7035 lines
192 KiB
JSON
7035 lines
192 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.20338786065026004,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742584228515625,
|
|
"epoch": 0.0002905540866432286,
|
|
"grad_norm": 4.90625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7837,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10156.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742587471008301,
|
|
"epoch": 0.0005811081732864572,
|
|
"grad_norm": 4.8125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7753,
|
|
"mean_token_accuracy": 9.267840650863945e-05,
|
|
"num_tokens": 20933.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.74257869720459,
|
|
"epoch": 0.0008716622599296859,
|
|
"grad_norm": 4.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7508,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 31298.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742635726928711,
|
|
"epoch": 0.0011622163465729145,
|
|
"grad_norm": 4.875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.697,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 40913.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742652702331544,
|
|
"epoch": 0.0014527704332161432,
|
|
"grad_norm": 4.28125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5798,
|
|
"mean_token_accuracy": 0.0007269373920280487,
|
|
"num_tokens": 49901.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742454719543456,
|
|
"epoch": 0.0017433245198593718,
|
|
"grad_norm": 4.0625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4688,
|
|
"mean_token_accuracy": 0.01560134175233543,
|
|
"num_tokens": 59328.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741775226593017,
|
|
"epoch": 0.0020338786065026006,
|
|
"grad_norm": 3.25,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3287,
|
|
"mean_token_accuracy": 0.037073963694274424,
|
|
"num_tokens": 68405.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.740037631988525,
|
|
"epoch": 0.002324432693145829,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2203,
|
|
"mean_token_accuracy": 0.037133642472326756,
|
|
"num_tokens": 77591.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.73731575012207,
|
|
"epoch": 0.0026149867797890577,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.1202,
|
|
"mean_token_accuracy": 0.03901108838617802,
|
|
"num_tokens": 88186.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.734606838226318,
|
|
"epoch": 0.0029055408664322865,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.0211,
|
|
"mean_token_accuracy": 0.04241710864007473,
|
|
"num_tokens": 97594.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.73211612701416,
|
|
"epoch": 0.003196094953075515,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9871,
|
|
"mean_token_accuracy": 0.03826836366206408,
|
|
"num_tokens": 107386.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.73102445602417,
|
|
"epoch": 0.0034866490397187436,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.9132,
|
|
"mean_token_accuracy": 0.03943221494555473,
|
|
"num_tokens": 116742.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.729645252227783,
|
|
"epoch": 0.0037772031263619723,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8519,
|
|
"mean_token_accuracy": 0.03962419871240854,
|
|
"num_tokens": 126520.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.727834033966065,
|
|
"epoch": 0.004067757213005201,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7907,
|
|
"mean_token_accuracy": 0.03989919070154428,
|
|
"num_tokens": 136382.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.724947452545166,
|
|
"epoch": 0.0043583112996484295,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7212,
|
|
"mean_token_accuracy": 0.03671109899878502,
|
|
"num_tokens": 146435.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.72182493209839,
|
|
"epoch": 0.004648865386291658,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6591,
|
|
"mean_token_accuracy": 0.037667426839470865,
|
|
"num_tokens": 156174.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.71723222732544,
|
|
"epoch": 0.004939419472934887,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5783,
|
|
"mean_token_accuracy": 0.04142397493124008,
|
|
"num_tokens": 165118.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.708585739135742,
|
|
"epoch": 0.005229973559578115,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.5252,
|
|
"mean_token_accuracy": 0.04036426953971386,
|
|
"num_tokens": 174401.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.697160243988037,
|
|
"epoch": 0.005520527646221344,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.443,
|
|
"mean_token_accuracy": 0.04118307866156101,
|
|
"num_tokens": 183533.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.683875274658202,
|
|
"epoch": 0.005811081732864573,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3596,
|
|
"mean_token_accuracy": 0.045602331310510634,
|
|
"num_tokens": 193296.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.665638542175293,
|
|
"epoch": 0.006101635819507801,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2242,
|
|
"mean_token_accuracy": 0.055989645794034,
|
|
"num_tokens": 202741.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.637047958374023,
|
|
"epoch": 0.00639218990615103,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1359,
|
|
"mean_token_accuracy": 0.05134495124220848,
|
|
"num_tokens": 212441.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.61000461578369,
|
|
"epoch": 0.006682743992794259,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9868,
|
|
"mean_token_accuracy": 0.04918566383421421,
|
|
"num_tokens": 220671.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.56931962966919,
|
|
"epoch": 0.006973298079437487,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9878,
|
|
"mean_token_accuracy": 0.04560479037463665,
|
|
"num_tokens": 231390.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.515452098846435,
|
|
"epoch": 0.007263852166080716,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.8241,
|
|
"mean_token_accuracy": 0.05023673102259636,
|
|
"num_tokens": 241137.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.430156230926514,
|
|
"epoch": 0.007554406252723945,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6778,
|
|
"mean_token_accuracy": 0.05138532817363739,
|
|
"num_tokens": 250627.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.353140926361084,
|
|
"epoch": 0.007844960339367173,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.5255,
|
|
"mean_token_accuracy": 0.05529710613191128,
|
|
"num_tokens": 259564.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.262280082702636,
|
|
"epoch": 0.008135514426010402,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4168,
|
|
"mean_token_accuracy": 0.05102897398173809,
|
|
"num_tokens": 268997.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.17268762588501,
|
|
"epoch": 0.00842606851265363,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.402,
|
|
"mean_token_accuracy": 0.04707291163504124,
|
|
