
|
"num_tokens": 2869735.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.315689992904663,
|
|
"epoch": 0.12055715658021134,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999932596188802,
|
|
"loss": 6.0386,
|
|
"mean_token_accuracy": 0.1542970433831215,
|
|
"num_tokens": 2881681.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.181022119522095,
|
|
"epoch": 0.12103746397694524,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999929916390331,
|
|
"loss": 6.056,
|
|
"mean_token_accuracy": 0.15502342879772185,
|
|
"num_tokens": 2893828.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.151469087600708,
|
|
"epoch": 0.12151777137367915,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999927184355169,
|
|
"loss": 6.0066,
|
|
"mean_token_accuracy": 0.15098301693797112,
|
|
"num_tokens": 2905804.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.467128086090088,
|
|
"epoch": 0.12199807877041306,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999924400083377,
|
|
"loss": 6.1586,
|
|
"mean_token_accuracy": 0.1455710545182228,
|
|
"num_tokens": 2918709.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.319386768341064,
|
|
"epoch": 0.12247838616714697,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999921563575022,
|
|
"loss": 6.0642,
|
|
"mean_token_accuracy": 0.15271421596407891,
|
|
"num_tokens": 2930589.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.2710630893707275,
|
|
"epoch": 0.12295869356388088,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999918674830169,
|
|
"loss": 6.1113,
|
|
"mean_token_accuracy": 0.15638522952795028,
|
|
"num_tokens": 2940635.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.209631776809692,
|
|
"epoch": 0.12343900096061479,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999915733848886,
|
|
"loss": 5.9893,
|
|
"mean_token_accuracy": 0.16009382605552674,
|
|
"num_tokens": 2952731.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.254339933395386,
|
|
"epoch": 0.1239193083573487,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499991274063124,
|
|
"loss": 6.0027,
|
|
"mean_token_accuracy": 0.15292049050331116,
|
|
"num_tokens": 2964544.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.232159233093261,
|
|
"epoch": 0.12439961575408261,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999909695177301,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.15449643433094024,
|
|
"num_tokens": 2974996.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.179900312423706,
|
|
"epoch": 0.12487992315081652,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499990659748714,
|
|
"loss": 5.9432,
|
|
"mean_token_accuracy": 0.1571894556283951,
|
|
"num_tokens": 2986496.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.219573736190796,
|
|
"epoch": 0.12536023054755044,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999903447560828,
|
|
"loss": 6.0192,
|
|
"mean_token_accuracy": 0.15346220433712005,
|
|
"num_tokens": 2997693.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.187070655822754,
|
|
"epoch": 0.12584053794428435,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999900245398439,
|
|
"loss": 5.9838,
|
|
"mean_token_accuracy": 0.15567831546068192,
|
|
"num_tokens": 3009155.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.1994123458862305,
|
|
"epoch": 0.12632084534101826,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999896991000047,
|
|
"loss": 5.8983,
|
|
"mean_token_accuracy": 0.16280196905136107,
|
|
"num_tokens": 3020944.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.196508884429932,
|
|
"epoch": 0.12680115273775217,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999893684365729,
|
|
"loss": 5.9764,
|
|
"mean_token_accuracy": 0.15572714507579805,
|
|
"num_tokens": 3032893.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.2522900104522705,
|
|
"epoch": 0.12728146013448607,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999890325495559,
|
|
"loss": 6.0422,
|
|
"mean_token_accuracy": 0.15061216205358505,
|
|
"num_tokens": 3044937.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.236337089538575,
|
|
"epoch": 0.12776176753121998,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999886914389617,
|
|
"loss": 5.9482,
|
|
"mean_token_accuracy": 0.15663958489894866,
|
|
"num_tokens": 3056071.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.207484483718872,
|
|
"epoch": 0.1282420749279539,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004999883451047981,
|
|
"loss": 6.1033,
|
|
"mean_token_accuracy": 0.14877638295292855,
|
|
"num_tokens": 3067724.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.282546806335449,
|
|
"epoch": 0.1287223823246878,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999879935470733,
|
|
"loss": 6.0031,
|
|
"mean_token_accuracy": 0.15774892270565033,
|
|
"num_tokens": 3080779.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.203645133972168,
|
|
"epoch": 0.1292026897214217,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999876367657954,
|
|
"loss": 5.983,
|
|
"mean_token_accuracy": 0.15141114294528962,
|
|
"num_tokens": 3091991.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.1824891567230225,
|
|
"epoch": 0.12968299711815562,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999872747609725,
|
|
"loss": 5.9692,
|
|
"mean_token_accuracy": 0.1594395488500595,
|
|
"num_tokens": 3103713.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.157753753662109,
|
|
"epoch": 0.13016330451488953,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999869075326132,
|
|
"loss": 5.9986,
|
|
"mean_token_accuracy": 0.15652020871639252,
|
|
"num_tokens": 3115891.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.1673017024993895,
|
|
"epoch": 0.13064361191162344,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499986535080726,
|
|
"loss": 5.863,
|
|
"mean_token_accuracy": 0.16944663524627684,
|
|
"num_tokens": 3126223.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.263637685775757,
|
|
"epoch": 0.13112391930835735,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999861574053196,
|
|
"loss": 6.0469,
|
|
"mean_token_accuracy": 0.15185227915644645,
|
|
"num_tokens": 3137537.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.291316413879395,
|
|
"epoch": 0.13160422670509125,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999857745064027,
|
|
"loss": 6.0566,
|
|
"mean_token_accuracy": 0.15476293712854386,
|
|
"num_tokens": 3150126.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.224523258209229,
|
|
"epoch": 0.13208453410182516,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000499985386383984,
|
|
"loss": 6.0252,
|
|
"mean_token_accuracy": 0.16093770563602447,
|
|
"num_tokens": 3161684.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.1495531558990475,
|
|
"epoch": 0.13256484149855907,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999849930380729,
|
|
"loss": 5.9115,
|
|
"mean_token_accuracy": 0.1610415458679199,
|
|
"num_tokens": 3172342.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.276655769348144,
|
|
"epoch": 0.13304514889529298,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999845944686781,
|
|
"loss": 6.0493,
|
|
"mean_token_accuracy": 0.16042522042989732,
|
|
"num_tokens": 3183681.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.0783570289611815,
|
|
"epoch": 0.1335254562920269,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999841906758093,
|
|
"loss": 5.8443,
|
|
"mean_token_accuracy": 0.16978892534971238,
|
|
"num_tokens": 3194122.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.182632637023926,
|
|
"epoch": 0.1340057636887608,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999837816594757,
|
|
"loss": 5.9101,
|
|
"mean_token_accuracy": 0.15853887796401978,
|
|
"num_tokens": 3205719.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.120689725875854,
|
|
"epoch": 0.1344860710854947,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999833674196865,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.15988747626543046,
|
|
"num_tokens": 3216262.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.274352216720581,
|
|
"epoch": 0.13496637848222862,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999829479564518,
|
|
"loss": 5.9881,
|
|
"mean_token_accuracy": 0.15093525648117065,
|
|
"num_tokens": 3228035.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.099394607543945,
|
|
"epoch": 0.13544668587896252,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499982523269781,
|
|
"loss": 5.8569,
|
|
"mean_token_accuracy": 0.1615759938955307,
|
|
"num_tokens": 3239643.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.1452710151672365,
|
|
"epoch": 0.13592699327569643,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999820933596842,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.16112854480743408,
|
|
"num_tokens": 3250293.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.273663139343261,
|
|
"epoch": 0.13640730067243034,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499981658226171,
|
|
"loss": 6.0604,
|
|
"mean_token_accuracy": 0.1478336751461029,
|
|
"num_tokens": 3262448.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.064791679382324,
|
|
"epoch": 0.13688760806916425,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000499981217869252,
|
|
"loss": 5.9375,
|
|
"mean_token_accuracy": 0.1563238486647606,
|
|
"num_tokens": 3274041.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.223370218276978,
|
|
"epoch": 0.1373679154658982,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499980772288937,
|
|
"loss": 5.9609,
|
|
"mean_token_accuracy": 0.1560527950525284,
|
|
"num_tokens": 3285063.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.2211846828460695,
|
|
"epoch": 0.1378482228626321,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999803214852367,
|
|
"loss": 5.9666,
|
|
"mean_token_accuracy": 0.15824110060930252,
|
|
"num_tokens": 3297031.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.197072458267212,
|
|
"epoch": 0.138328530259366,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999798654581613,
|
|
"loss": 5.9573,
|
|
"mean_token_accuracy": 0.1547205552458763,
|
|
"num_tokens": 3308187.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.1259284019470215,
|
|
"epoch": 0.13880883765609991,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999794042077214,
|
|
"loss": 5.9549,
|
|
"mean_token_accuracy": 0.15546363741159439,
|
|
"num_tokens": 3319439.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.13748083114624,
|
|
"epoch": 0.13928914505283382,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999789377339279,
|
|
"loss": 5.9476,
|
|
"mean_token_accuracy": 0.1569317899644375,
|
|
"num_tokens": 3331715.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.180286836624146,
|
|
"epoch": 0.13976945244956773,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999784660367915,
|
|
"loss": 5.9111,
|
|
"mean_token_accuracy": 0.1668043091893196,
|
|
"num_tokens": 3342281.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.317266130447388,
|
|
"epoch": 0.14024975984630164,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999779891163231,
|
|
"loss": 6.0415,
|
|
"mean_token_accuracy": 0.15571026653051376,
|
|
"num_tokens": 3353530.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.173709964752197,
|
|
"epoch": 0.14073006724303555,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999775069725339,
|
|
"loss": 6.0136,
|
|
"mean_token_accuracy": 0.1503211587667465,
|
|
"num_tokens": 3365083.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.193066835403442,
|
|
"epoch": 0.14121037463976946,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000499977019605435,
|
|
"loss": 5.9567,
|
|
"mean_token_accuracy": 0.1586127519607544,
|
|
"num_tokens": 3378143.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.231985139846802,
|
|
"epoch": 0.14169068203650337,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999765270150378,
|
|
"loss": 5.9434,
|
|
"mean_token_accuracy": 0.15540362745523453,
|
|
"num_tokens": 3388438.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.073607492446899,
|
|
"epoch": 0.14217098943323728,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999760292013536,
|
|
"loss": 5.9317,
|
|
"mean_token_accuracy": 0.15305885225534438,
|
|
"num_tokens": 3400150.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.05352897644043,
|
|
"epoch": 0.14265129682997119,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999755261643941,
|
|
"loss": 5.8103,
|
|
"mean_token_accuracy": 0.16507214605808257,
|
|
"num_tokens": 3411073.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.134577798843384,
|
|
"epoch": 0.1431316042267051,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999750179041709,
|
|
"loss": 5.8741,
|
|
"mean_token_accuracy": 0.16767435967922212,
|
|
"num_tokens": 3422196.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.159465312957764,
|
|
"epoch": 0.143611911623439,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999745044206959,
|
|
"loss": 5.9009,
|
|
"mean_token_accuracy": 0.1562460742890835,
|
|
"num_tokens": 3432568.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.127354431152344,
|
|
"epoch": 0.1440922190201729,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999739857139809,
|
|
"loss": 5.9367,
|
|
"mean_token_accuracy": 0.15517643094062805,
|
|
"num_tokens": 3444096.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.010924053192139,
|
|
"epoch": 0.14457252641690682,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499973461784038,
|
|
"loss": 5.8178,
|
|
"mean_token_accuracy": 0.1662687450647354,
|
|
"num_tokens": 3455502.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.194846773147583,
|
|
"epoch": 0.14505283381364073,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999729326308792,
|
|
"loss": 5.9572,
|
|
"mean_token_accuracy": 0.15202689915895462,
|
|
"num_tokens": 3467753.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.0664191246032715,
|
|
"epoch": 0.14553314121037464,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499972398254517,
|
|
"loss": 5.8367,
|
|
"mean_token_accuracy": 0.16286454498767852,
|
|
"num_tokens": 3479206.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.076856899261474,
|
|
"epoch": 0.14601344860710855,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499971858654964,
|
|
"loss": 5.8593,
|
|
"mean_token_accuracy": 0.16005290448665618,
|
|
"num_tokens": 3490914.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.107672548294067,
|
|
"epoch": 0.14649375600384246,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999713138322321,
|
|
"loss": 5.9129,
|
|
"mean_token_accuracy": 0.15861159712076187,
|
|
"num_tokens": 3502932.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.048060894012451,
|
|
"epoch": 0.14697406340057637,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999707637863346,
|
|
"loss": 5.8432,
|
|
"mean_token_accuracy": 0.16491712927818297,
|
|
"num_tokens": 3515071.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.208171081542969,
|
|
"epoch": 0.14745437079731027,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999702085172838,
|
|
"loss": 5.956,
|
|
"mean_token_accuracy": 0.15694793835282325,
|
|
"num_tokens": 3526845.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.080694913864136,
|
|
"epoch": 0.14793467819404418,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999696480250929,
|
|
"loss": 5.9302,
|
|
"mean_token_accuracy": 0.15591877698898315,
|
|
"num_tokens": 3538810.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.19296703338623,
|
|
"epoch": 0.1484149855907781,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999690823097747,
|
|
"loss": 5.8928,
|
|
"mean_token_accuracy": 0.16947668939828872,
|
|
"num_tokens": 3549570.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.932188987731934,
|
|
"epoch": 0.148895292987512,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999685113713426,
|
|
"loss": 5.8191,
|
|
"mean_token_accuracy": 0.17169633954763414,
|
|
"num_tokens": 3560937.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.1333564758300785,
|
|
"epoch": 0.1493756003842459,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999679352098096,
|
|
"loss": 5.8803,
|
|
"mean_token_accuracy": 0.15864527374505996,
|
|
"num_tokens": 3573603.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.011618709564209,
|
|
"epoch": 0.14985590778097982,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999673538251891,
|
|
"loss": 5.8981,
|
|
"mean_token_accuracy": 0.15385786592960357,
|
|
"num_tokens": 3586034.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.00404839515686,
|
|
"epoch": 0.15033621517771373,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999667672174947,
|
|
"loss": 5.8189,
|
|
"mean_token_accuracy": 0.16424645334482194,
|
|
"num_tokens": 3597305.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.191066837310791,
|
|
"epoch": 0.15081652257444764,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00049996617538674,
|
|
"loss": 5.9702,
|
|
"mean_token_accuracy": 0.1484566166996956,
|
|
"num_tokens": 3608758.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.153885412216186,
|
|
"epoch": 0.15129682997118155,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999655783329386,
|
|
"loss": 5.9672,
|
|
"mean_token_accuracy": 0.15386258363723754,
|
|
"num_tokens": 3621508.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.169773721694947,
|
|
"epoch": 0.15177713736791545,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999649760561046,
|
|
"loss": 5.9308,
|
|
"mean_token_accuracy": 0.16340596377849578,
|
|
"num_tokens": 3634826.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.981789493560791,
|
|
"epoch": 0.15225744476464936,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999643685562519,
|
|
"loss": 5.7666,
|
|
"mean_token_accuracy": 0.1757243752479553,
|
|
"num_tokens": 3646427.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.004927635192871,
|
|
"epoch": 0.15273775216138327,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999637558333945,
|
|
"loss": 5.8887,
|
|
"mean_token_accuracy": 0.15893372148275375,
|
|
"num_tokens": 3658157.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.1353168964385985,
|
|
"epoch": 0.15321805955811718,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999631378875467,
|
|
"loss": 5.9238,
|
|
"mean_token_accuracy": 0.1580752119421959,
|
|
"num_tokens": 3670695.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.068402624130249,
|
|
"epoch": 0.15369836695485112,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999625147187228,
|
|
"loss": 5.849,
|
|
"mean_token_accuracy": 0.1568877637386322,
|
|
"num_tokens": 3683379.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.990541267395019,
|
|
"epoch": 0.15417867435158503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999618863269373,
|
|
"loss": 5.7307,
|
|
"mean_token_accuracy": 0.1711265340447426,
|
|
"num_tokens": 3694411.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.020878696441651,
|
|
"epoch": 0.15465898174831894,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999612527122049,
|
|
"loss": 5.7903,
|
|
"mean_token_accuracy": 0.17313530445098876,
|
|
"num_tokens": 3703833.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.092939853668213,
|
|
"epoch": 0.15513928914505284,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999606138745402,
|
|
"loss": 5.8829,
|
|
"mean_token_accuracy": 0.16913823708891867,
|
|
"num_tokens": 3714702.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.116333436965943,
|
|
"epoch": 0.15561959654178675,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999599698139581,
|
|
"loss": 5.9569,
|
|
"mean_token_accuracy": 0.1631587713956833,
|
|
"num_tokens": 3726000.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.10855507850647,
|
|
"epoch": 0.15609990393852066,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999593205304734,
|
|
"loss": 5.8364,
|
|
"mean_token_accuracy": 0.16023391783237456,
|
|
"num_tokens": 3737681.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.14500937461853,
|
|
"epoch": 0.15658021133525457,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999586660241012,
|
|
"loss": 5.9447,
|
|
"mean_token_accuracy": 0.160616697371006,
|
|
"num_tokens": 3749477.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.12149567604065,
|
|
"epoch": 0.15706051873198848,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999580062948569,
|
|
"loss": 5.9172,
|
|
"mean_token_accuracy": 0.1614253431558609,
|
|
"num_tokens": 3759946.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.0639667987823485,
|
|
"epoch": 0.1575408261287224,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999573413427556,
|
|
"loss": 5.8266,
|
|
"mean_token_accuracy": 0.16388048231601715,
|
|
"num_tokens": 3770147.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.004903888702392,
|
|
"epoch": 0.1580211335254563,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999566711678128,
|
|
"loss": 5.767,
|
|
"mean_token_accuracy": 0.1723175212740898,
|
|
"num_tokens": 3781028.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.069085884094238,
|
|
"epoch": 0.1585014409221902,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999559957700442,
|
|
"loss": 5.8428,
|
|
"mean_token_accuracy": 0.15636067241430282,
|
|
"num_tokens": 3792529.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.152743339538574,
|
|
"epoch": 0.15898174831892412,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999553151494653,
|
|
"loss": 5.9606,
|
|
"mean_token_accuracy": 0.1637915402650833,
|
|
"num_tokens": 3804630.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.922781467437744,
|
|
"epoch": 0.15946205571565802,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999546293060919,
|
|
"loss": 5.7158,
|
|
"mean_token_accuracy": 0.17320418655872344,
|
|
"num_tokens": 3815019.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.049646425247192,
|
|
"epoch": 0.15994236311239193,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00049995393823994,
|
|
"loss": 5.8234,
|
|
"mean_token_accuracy": 0.1636585474014282,
|
|
"num_tokens": 3825648.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.026040172576904,
|
|
"epoch": 0.16042267050912584,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999532419510255,
|
|
"loss": 5.8164,
|
|
"mean_token_accuracy": 0.16479419767856598,
|
|
"num_tokens": 3836779.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.989983701705933,
|
|
"epoch": 0.16090297790585975,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000499952540439365,
|
|
"loss": 5.8646,
|
|
"mean_token_accuracy": 0.15931192189455032,
|
|
"num_tokens": 3848350.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.11183648109436,
|
|
"epoch": 0.16138328530259366,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999518337049743,
|
|
"loss": 5.8669,
|
|
"mean_token_accuracy": 0.1647849351167679,
|
|
"num_tokens": 3859792.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.995904159545899,
|
|
"epoch": 0.16186359269932757,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00049995112174787,
|
|
"loss": 5.7954,
|
|
"mean_token_accuracy": 0.16283721774816512,
|
|
"num_tokens": 3871539.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.97924485206604,
|
|
"epoch": 0.16234390009606148,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999504045680687,
|
|
"loss": 5.8367,
|
|
"mean_token_accuracy": 0.16463494151830674,
|
|
"num_tokens": 3885025.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.080967426300049,
|
|
"epoch": 0.1628242074927954,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999496821655869,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.15857994854450225,
|
|
"num_tokens": 3896519.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.997112226486206,
|
|
"epoch": 0.1633045148895293,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999489545404414,
|
|
"loss": 5.7113,
|
|
"mean_token_accuracy": 0.17066951245069503,
|
|
"num_tokens": 3907636.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.9870668888092045,
|
|
"epoch": 0.1637848222862632,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999482216926493,
|
|
"loss": 5.8384,
|
|
"mean_token_accuracy": 0.163779553771019,
|
|
"num_tokens": 3917746.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.008543872833252,
|
|
"epoch": 0.1642651296829971,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999474836222273,
|
|
"loss": 5.7528,
|
|
"mean_token_accuracy": 0.17229880392551422,
|
|
"num_tokens": 3928901.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.945359992980957,
|
|
"epoch": 0.16474543707973102,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999467403291928,
|
|
"loss": 5.8697,
|
|
"mean_token_accuracy": 0.16977249681949616,
|
|
"num_tokens": 3940094.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.0175048351287845,
|
|
"epoch": 0.16522574447646493,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999459918135628,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.16932837516069413,
|
|
"num_tokens": 3952714.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.957449197769165,
|
|
"epoch": 0.16570605187319884,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499945238075355,
|
|
"loss": 5.7765,
|
|
"mean_token_accuracy": 0.17156666070222854,
|
|
"num_tokens": 3963107.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.997269058227539,
|
|
"epoch": 0.16618635926993275,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999444791145865,
|
|
"loss": 5.7201,
|
|
"mean_token_accuracy": 0.17136422395706177,
|
|
"num_tokens": 3973454.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.957687091827393,
|
|
"epoch": 0.16666666666666666,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004999437149312754,
|
|
"loss": 5.8376,
|
|
"mean_token_accuracy": 0.16045114547014236,
|
|
"num_tokens": 3985352.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.06954288482666,
|
|
"epoch": 0.16714697406340057,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499942945525439,
|
|
"loss": 5.8073,
|
|
"mean_token_accuracy": 0.16711310297250748,
|
|
"num_tokens": 3997088.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.022908163070679,
|
|
"epoch": 0.16762728146013448,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999421708970954,
|
|
"loss": 5.8296,
|
|
"mean_token_accuracy": 0.1672845110297203,
|
|
"num_tokens": 4008551.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.984336090087891,
|
|
"epoch": 0.16810758885686838,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999413910462625,
|
|
"loss": 5.8538,
|
|
"mean_token_accuracy": 0.16413284242153167,
|
|
"num_tokens": 4020886.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.037380981445312,
|
|
"epoch": 0.1685878962536023,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999406059729586,
|
|
"loss": 5.7928,
|
|
"mean_token_accuracy": 0.1607511505484581,
|
|
"num_tokens": 4032811.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.000102710723877,
|
|
"epoch": 0.1690682036503362,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999398156772016,
|
|
"loss": 5.7873,
|
|
"mean_token_accuracy": 0.16719766408205033,
|
|
"num_tokens": 4044102.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.994718551635742,
|
|
"epoch": 0.16954851104707014,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00049993902015901,
|
|
"loss": 5.8172,
|
|
"mean_token_accuracy": 0.1658351629972458,
|
|
"num_tokens": 4056849.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.982969522476196,
|
|
"epoch": 0.17002881844380405,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999382194184023,
|
|
"loss": 5.8477,
|
|
"mean_token_accuracy": 0.1637706860899925,
|
|
"num_tokens": 4068392.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.006184864044189,
|
|
"epoch": 0.17050912584053796,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999374134553972,
|
|
"loss": 5.6927,
|
|
"mean_token_accuracy": 0.170590578019619,
|
|
"num_tokens": 4078949.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.909857225418091,
|
|
"epoch": 0.17098943323727187,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999366022700131,
|
|
"loss": 5.8275,
|
|
"mean_token_accuracy": 0.16255858093500136,
|
|
"num_tokens": 4091432.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.954454946517944,
|
|
"epoch": 0.17146974063400577,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999357858622691,
|
|
"loss": 5.6951,
|
|
"mean_token_accuracy": 0.17012006640434266,
|
|
"num_tokens": 4103001.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.986909008026123,
|
|
"epoch": 0.17195004803073968,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999349642321842,
|
|
"loss": 5.8051,
|
|
"mean_token_accuracy": 0.16849690526723862,
|
|
"num_tokens": 4113570.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.007826423645019,
|
|
"epoch": 0.1724303554274736,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999341373797772,
|
|
"loss": 5.8411,
|
|
"mean_token_accuracy": 0.16274026334285735,
|
|
"num_tokens": 4125658.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.95023546218872,
|
|
"epoch": 0.1729106628242075,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999333053050675,
|
|
"loss": 5.7205,
|
|
"mean_token_accuracy": 0.16709082424640656,
|
|
"num_tokens": 4137424.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.011094236373902,
|
|
"epoch": 0.1733909702209414,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999324680080744,
|
|
"loss": 5.764,
|
|
"mean_token_accuracy": 0.16791113317012787,
|
|
"num_tokens": 4149772.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.905970621109009,
|
|
"epoch": 0.17387127761767532,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999316254888172,
|
|
"loss": 5.7669,
|
|
"mean_token_accuracy": 0.1733687311410904,
|
|
"num_tokens": 4162246.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.918607234954834,
|
|
"epoch": 0.17435158501440923,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999307777473157,
|
|
"loss": 5.7396,
|
|
"mean_token_accuracy": 0.17318637818098068,
|
|
"num_tokens": 4173449.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.937366533279419,
|
|
"epoch": 0.17483189241114314,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999299247835893,
|
|
"loss": 5.7745,
|
|
"mean_token_accuracy": 0.1625620037317276,
|
|
"num_tokens": 4185560.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.952977085113526,
|
|
"epoch": 0.17531219980787704,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000499929066597658,
|
|
"loss": 5.8399,
|
|
"mean_token_accuracy": 0.15584002658724785,
|
|
"num_tokens": 4197843.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.022454023361206,
|
|
"epoch": 0.17579250720461095,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999282031895418,
|
|
"loss": 5.6884,
|
|
"mean_token_accuracy": 0.16660276055335999,
|
|
"num_tokens": 4210009.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.942712497711182,
|
|
"epoch": 0.17627281460134486,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999273345592604,
|
|
"loss": 5.932,
|
|
"mean_token_accuracy": 0.15909999683499337,
|
|
"num_tokens": 4222213.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.004428100585938,
|
|
"epoch": 0.17675312199807877,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999264607068343,
|
|
"loss": 5.778,
|
|
"mean_token_accuracy": 0.16446920484304428,
|
|
"num_tokens": 4234159.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.9961381435394285,
|
|
"epoch": 0.17723342939481268,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999255816322837,
|
|
"loss": 5.8588,
|
|
"mean_token_accuracy": 0.15720370337367057,
|
|
"num_tokens": 4246358.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.0960883617401125,
|
|
"epoch": 0.1777137367915466,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499924697335629,
|
|
"loss": 5.9093,
|
|
"mean_token_accuracy": 0.1618606984615326,
|
|
"num_tokens": 4256698.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.9101903438568115,
|
|
"epoch": 0.1781940441882805,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999238078168906,
|
|
"loss": 5.7071,
|
|
"mean_token_accuracy": 0.17886844426393508,
|
|
"num_tokens": 4268007.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.921268177032471,
|
|
"epoch": 0.1786743515850144,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999229130760894,
|
|
"loss": 5.8127,
|
|
"mean_token_accuracy": 0.1664751335978508,
|
|
"num_tokens": 4278583.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.077760076522827,
|
|
"epoch": 0.17915465898174832,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499922013113246,
|
|
"loss": 5.8361,
|
|
"mean_token_accuracy": 0.1648312047123909,
|
|
"num_tokens": 4290267.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.952408123016357,
|
|
"epoch": 0.17963496637848222,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999211079283814,
|
|
"loss": 5.732,
|
|
"mean_token_accuracy": 0.17537832111120225,
|
|
"num_tokens": 4302177.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.861745071411133,
|
|
"epoch": 0.18011527377521613,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999201975215164,
|
|
"loss": 5.6961,
|
|
"mean_token_accuracy": 0.17122533768415452,
|
|
"num_tokens": 4312420.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.0441864967346195,
|
|
"epoch": 0.18059558117195004,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999192818926725,
|
|
"loss": 5.8215,
|
|
"mean_token_accuracy": 0.16632328778505326,
|
|
"num_tokens": 4324050.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.901515483856201,
|
|
"epoch": 0.18107588856868395,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999183610418706,
|
|
"loss": 5.6771,
|
|
"mean_token_accuracy": 0.17299336791038514,
|
|
"num_tokens": 4335336.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.852649927139282,
|
|
"epoch": 0.18155619596541786,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999174349691322,
|
|
"loss": 5.6976,
|
|
"mean_token_accuracy": 0.17085058093070984,
|
|
"num_tokens": 4346239.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.94590048789978,
|
|
"epoch": 0.18203650336215177,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999165036744788,
|
|
"loss": 5.7683,
|
|
"mean_token_accuracy": 0.1664916157722473,
|
|
"num_tokens": 4357195.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.875322675704956,
|
|
"epoch": 0.18251681075888568,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999155671579322,
|
|
"loss": 5.702,
|
|
"mean_token_accuracy": 0.17268287986516953,
|
|
"num_tokens": 4368905.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.9875816822052,
|
|
"epoch": 0.1829971181556196,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000499914625419514,
|
|
"loss": 5.7837,
|
|
"mean_token_accuracy": 0.16864405423402787,
|
|
"num_tokens": 4381708.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.846536254882812,
|
|
"epoch": 0.1834774255523535,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999136784592459,
|
|
"loss": 5.6212,
|
|
"mean_token_accuracy": 0.18302999585866928,
|
|
"num_tokens": 4393199.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.953036260604859,
|
|
"epoch": 0.1839577329490874,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999127262771502,
|
|
"loss": 5.8543,
|
|
"mean_token_accuracy": 0.1608387678861618,
|
|
"num_tokens": 4403986.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.000203561782837,
|
|
"epoch": 0.1844380403458213,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999117688732487,
|
|
"loss": 5.7057,
|
|
"mean_token_accuracy": 0.1680416136980057,
|
|
"num_tokens": 4415574.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.88345947265625,
|
|
"epoch": 0.18491834774255522,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999108062475638,
|
|
"loss": 5.7319,
|
|
"mean_token_accuracy": 0.1759544551372528,
|
|
"num_tokens": 4427579.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.957119131088257,
|
|
"epoch": 0.18539865513928913,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499909838400118,
|
|
"loss": 5.7284,
|
|
"mean_token_accuracy": 0.16889535933732985,
|
|
"num_tokens": 4440045.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.919621276855469,
|
|
"epoch": 0.18587896253602307,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999088653309334,
|
|
"loss": 5.8139,
|
|
"mean_token_accuracy": 0.16399643570184708,
|
|
"num_tokens": 4450129.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.897604322433471,
|
|
"epoch": 0.18635926993275698,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999078870400329,
|
|
"loss": 5.6809,
|
|
"mean_token_accuracy": 0.17348277419805527,
|
|
"num_tokens": 4461378.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.938304138183594,
|
|
"epoch": 0.18683957732949089,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999069035274391,
|
|
"loss": 5.7048,
|
|
"mean_token_accuracy": 0.17170394212007523,
|
|
"num_tokens": 4472249.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.890934944152832,
|
|
"epoch": 0.1873198847262248,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999059147931747,
|
|
"loss": 5.7677,
|
|
"mean_token_accuracy": 0.1649644747376442,
|
|
"num_tokens": 4484243.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.9530542373657225,
|
|
"epoch": 0.1878001921229587,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999049208372629,
|
|
"loss": 5.6785,
|
|
"mean_token_accuracy": 0.16919784843921662,
|
|
"num_tokens": 4495776.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.853722429275512,
|
|
"epoch": 0.1882804995196926,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999039216597267,
|
|
"loss": 5.7548,
|
|
"mean_token_accuracy": 0.16727603524923323,
|
|
"num_tokens": 4508581.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.9266825199127195,
|
|
"epoch": 0.18876080691642652,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999029172605892,
|
|
"loss": 5.6903,
|
|
"mean_token_accuracy": 0.16845113039016724,
|
|
"num_tokens": 4519753.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.900664520263672,
|
|
"epoch": 0.18924111431316043,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999019076398738,
|
|
"loss": 5.7549,
|
|
"mean_token_accuracy": 0.1777681216597557,
|
|
"num_tokens": 4529957.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.974092197418213,
|
|
"epoch": 0.18972142170989434,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499900892797604,
|
|
"loss": 5.7312,
|
|
"mean_token_accuracy": 0.16701797097921373,
|
|
"num_tokens": 4541181.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.877654409408569,
|
|
"epoch": 0.19020172910662825,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998998727338031,
|
|
"loss": 5.6976,
|
|
"mean_token_accuracy": 0.17389168292284013,
|
|
"num_tokens": 4552235.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.943179893493652,
|
|
"epoch": 0.19068203650336216,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998988474484952,
|
|
"loss": 5.7062,
|
|
"mean_token_accuracy": 0.1742224797606468,
|
|
"num_tokens": 4564045.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.8289329528808596,
|
|
"epoch": 0.19116234390009607,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998978169417038,
|
|
"loss": 5.6085,
|
|
"mean_token_accuracy": 0.1785324141383171,
|
|
"num_tokens": 4575141.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.852221345901489,
|
|
"epoch": 0.19164265129682997,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998967812134529,
|
|
"loss": 5.723,
|
|
"mean_token_accuracy": 0.17090109288692473,
|
|
"num_tokens": 4586491.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.9237569808959964,
|
|
"epoch": 0.19212295869356388,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998957402637664,
|
|
"loss": 5.7187,
|
|
"mean_token_accuracy": 0.1718616396188736,
|
|
"num_tokens": 4598421.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.894081020355225,
|
|
"epoch": 0.1926032660902978,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998946940926687,
|
|
"loss": 5.6331,
|
|
"mean_token_accuracy": 0.18285724371671677,
|
|
"num_tokens": 4609688.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.882657384872436,
|
|
"epoch": 0.1930835734870317,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499893642700184,
|
|
"loss": 5.7813,
|
|
"mean_token_accuracy": 0.16469601690769195,
|
|
"num_tokens": 4621169.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.9078254222869875,
|
|
"epoch": 0.1935638808837656,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998925860863368,
|
|
"loss": 5.7242,
|
|
"mean_token_accuracy": 0.17368836998939513,
|
|
"num_tokens": 4632512.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.930011129379272,
|
|
"epoch": 0.19404418828049952,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998915242511516,
|
|
"loss": 5.7794,
|
|
"mean_token_accuracy": 0.16718964874744416,
|
|
"num_tokens": 4644356.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.975589370727539,
|
|
"epoch": 0.19452449567723343,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998904571946528,
|
|
"loss": 5.7641,
|
|
"mean_token_accuracy": 0.1692662999033928,
|
|
"num_tokens": 4655601.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.867001152038574,
|
|
"epoch": 0.19500480307396734,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998893849168655,
|
|
"loss": 5.7754,
|
|
"mean_token_accuracy": 0.16788003146648406,
|
|
"num_tokens": 4666813.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.944839382171631,
|
|
"epoch": 0.19548511047070125,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998883074178144,
|
|
"loss": 5.7572,
|
|
"mean_token_accuracy": 0.16712078899145127,
|
|
"num_tokens": 4679123.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.912882137298584,
|
|
"epoch": 0.19596541786743515,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998872246975247,
|
|
"loss": 5.6571,
|
|
"mean_token_accuracy": 0.17612844854593276,
|
|
"num_tokens": 4690485.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.806442070007324,
|
|
"epoch": 0.19644572526416906,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998861367560213,
|
|
"loss": 5.6748,
|
|
"mean_token_accuracy": 0.1725752279162407,
|
|
"num_tokens": 4702573.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.891991090774536,
|
|
"epoch": 0.19692603266090297,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998850435933296,
|
|
"loss": 5.7193,
|
|
"mean_token_accuracy": 0.16615684181451798,
|
|
"num_tokens": 4715653.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.882970237731934,
|
|
"epoch": 0.19740634005763688,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998839452094749,
|
|
"loss": 5.7004,
|
|
"mean_token_accuracy": 0.17499724328517913,
|
|
"num_tokens": 4726051.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.934564733505249,
|
|
"epoch": 0.1978866474543708,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998828416044829,
|
|
"loss": 5.8515,
|
|
"mean_token_accuracy": 0.16007425934076308,
|
|
"num_tokens": 4738176.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.954264545440674,
|
|
"epoch": 0.1983669548511047,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499881732778379,
|
|
"loss": 5.7522,
|
|
"mean_token_accuracy": 0.16757950633764268,
|
|
"num_tokens": 4749547.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.838888549804688,
|
|
"epoch": 0.1988472622478386,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499880618731189,
|
|
"loss": 5.6566,
|
|
"mean_token_accuracy": 0.1716952934861183,
|
|
"num_tokens": 4761967.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.834277725219726,
|
|
"epoch": 0.19932756964457252,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998794994629388,
|
|
"loss": 5.6485,
|
|
"mean_token_accuracy": 0.1739967070519924,
|
|
"num_tokens": 4772847.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.932867527008057,
|
|
"epoch": 0.19980787704130643,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998783749736545,
|
|
"loss": 5.7124,
|
|
"mean_token_accuracy": 0.17146496921777726,
|
|
"num_tokens": 4785049.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.833413457870483,
|
|
"epoch": 0.20028818443804033,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998772452633619,
|
|
"loss": 5.7374,
|
|
"mean_token_accuracy": 0.16502633690834045,
|
|
"num_tokens": 4797484.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.859524917602539,
|
|
"epoch": 0.20076849183477424,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998761103320876,
|
|
"loss": 5.6756,
|
|
"mean_token_accuracy": 0.1668789342045784,
|
|
"num_tokens": 4809422.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.92907977104187,
|
|
"epoch": 0.20124879923150815,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998749701798577,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.16625447124242781,
|
|
"num_tokens": 4822340.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.860308408737183,
|
|
"epoch": 0.2017291066282421,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998738248066986,
|
|
"loss": 5.6352,
|
|
"mean_token_accuracy": 0.17530228346586227,
|
|
"num_tokens": 4833382.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.790621709823609,
|
|
"epoch": 0.202209414024976,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998726742126372,
|
|
"loss": 5.6433,
|
|
"mean_token_accuracy": 0.1817137658596039,
|
|
"num_tokens": 4843916.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.927386856079101,
|
|
"epoch": 0.2026897214217099,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998715183976999,
|
|
"loss": 5.6436,
|
|
"mean_token_accuracy": 0.17769275158643721,
|
|
"num_tokens": 4855420.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.880288362503052,
|
|
"epoch": 0.20317002881844382,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998703573619137,
|
|
"loss": 5.7103,
|
|
"mean_token_accuracy": 0.1706031545996666,
|
|
"num_tokens": 4866540.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.895658493041992,
|
|
"epoch": 0.20365033621517772,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004998691911053056,
|
|
"loss": 5.6665,
|
|
"mean_token_accuracy": 0.17256468534469604,
|
|
"num_tokens": 4878860.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.960579252243042,
|
|
"epoch": 0.20413064361191163,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998680196279026,
|
|
"loss": 5.7883,
|
|
"mean_token_accuracy": 0.16489656418561935,
|
|
"num_tokens": 4889692.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.970758581161499,
|
|
"epoch": 0.20461095100864554,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998668429297319,
|
|
"loss": 5.7306,
|
|
"mean_token_accuracy": 0.17567683905363082,
|
|
"num_tokens": 4901337.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.75920524597168,
|
|
"epoch": 0.20509125840537945,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998656610108208,
|
|
"loss": 5.6448,
|
|
"mean_token_accuracy": 0.17524836361408233,
|
|
"num_tokens": 4912340.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.857781076431275,
|
|
"epoch": 0.20557156580211336,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998644738711969,
|
|
"loss": 5.6107,
|
|
"mean_token_accuracy": 0.17784797996282578,
|
|
"num_tokens": 4923429.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.896333789825439,
|
|
"epoch": 0.20605187319884727,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998632815108874,
|
|
"loss": 5.6966,
|
|
"mean_token_accuracy": 0.16947495341300964,
|
|
"num_tokens": 4934826.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.881085062026978,
|
|
"epoch": 0.20653218059558118,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998620839299203,
|
|
"loss": 5.7698,
|
|
"mean_token_accuracy": 0.17018158435821534,
|
|
"num_tokens": 4945189.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.932586526870727,
|
|
"epoch": 0.2070124879923151,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998608811283233,
|
|
"loss": 5.6551,
|
|
"mean_token_accuracy": 0.1765411153435707,
|
|
"num_tokens": 4956060.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.858534717559815,
|
|
"epoch": 0.207492795389049,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998596731061244,
|
|
"loss": 5.6717,
|
|
"mean_token_accuracy": 0.17596034407615663,
|
|
"num_tokens": 4967710.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.857064723968506,
|
|
"epoch": 0.2079731027857829,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998584598633516,
|
|
"loss": 5.6789,
|
|
"mean_token_accuracy": 0.16734581142663957,
|
|
"num_tokens": 4978716.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.874509191513061,
|
|
"epoch": 0.2084534101825168,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998572414000329,
|
|
"loss": 5.7568,
|
|
"mean_token_accuracy": 0.17040223330259324,
|
|
"num_tokens": 4990742.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.943036460876465,
|
|
"epoch": 0.20893371757925072,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998560177161969,
|
|
"loss": 5.7077,
|
|
"mean_token_accuracy": 0.17488541454076767,
|
|
"num_tokens": 5001217.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.846171760559082,
|
|
"epoch": 0.20941402497598463,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998547888118718,
|
|
"loss": 5.6921,
|
|
"mean_token_accuracy": 0.16881862580776213,
|
|
"num_tokens": 5012750.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.865509939193726,
|
|
"epoch": 0.20989433237271854,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998535546870862,
|
|
"loss": 5.7001,
|
|
"mean_token_accuracy": 0.17801522612571716,
|
|
"num_tokens": 5023818.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.9269102096557615,
|
|
"epoch": 0.21037463976945245,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998523153418687,
|
|
"loss": 5.7176,
|
|
"mean_token_accuracy": 0.16671841889619826,
|
|
"num_tokens": 5035060.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.882676506042481,
|
|
"epoch": 0.21085494716618636,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998510707762481,
|
|
"loss": 5.7578,
|
|
"mean_token_accuracy": 0.16046614050865174,
|
|
"num_tokens": 5046780.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.83563551902771,
|
|
"epoch": 0.21133525456292027,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998498209902533,
|
|
"loss": 5.725,
|
|
"mean_token_accuracy": 0.16501837968826294,
|
|
"num_tokens": 5059152.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.86581654548645,
|
|
"epoch": 0.21181556195965417,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998485659839134,
|
|
"loss": 5.6854,
|
|
"mean_token_accuracy": 0.17040270417928696,
|
|
"num_tokens": 5069971.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.828520584106445,
|
|
"epoch": 0.21229586935638808,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998473057572575,
|
|
"loss": 5.6356,
|
|
"mean_token_accuracy": 0.17748477756977082,
|
|
"num_tokens": 5080640.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.851872873306275,
|
|
"epoch": 0.212776176753122,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998460403103146,
|
|
"loss": 5.6556,
|
|
"mean_token_accuracy": 0.17382104694843292,
|
|
"num_tokens": 5091240.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.754461812973022,
|
|
"epoch": 0.2132564841498559,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998447696431146,
|
|
"loss": 5.6739,
|
|
"mean_token_accuracy": 0.17026726752519608,
|
|
"num_tokens": 5102920.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.863530015945434,
|
|
"epoch": 0.2137367915465898,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998434937556865,
|
|
"loss": 5.7022,
|
|
"mean_token_accuracy": 0.17059502005577087,
|
|
"num_tokens": 5114240.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.968141746520996,
|
|
"epoch": 0.21421709894332372,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998422126480602,
|
|
"loss": 5.9208,
|
|
"mean_token_accuracy": 0.15989586859941482,
|
|
"num_tokens": 5126653.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.998019552230835,
|
|
"epoch": 0.21469740634005763,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998409263202653,
|
|
"loss": 5.7689,
|
|
"mean_token_accuracy": 0.1644024819135666,
|
|
"num_tokens": 5138778.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.859699535369873,
|
|
"epoch": 0.21517771373679154,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998396347723318,
|
|
"loss": 5.6178,
|
|
"mean_token_accuracy": 0.17633095979690552,
|
|
"num_tokens": 5148670.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.68029522895813,
|
|
"epoch": 0.21565802113352545,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998383380042895,
|
|
"loss": 5.6738,
|
|
"mean_token_accuracy": 0.16909311413764955,
|
|
"num_tokens": 5159617.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.894393634796143,
|
|
"epoch": 0.21613832853025935,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998370360161688,
|
|
"loss": 5.7241,
|
|
"mean_token_accuracy": 0.1727438136935234,
|
|
"num_tokens": 5171225.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.87973313331604,
|
|
"epoch": 0.21661863592699326,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998357288079996,
|
|
"loss": 5.7253,
|
|
"mean_token_accuracy": 0.1702487885951996,
|
|
"num_tokens": 5184586.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.842156028747558,
|
|
"epoch": 0.21709894332372717,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998344163798125,
|
|
"loss": 5.6896,
|
|
"mean_token_accuracy": 0.1675831764936447,
|
|
"num_tokens": 5196089.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.868828344345093,
|
|
"epoch": 0.21757925072046108,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998330987316379,
|
|
"loss": 5.7183,
|
|
"mean_token_accuracy": 0.17021534889936446,
|
|
"num_tokens": 5207581.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.812510681152344,
|
|
"epoch": 0.21805955811719502,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998317758635062,
|
|
"loss": 5.672,
|
|
"mean_token_accuracy": 0.1744819015264511,
|
|
"num_tokens": 5219108.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.866617918014526,
|
|
"epoch": 0.21853986551392893,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004998304477754484,
|
|
"loss": 5.6265,
|
|
"mean_token_accuracy": 0.17249889373779298,
|
|
"num_tokens": 5231137.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.9222582340240475,
|
|
"epoch": 0.21902017291066284,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998291144674952,
|
|
"loss": 5.692,
|
|
"mean_token_accuracy": 0.17626527845859527,
|
|
"num_tokens": 5243528.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.688040399551392,
|
|
"epoch": 0.21950048030739674,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998277759396776,
|
|
"loss": 5.6408,
|
|
"mean_token_accuracy": 0.17519633769989013,
|
|
"num_tokens": 5254869.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.8270598411560055,
|
|
"epoch": 0.21998078770413065,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998264321920265,
|
|
"loss": 5.6233,
|
|
"mean_token_accuracy": 0.17478814274072646,
|
|
"num_tokens": 5265370.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.782625007629394,
|
|
"epoch": 0.22046109510086456,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004998250832245734,
|
|
"loss": 5.5868,
|
|
"mean_token_accuracy": 0.17365225702524184,
|
|
"num_tokens": 5277812.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.785833263397217,
|
|
"epoch": 0.22094140249759847,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998237290373494,
|
|
"loss": 5.5975,
|
|
"mean_token_accuracy": 0.1779184401035309,
|
|
"num_tokens": 5289107.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.854281902313232,
|
|
"epoch": 0.22142170989433238,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499822369630386,
|
|
"loss": 5.6323,
|
|
"mean_token_accuracy": 0.16607864797115326,
|
|
"num_tokens": 5301489.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.707867050170899,
|
|
"epoch": 0.2219020172910663,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998210050037148,
|
|
"loss": 5.6112,
|
|
"mean_token_accuracy": 0.18296469151973724,
|
|
"num_tokens": 5313273.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.856090545654297,
|
|
"epoch": 0.2223823246878002,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998196351573674,
|
|
"loss": 5.6373,
|
|
"mean_token_accuracy": 0.17394472658634186,
|
|
"num_tokens": 5324227.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.788931798934937,
|
|
"epoch": 0.2228626320845341,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998182600913757,
|
|
"loss": 5.583,
|
|
"mean_token_accuracy": 0.18264526426792144,
|
|
"num_tokens": 5335138.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.759352922439575,
|
|
"epoch": 0.22334293948126802,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998168798057715,
|
|
"loss": 5.653,
|
|
"mean_token_accuracy": 0.17215364724397658,
|
|
"num_tokens": 5345652.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.9246961116790775,
|
|
"epoch": 0.22382324687800192,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000499815494300587,
|
|
"loss": 5.7542,
|
|
"mean_token_accuracy": 0.17096742987632751,
|
|
"num_tokens": 5357999.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.823276519775391,
|
|
"epoch": 0.22430355427473583,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998141035758542,
|
|
"loss": 5.6663,
|
|
"mean_token_accuracy": 0.1679316133260727,
|
|
"num_tokens": 5369032.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.7893892288208,
|
|
"epoch": 0.22478386167146974,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998127076316054,
|
|
"loss": 5.5745,
|
|
"mean_token_accuracy": 0.17573301643133163,
|
|
"num_tokens": 5380692.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.879408645629883,
|
|
"epoch": 0.22526416906820365,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998113064678734,
|
|
"loss": 5.7149,
|
|
"mean_token_accuracy": 0.16351985484361647,
|
|
"num_tokens": 5392320.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.847774028778076,
|
|
"epoch": 0.22574447646493756,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998099000846901,
|
|
"loss": 5.6944,
|
|
"mean_token_accuracy": 0.17470421493053437,
|
|
"num_tokens": 5404374.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.8599584102630615,
|
|
"epoch": 0.22622478386167147,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998084884820887,
|
|
"loss": 5.6919,
|
|
"mean_token_accuracy": 0.17138945311307907,
|
|
"num_tokens": 5415471.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.725100374221801,
|
|
"epoch": 0.22670509125840538,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998070716601016,
|
|
"loss": 5.5278,
|
|
"mean_token_accuracy": 0.1748400166630745,
|
|
"num_tokens": 5425971.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.821712970733643,
|
|
"epoch": 0.2271853986551393,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998056496187618,
|
|
"loss": 5.643,
|
|
"mean_token_accuracy": 0.18096609115600587,
|
|
"num_tokens": 5437638.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.868386602401733,
|
|
"epoch": 0.2276657060518732,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998042223581025,
|
|
"loss": 5.6804,
|
|
"mean_token_accuracy": 0.17262430340051652,
|
|
"num_tokens": 5449463.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.809400224685669,
|
|
"epoch": 0.2281460134486071,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998027898781565,
|
|
"loss": 5.7433,
|
|
"mean_token_accuracy": 0.17124372869729995,
|
|
"num_tokens": 5460972.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.957932853698731,
|
|
"epoch": 0.228626320845341,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998013521789574,
|
|
"loss": 5.7479,
|
|
"mean_token_accuracy": 0.16553788185119628,
|
|
"num_tokens": 5471575.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.7461940288543705,
|
|
"epoch": 0.22910662824207492,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997999092605384,
|
|
"loss": 5.5997,
|
|
"mean_token_accuracy": 0.17493402808904648,
|
|
"num_tokens": 5483041.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.791565608978272,
|
|
"epoch": 0.22958693563880883,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499798461122933,
|
|
"loss": 5.6139,
|
|
"mean_token_accuracy": 0.17630909532308578,
|
|
"num_tokens": 5494040.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.730622577667236,
|
|
"epoch": 0.23006724303554274,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997970077661748,
|
|
"loss": 5.5301,
|
|
"mean_token_accuracy": 0.18595009446144103,
|
|
"num_tokens": 5504217.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.860027551651001,
|
|
"epoch": 0.23054755043227665,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997955491902977,
|
|
"loss": 5.6931,
|
|
"mean_token_accuracy": 0.1753861427307129,
|
|
"num_tokens": 5517180.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.809644031524658,
|
|
"epoch": 0.23102785782901056,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997940853953354,
|
|
"loss": 5.6133,
|
|
"mean_token_accuracy": 0.16900173276662828,
|
|
"num_tokens": 5529626.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.8468280792236325,
|
|
"epoch": 0.23150816522574447,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499792616381322,
|
|
"loss": 5.6331,
|
|
"mean_token_accuracy": 0.18043770939111708,
|
|
"num_tokens": 5540844.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.825774335861206,
|
|
"epoch": 0.23198847262247838,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997911421482914,
|
|
"loss": 5.5693,
|
|
"mean_token_accuracy": 0.17974244356155394,
|
|
"num_tokens": 5551732.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.826646757125855,
|
|
"epoch": 0.23246878001921228,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499789662696278,
|
|
"loss": 5.7035,
|
|
"mean_token_accuracy": 0.17213926166296006,
|
|
"num_tokens": 5562692.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.881090593338013,
|
|
"epoch": 0.2329490874159462,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997881780253162,
|
|
"loss": 5.6863,
|
|
"mean_token_accuracy": 0.1720082253217697,
|
|
"num_tokens": 5574884.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.716880178451538,
|
|
"epoch": 0.2334293948126801,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997866881354403,
|
|
"loss": 5.5235,
|
|
"mean_token_accuracy": 0.18319079726934434,
|
|
"num_tokens": 5585335.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.850202512741089,
|
|
"epoch": 0.23390970220941404,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499785193026685,
|
|
"loss": 5.693,
|
|
"mean_token_accuracy": 0.17081967294216155,
|
|
"num_tokens": 5597621.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.732968711853028,
|
|
"epoch": 0.23439000960614795,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004997836926990851,
|
|
"loss": 5.5419,
|
|
"mean_token_accuracy": 0.17374192029237748,
|
|
"num_tokens": 5609795.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.789673328399658,
|
|
"epoch": 0.23487031700288186,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997821871526752,
|
|
"loss": 5.6515,
|
|
"mean_token_accuracy": 0.173040571808815,
|
|
"num_tokens": 5621473.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.892057800292969,
|
|
"epoch": 0.23535062439961577,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004997806763874905,
|
|
"loss": 5.7274,
|
|
"mean_token_accuracy": 0.17482251077890396,
|
|
"num_tokens": 5635064.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.872965669631958,
|
|
"epoch": 0.23583093179634967,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997791604035659,
|
|
"loss": 5.7318,
|
|
"mean_token_accuracy": 0.16974180638790132,
|
|
"num_tokens": 5647382.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.820632839202881,
|
|
"epoch": 0.23631123919308358,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997776392009366,
|
|
"loss": 5.6123,
|
|
"mean_token_accuracy": 0.17786978632211686,
|
|
"num_tokens": 5658417.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.767739152908325,
|
|
"epoch": 0.2367915465898175,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997761127796381,
|
|
"loss": 5.5868,
|
|
"mean_token_accuracy": 0.18194347023963928,
|
|
"num_tokens": 5672190.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.772201633453369,
|
|
"epoch": 0.2372718539865514,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997745811397056,
|
|
"loss": 5.595,
|
|
"mean_token_accuracy": 0.1847991779446602,
|
|
"num_tokens": 5684259.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.729731464385987,
|
|
"epoch": 0.2377521613832853,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997730442811748,
|
|
"loss": 5.665,
|
|
"mean_token_accuracy": 0.1762344717979431,
|
|
"num_tokens": 5696505.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.725724935531616,
|
|
"epoch": 0.23823246878001922,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997715022040814,
|
|
"loss": 5.5436,
|
|
"mean_token_accuracy": 0.181012162566185,
|
|
"num_tokens": 5707638.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.763240242004395,
|
|
"epoch": 0.23871277617675313,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000499769954908461,
|
|
"loss": 5.5307,
|
|
"mean_token_accuracy": 0.18620459139347076,
|
|
"num_tokens": 5719098.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.9247959613800045,
|
|
"epoch": 0.23919308357348704,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004997684023943498,
|
|
"loss": 5.7533,
|
|
"mean_token_accuracy": 0.1620590589940548,
|
|
"num_tokens": 5731345.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.656325578689575,
|
|
"epoch": 0.23967339097022095,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997668446617837,
|
|
"loss": 5.5052,
|
|
"mean_token_accuracy": 0.17786547392606736,
|
|
"num_tokens": 5742604.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.699789333343506,
|
|
"epoch": 0.24015369836695485,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997652817107989,
|
|
"loss": 5.5488,
|
|
"mean_token_accuracy": 0.18326022177934648,
|
|
"num_tokens": 5754620.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.7961937427520756,
|
|
"epoch": 0.24063400576368876,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997637135414315,
|
|
"loss": 5.6351,
|
|
"mean_token_accuracy": 0.17634303420782088,
|
|
"num_tokens": 5764764.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.671987724304199,
|
|
"epoch": 0.24111431316042267,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997621401537183,
|
|
"loss": 5.518,
|
|
"mean_token_accuracy": 0.1902934193611145,
|
|
"num_tokens": 5776341.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.814025735855102,
|
|
"epoch": 0.24159462055715658,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997605615476955,
|
|
"loss": 5.7332,
|
|
"mean_token_accuracy": 0.16588144600391388,
|
|
"num_tokens": 5788604.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.8230632781982425,
|
|
"epoch": 0.2420749279538905,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997589777234,
|
|
"loss": 5.6187,
|
|
"mean_token_accuracy": 0.17924741953611373,
|
|
"num_tokens": 5800194.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.757388734817505,
|
|
"epoch": 0.2425552353506244,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004997573886808684,
|
|
"loss": 5.6105,
|
|
"mean_token_accuracy": 0.18109989911317825,
|
|
"num_tokens": 5812978.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.717558908462524,
|
|
"epoch": 0.2430355427473583,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004997557944201375,
|
|
"loss": 5.5223,
|
|
"mean_token_accuracy": 0.18487590551376343,
|
|
"num_tokens": 5824591.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.745062971115113,
|
|
"epoch": 0.24351585014409222,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997541949412445,
|
|
"loss": 5.6232,
|
|
"mean_token_accuracy": 0.17669124156236649,
|
|
"num_tokens": 5835970.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.807597732543945,
|
|
"epoch": 0.24399615754082613,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997525902442266,
|
|
"loss": 5.6655,
|
|
"mean_token_accuracy": 0.16880002468824387,
|
|
"num_tokens": 5848687.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.737652158737182,
|
|
"epoch": 0.24447646493756003,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997509803291207,
|
|
"loss": 5.6236,
|
|
"mean_token_accuracy": 0.17555102258920668,
|
|
"num_tokens": 5860820.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.786333513259888,
|
|
"epoch": 0.24495677233429394,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997493651959647,
|
|
"loss": 5.5817,
|
|
"mean_token_accuracy": 0.1809490665793419,
|
|
"num_tokens": 5872172.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.769077014923096,
|
|
"epoch": 0.24543707973102785,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997477448447955,
|
|
"loss": 5.6131,
|
|
"mean_token_accuracy": 0.17688888311386108,
|
|
"num_tokens": 5883574.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.845238304138183,
|
|
"epoch": 0.24591738712776176,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997461192756512,
|
|
"loss": 5.7127,
|
|
"mean_token_accuracy": 0.17598800212144852,
|
|
"num_tokens": 5895223.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.752743339538574,
|
|
"epoch": 0.24639769452449567,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997444884885694,
|
|
"loss": 5.5381,
|
|
"mean_token_accuracy": 0.17801688909530639,
|
|
"num_tokens": 5907448.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.751397609710693,
|
|
"epoch": 0.24687800192122958,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997428524835879,
|
|
"loss": 5.6376,
|
|
"mean_token_accuracy": 0.17808189988136292,
|
|
"num_tokens": 5918906.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.76022891998291,
|
|
"epoch": 0.2473583093179635,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004997412112607446,
|
|
"loss": 5.5087,
|
|
"mean_token_accuracy": 0.1827773556113243,
|
|
"num_tokens": 5930619.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.866481304168701,
|
|
"epoch": 0.2478386167146974,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997395648200778,
|
|
"loss": 5.6996,
|
|
"mean_token_accuracy": 0.17080766707658768,
|
|
"num_tokens": 5941554.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.791235780715942,
|
|
"epoch": 0.2483189241114313,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997379131616257,
|
|
"loss": 5.6619,
|
|
"mean_token_accuracy": 0.1792931705713272,
|
|
"num_tokens": 5953309.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.759411811828613,
|
|
"epoch": 0.24879923150816521,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997362562854266,
|
|
"loss": 5.6054,
|
|
"mean_token_accuracy": 0.17513532787561417,
|
|
"num_tokens": 5965286.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.769647026062012,
|
|
"epoch": 0.24927953890489912,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997345941915187,
|
|
"loss": 5.5939,
|
|
"mean_token_accuracy": 0.17810043692588806,
|
|
"num_tokens": 5976558.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.803124666213989,
|
|
"epoch": 0.24975984630163303,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997329268799412,
|
|
"loss": 5.6373,
|
|
"mean_token_accuracy": 0.16947010159492493,
|
|
"num_tokens": 5988793.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.7729833126068115,
|
|
"epoch": 0.25024015369836694,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997312543507322,
|
|
"loss": 5.6674,
|
|
"mean_token_accuracy": 0.16540661901235582,
|
|
"num_tokens": 5999825.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.835980033874511,
|
|
"epoch": 0.2507204610951009,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997295766039309,
|
|
"loss": 5.6256,
|
|
"mean_token_accuracy": 0.17402078658342363,
|
|
"num_tokens": 6010408.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.805245876312256,
|
|
"epoch": 0.25120076849183476,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997278936395761,
|
|
"loss": 5.6382,
|
|
"mean_token_accuracy": 0.17170876115560532,
|
|
"num_tokens": 6021161.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.706417989730835,
|
|
"epoch": 0.2516810758885687,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997262054577071,
|
|
"loss": 5.4989,
|
|
"mean_token_accuracy": 0.18810333758592607,
|
|
"num_tokens": 6032646.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.766708278656006,
|
|
"epoch": 0.2521613832853026,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997245120583627,
|
|
"loss": 5.6631,
|
|
"mean_token_accuracy": 0.17591285109519958,
|
|
"num_tokens": 6045019.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.657771825790405,
|
|
"epoch": 0.2526416906820365,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997228134415825,
|
|
"loss": 5.5197,
|
|
"mean_token_accuracy": 0.1834905758500099,
|
|
"num_tokens": 6055644.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.78851728439331,
|
|
"epoch": 0.2531219980787704,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997211096074059,
|
|
"loss": 5.5693,
|
|
"mean_token_accuracy": 0.17673975676298143,
|
|
"num_tokens": 6066488.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.758440732955933,
|
|
"epoch": 0.25360230547550433,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997194005558722,
|
|
"loss": 5.5965,
|
|
"mean_token_accuracy": 0.17449358552694322,
|
|
"num_tokens": 6078300.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.734206056594848,
|
|
"epoch": 0.2540826128722382,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997176862870216,
|
|
"loss": 5.5783,
|
|
"mean_token_accuracy": 0.1704964131116867,
|
|
"num_tokens": 6089180.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.757406234741211,
|
|
"epoch": 0.25456292026897215,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997159668008933,
|
|
"loss": 5.618,
|
|
"mean_token_accuracy": 0.1837097004055977,
|
|
"num_tokens": 6099093.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.790572166442871,
|
|
"epoch": 0.25504322766570603,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997142420975277,
|
|
"loss": 5.626,
|
|
"mean_token_accuracy": 0.1773001104593277,
|
|
"num_tokens": 6109161.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.86971116065979,
|
|
"epoch": 0.25552353506243997,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997125121769647,
|
|
"loss": 5.7318,
|
|
"mean_token_accuracy": 0.17001539021730422,
|
|
"num_tokens": 6120031.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.717875337600708,
|
|
"epoch": 0.25600384245917385,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997107770392444,
|
|
"loss": 5.5845,
|
|
"mean_token_accuracy": 0.17720083743333817,
|
|
"num_tokens": 6132398.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.713741064071655,
|
|
"epoch": 0.2564841498559078,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499709036684407,
|
|
"loss": 5.5721,
|
|
"mean_token_accuracy": 0.17020554840564728,
|
|
"num_tokens": 6143099.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.752955055236816,
|
|
"epoch": 0.25696445725264166,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997072911124932,
|
|
"loss": 5.5384,
|
|
"mean_token_accuracy": 0.1820665955543518,
|
|
"num_tokens": 6154815.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.738240671157837,
|
|
"epoch": 0.2574447646493756,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997055403235432,
|
|
"loss": 5.596,
|
|
"mean_token_accuracy": 0.1818923234939575,
|
|
"num_tokens": 6165696.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.721163082122803,
|
|
"epoch": 0.2579250720461095,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997037843175978,
|
|
"loss": 5.6238,
|
|
"mean_token_accuracy": 0.17519191652536392,
|
|
"num_tokens": 6177042.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.815378189086914,
|
|
"epoch": 0.2584053794428434,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997020230946978,
|
|
"loss": 5.6137,
|
|
"mean_token_accuracy": 0.1751040607690811,
|
|
"num_tokens": 6188691.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.593714475631714,
|
|
"epoch": 0.25888568683957736,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997002566548841,
|
|
"loss": 5.4373,
|
|
"mean_token_accuracy": 0.18646937608718872,
|
|
"num_tokens": 6199931.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.693178653717041,
|
|
"epoch": 0.25936599423631124,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996984849981976,
|
|
"loss": 5.5608,
|
|
"mean_token_accuracy": 0.18345472663640977,
|
|
"num_tokens": 6211493.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.793261384963989,
|
|
"epoch": 0.2598463016330452,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996967081246794,
|
|
"loss": 5.5049,
|
|
"mean_token_accuracy": 0.1827055275440216,
|
|
"num_tokens": 6221505.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.638125991821289,
|
|
"epoch": 0.26032660902977905,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996949260343711,
|
|
"loss": 5.4859,
|
|
"mean_token_accuracy": 0.1871676668524742,
|
|
"num_tokens": 6233058.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.74454984664917,
|
|
"epoch": 0.260806916426513,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004996931387273137,
|
|
"loss": 5.6297,
|
|
"mean_token_accuracy": 0.17437177300453185,
|
|
"num_tokens": 6244713.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.722670650482177,
|
|
"epoch": 0.2612872238232469,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004996913462035487,
|
|
"loss": 5.5861,
|
|
"mean_token_accuracy": 0.17518469989299773,
|
|
"num_tokens": 6256624.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.696188926696777,
|
|
"epoch": 0.2617675312199808,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000499689548463118,
|
|
"loss": 5.5856,
|
|
"mean_token_accuracy": 0.1736260771751404,
|
|
"num_tokens": 6271134.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.834325361251831,
|
|
"epoch": 0.2622478386167147,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996877455060631,
|
|
"loss": 5.6834,
|
|
"mean_token_accuracy": 0.17634287625551223,
|
|
"num_tokens": 6282082.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.782431602478027,
|
|
"epoch": 0.2627281460134486,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996859373324259,
|
|
"loss": 5.6268,
|
|
"mean_token_accuracy": 0.17405078411102295,
|
|
"num_tokens": 6293869.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.699025249481201,
|
|
"epoch": 0.2632084534101825,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996841239422485,
|
|
"loss": 5.6431,
|
|
"mean_token_accuracy": 0.17163943350315095,
|
|
"num_tokens": 6304858.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.754938268661499,
|
|
"epoch": 0.26368876080691644,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996823053355729,
|
|
"loss": 5.57,
|
|
"mean_token_accuracy": 0.1830345094203949,
|
|
"num_tokens": 6316738.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.702230501174927,
|
|
"epoch": 0.2641690682036503,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996804815124413,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.17675393223762512,
|
|
"num_tokens": 6327691.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.787597799301148,
|
|
"epoch": 0.26464937560038426,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996786524728962,
|
|
"loss": 5.5546,
|
|
"mean_token_accuracy": 0.17426677644252778,
|
|
"num_tokens": 6337964.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.631732511520386,
|
|
"epoch": 0.26512968299711814,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996768182169797,
|
|
"loss": 5.5951,
|
|
"mean_token_accuracy": 0.1759776771068573,
|
|
"num_tokens": 6349764.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.8128139019012455,
|
|
"epoch": 0.2656099903938521,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996749787447349,
|
|
"loss": 5.5442,
|
|
"mean_token_accuracy": 0.18075311332941055,
|
|
"num_tokens": 6360538.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.763351535797119,
|
|
"epoch": 0.26609029779058596,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499673134056204,
|
|
"loss": 5.5971,
|
|
"mean_token_accuracy": 0.17973174452781676,
|
|
"num_tokens": 6372489.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.568811273574829,
|
|
"epoch": 0.2665706051873199,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996712841514303,
|
|
"loss": 5.4534,
|
|
"mean_token_accuracy": 0.18664876222610474,
|
|
"num_tokens": 6383080.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.806193017959595,
|
|
"epoch": 0.2670509125840538,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996694290304563,
|
|
"loss": 5.5947,
|
|
"mean_token_accuracy": 0.1757884532213211,
|
|
"num_tokens": 6394033.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.715140008926392,
|
|
"epoch": 0.2675312199807877,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004996675686933255,
|
|
"loss": 5.5966,
|
|
"mean_token_accuracy": 0.17878031432628633,
|
|
"num_tokens": 6404990.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.701559019088745,
|
|
"epoch": 0.2680115273775216,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004996657031400807,
|
|
"loss": 5.4591,
|
|
"mean_token_accuracy": 0.18534403890371323,
|
|
"num_tokens": 6416337.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.730889225006104,
|
|
"epoch": 0.26849183477425553,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996638323707655,
|
|
"loss": 5.6656,
|
|
"mean_token_accuracy": 0.1772391602396965,
|
|
"num_tokens": 6427892.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.69028673171997,
|
|
"epoch": 0.2689721421709894,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996619563854232,
|
|
"loss": 5.5753,
|
|
"mean_token_accuracy": 0.17937600910663604,
|
|
"num_tokens": 6440300.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.708783483505249,
|
|
"epoch": 0.26945244956772335,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004996600751840974,
|
|
"loss": 5.4715,
|
|
"mean_token_accuracy": 0.18418766260147096,
|
|
"num_tokens": 6451499.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.712428379058838,
|
|
"epoch": 0.26993275696445723,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996581887668317,
|
|
"loss": 5.5552,
|
|
"mean_token_accuracy": 0.18144695162773133,
|
|
"num_tokens": 6462882.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.6650489330291744,
|
|
"epoch": 0.27041306436119117,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00049965629713367,
|
|
"loss": 5.5474,
|
|
"mean_token_accuracy": 0.1818947970867157,
|
|
"num_tokens": 6475042.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.719558477401733,
|
|
"epoch": 0.27089337175792505,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004996544002846561,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.1792231246829033,
|
|
"num_tokens": 6486245.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.716267395019531,
|
|
"epoch": 0.271373679154659,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996524982198343,
|
|
"loss": 5.4454,
|
|
"mean_token_accuracy": 0.18538395762443544,
|
|
"num_tokens": 6497705.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.62763934135437,
|
|
"epoch": 0.27185398655139287,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996505909392485,
|
|
"loss": 5.493,
|
|
"mean_token_accuracy": 0.1870325416326523,
|
|
"num_tokens": 6508193.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.725575256347656,
|
|
"epoch": 0.2723342939481268,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996486784429429,
|
|
"loss": 5.6016,
|
|
"mean_token_accuracy": 0.1776618927717209,
|
|
"num_tokens": 6519228.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.755958032608032,
|
|
"epoch": 0.2728146013448607,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996467607309622,
|
|
"loss": 5.5569,
|
|
"mean_token_accuracy": 0.180067740380764,
|
|
"num_tokens": 6530743.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.667758846282959,
|
|
"epoch": 0.2732949087415946,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004996448378033507,
|
|
"loss": 5.4455,
|
|
"mean_token_accuracy": 0.17980502247810365,
|
|
"num_tokens": 6542588.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.799747085571289,
|
|
"epoch": 0.2737752161383285,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004996429096601532,
|
|
"loss": 5.6474,
|
|
"mean_token_accuracy": 0.16938518285751342,
|
|
"num_tokens": 6554002.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.747061014175415,
|
|
"epoch": 0.27425552353506244,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996409763014144,
|
|
"loss": 5.6675,
|
|
"mean_token_accuracy": 0.17701776772737504,
|
|
"num_tokens": 6563358.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.683012914657593,
|
|
"epoch": 0.2747358309317964,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996390377271791,
|
|
"loss": 5.5797,
|
|
"mean_token_accuracy": 0.17571266144514083,
|
|
"num_tokens": 6574986.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.7084794521331785,
|
|
"epoch": 0.27521613832853026,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996370939374924,
|
|
"loss": 5.5201,
|
|
"mean_token_accuracy": 0.18074920177459716,
|
|
"num_tokens": 6587407.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.637572622299194,
|
|
"epoch": 0.2756964457252642,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996351449323994,
|
|
"loss": 5.4845,
|
|
"mean_token_accuracy": 0.1878151774406433,
|
|
"num_tokens": 6598483.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.657338953018188,
|
|
"epoch": 0.2761767531219981,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004996331907119455,
|
|
"loss": 5.5831,
|
|
"mean_token_accuracy": 0.18044260144233704,
|
|
"num_tokens": 6610749.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.7645731449127195,
|
|
"epoch": 0.276657060518732,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996312312761758,
|
|
"loss": 5.5559,
|
|
"mean_token_accuracy": 0.17333664000034332,
|
|
"num_tokens": 6621762.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.7428053855896,
|
|
"epoch": 0.2771373679154659,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499629266625136,
|
|
"loss": 5.6111,
|
|
"mean_token_accuracy": 0.1731718435883522,
|
|
"num_tokens": 6632296.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.733534955978394,
|
|
"epoch": 0.27761767531219983,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996272967588715,
|
|
"loss": 5.5609,
|
|
"mean_token_accuracy": 0.18240266442298889,
|
|
"num_tokens": 6644413.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.6336816310882565,
|
|
"epoch": 0.2780979827089337,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004996253216774283,
|
|
"loss": 5.5107,
|
|
"mean_token_accuracy": 0.17972939014434813,
|
|
"num_tokens": 6655634.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.676279926300049,
|
|
"epoch": 0.27857829010566765,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996233413808521,
|
|
"loss": 5.4526,
|
|
"mean_token_accuracy": 0.1837601602077484,
|
|
"num_tokens": 6665977.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.707220602035522,
|
|
"epoch": 0.27905859750240153,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004996213558691889,
|
|
"loss": 5.5186,
|
|
"mean_token_accuracy": 0.17969339042901994,
|
|
"num_tokens": 6677165.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.689497089385986,
|
|
"epoch": 0.27953890489913547,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004996193651424848,
|
|
"loss": 5.6297,
|
|
"mean_token_accuracy": 0.17441802322864533,
|
|
"num_tokens": 6688182.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.734628391265869,
|
|
"epoch": 0.28001921229586935,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499617369200786,
|
|
"loss": 5.5334,
|
|
"mean_token_accuracy": 0.17993243038654327,
|
|
"num_tokens": 6699259.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.700125122070313,
|
|
"epoch": 0.2804995196926033,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996153680441389,
|
|
"loss": 5.5422,
|
|
"mean_token_accuracy": 0.17867809534072876,
|
|
"num_tokens": 6710071.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.7129288673400875,
|
|
"epoch": 0.28097982708933716,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00049961336167259,
|
|
"loss": 5.583,
|
|
"mean_token_accuracy": 0.1841597393155098,
|
|
"num_tokens": 6720267.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.750007438659668,
|
|
"epoch": 0.2814601344860711,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996113500861857,
|
|
"loss": 5.5725,
|
|
"mean_token_accuracy": 0.17657051831483841,
|
|
"num_tokens": 6730919.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.648938798904419,
|
|
"epoch": 0.281940441882805,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996093332849729,
|
|
"loss": 5.557,
|
|
"mean_token_accuracy": 0.18033284693956375,
|
|
"num_tokens": 6741480.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.743981218338012,
|
|
"epoch": 0.2824207492795389,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996073112689983,
|
|
"loss": 5.4984,
|
|
"mean_token_accuracy": 0.18312747478485109,
|
|
"num_tokens": 6751816.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.641194438934326,
|
|
"epoch": 0.2829010566762728,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996052840383088,
|
|
"loss": 5.5253,
|
|
"mean_token_accuracy": 0.18100760728120804,
|
|
"num_tokens": 6763645.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.674943637847901,
|
|
"epoch": 0.28338136407300674,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996032515929516,
|
|
"loss": 5.4977,
|
|
"mean_token_accuracy": 0.18233796060085297,
|
|
"num_tokens": 6774795.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.654936456680298,
|
|
"epoch": 0.2838616714697406,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004996012139329738,
|
|
"loss": 5.4161,
|
|
"mean_token_accuracy": 0.18299856334924697,
|
|
"num_tokens": 6786833.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.561620569229126,
|
|
"epoch": 0.28434197886647455,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004995991710584228,
|
|
"loss": 5.4387,
|
|
"mean_token_accuracy": 0.18655213415622712,
|
|
"num_tokens": 6798134.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.713185405731201,
|
|
"epoch": 0.28482228626320844,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004995971229693459,
|
|
"loss": 5.5493,
|
|
"mean_token_accuracy": 0.18436760306358338,
|
|
"num_tokens": 6807918.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.7143415927886965,
|
|
"epoch": 0.28530259365994237,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995950696657909,
|
|
"loss": 5.621,
|
|
"mean_token_accuracy": 0.17538298517465592,
|
|
"num_tokens": 6818689.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.6440918922424315,
|
|
"epoch": 0.28578290105667625,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004995930111478051,
|
|
"loss": 5.4919,
|
|
"mean_token_accuracy": 0.18471998870372772,
|
|
"num_tokens": 6830871.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.652586126327515,
|
|
"epoch": 0.2862632084534102,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995909474154365,
|
|
"loss": 5.4778,
|
|
"mean_token_accuracy": 0.18692311495542527,
|
|
"num_tokens": 6842857.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.662132787704468,
|
|
"epoch": 0.28674351585014407,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995888784687331,
|
|
"loss": 5.5059,
|
|
"mean_token_accuracy": 0.18150224685668945,
|
|
"num_tokens": 6853727.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.701902055740357,
|
|
"epoch": 0.287223823246878,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004995868043077428,
|
|
"loss": 5.5331,
|
|
"mean_token_accuracy": 0.181272454559803,
|
|
"num_tokens": 6864728.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.726777839660644,
|
|
"epoch": 0.2877041306436119,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995847249325137,
|
|
"loss": 5.606,
|
|
"mean_token_accuracy": 0.17763462513685227,
|
|
"num_tokens": 6875001.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.619700956344604,
|
|
"epoch": 0.2881844380403458,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004995826403430942,
|
|
"loss": 5.4719,
|
|
"mean_token_accuracy": 0.18973949998617173,
|
|
"num_tokens": 6887026.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.2881844380403458,
|
|
"eval_entropy": 5.502902475233371,
|
|
"eval_loss": 5.541717052459717,
|
|
"eval_mean_token_accuracy": 0.1886520807157706,
|
|
"eval_num_tokens": 6887026.0,
|
|
"eval_runtime": 20.5725,
|
|
"eval_samples_per_second": 1595.091,
|
|
"eval_steps_per_second": 199.392,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 104090,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 3000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1569715698401280.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|