"num_tokens": 278989.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.068906784057617,
|
|
"epoch": 0.008716622599296859,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.2195,
|
|
"mean_token_accuracy": 0.04823922924697399,
|
|
"num_tokens": 288770.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.884156227111816,
|
|
"epoch": 0.009007176685940088,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1604,
|
|
"mean_token_accuracy": 0.05296766012907028,
|
|
"num_tokens": 298368.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.749515438079834,
|
|
"epoch": 0.009297730772583316,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9887,
|
|
"mean_token_accuracy": 0.054083061218261716,
|
|
"num_tokens": 307437.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.539670753479005,
|
|
"epoch": 0.009588284859226545,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.931,
|
|
"mean_token_accuracy": 0.05368399284780025,
|
|
"num_tokens": 317842.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.367785167694091,
|
|
"epoch": 0.009878838945869774,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7746,
|
|
"mean_token_accuracy": 0.056211471930146216,
|
|
"num_tokens": 327455.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.106531143188477,
|
|
"epoch": 0.010169393032513002,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.7023,
|
|
"mean_token_accuracy": 0.059121083468198776,
|
|
"num_tokens": 338593.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.891216564178468,
|
|
"epoch": 0.01045994711915623,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.6717,
|
|
"mean_token_accuracy": 0.060001150518655774,
|
|
"num_tokens": 348278.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.690237522125244,
|
|
"epoch": 0.01075050120579946,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5848,
|
|
"mean_token_accuracy": 0.060652027279138564,
|
|
"num_tokens": 358293.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.500970458984375,
|
|
"epoch": 0.011041055292442687,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.6462,
|
|
"mean_token_accuracy": 0.06345079019665718,
|
|
"num_tokens": 368177.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.432841682434082,
|
|
"epoch": 0.011331609379085917,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.5041,
|
|
"mean_token_accuracy": 0.06438801400363445,
|
|
"num_tokens": 377258.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.328762531280518,
|
|
"epoch": 0.011622163465729146,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.515,
|
|
"mean_token_accuracy": 0.06462946832180023,
|
|
"num_tokens": 385931.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.228355598449706,
|
|
"epoch": 0.011912717552372373,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4262,
|
|
"mean_token_accuracy": 0.06731356121599674,
|
|
"num_tokens": 394370.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.163572025299072,
|
|
"epoch": 0.012203271639015602,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.5127,
|
|
"mean_token_accuracy": 0.06187250129878521,
|
|
"num_tokens": 405167.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.144425964355468,
|
|
"epoch": 0.012493825725658832,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.4823,
|
|
"mean_token_accuracy": 0.06424942426383495,
|
|
"num_tokens": 414954.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.074434852600097,
|
|
"epoch": 0.01278437981230206,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.4379,
|
|
"mean_token_accuracy": 0.07021872885525227,
|
|
"num_tokens": 423806.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.100719451904297,
|
|
"epoch": 0.013074933898945288,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.4049,
|
|
"mean_token_accuracy": 0.07006631046533585,
|
|
"num_tokens": 433416.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.068440341949463,
|
|
"epoch": 0.013365487985588518,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.4086,
|
|
"mean_token_accuracy": 0.0656484205275774,
|
|
"num_tokens": 443237.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.008077144622803,
|
|
"epoch": 0.013656042072231747,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.3811,
|
|
"mean_token_accuracy": 0.07138268202543259,
|
|
"num_tokens": 452334.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.95733003616333,
|
|
"epoch": 0.013946596158874974,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.451,
|
|
"mean_token_accuracy": 0.07069577798247337,
|
|
"num_tokens": 462604.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.985943031311035,
|
|
"epoch": 0.014237150245518203,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3342,
|
|
"mean_token_accuracy": 0.07418472990393639,
|
|
"num_tokens": 472105.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.985353708267212,
|
|
"epoch": 0.014527704332161433,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.3573,
|
|
"mean_token_accuracy": 0.07192124761641025,
|
|
"num_tokens": 481873.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.852858924865723,
|
|
"epoch": 0.01481825841880466,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.3134,
|
|
"mean_token_accuracy": 0.07094009146094322,
|
|
"num_tokens": 490776.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.97090711593628,
|
|
"epoch": 0.01510881250544789,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3459,
|
|
"mean_token_accuracy": 0.06945950090885163,
|
|
"num_tokens": 500237.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.988322401046753,
|
|
"epoch": 0.015399366592091119,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.3569,
|
|
"mean_token_accuracy": 0.0719369538128376,
|
|
"num_tokens": 509449.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.863973140716553,
|
|
"epoch": 0.015689920678734346,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3463,
|
|
"mean_token_accuracy": 0.07629362866282463,
|
|
"num_tokens": 519335.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.850080347061157,
|
|
"epoch": 0.015980474765377575,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.269,
|
|
"mean_token_accuracy": 0.07348301075398922,
|
|
"num_tokens": 529108.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.803001642227173,
|
|
"epoch": 0.016271028852020804,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.3421,
|
|
"mean_token_accuracy": 0.07442944496870041,
|
|
"num_tokens": 539409.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.8401947021484375,
|
|
"epoch": 0.016561582938664034,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.269,
|
|
"mean_token_accuracy": 0.07476447969675064,
|
|
"num_tokens": 549790.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.773062610626221,
|
|
"epoch": 0.01685213702530726,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.2597,
|
|
"mean_token_accuracy": 0.07743276208639145,
|
|
"num_tokens": 559343.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.833370351791382,
|
|
"epoch": 0.01714269111195049,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.306,
|
|
"mean_token_accuracy": 0.07507650516927242,
|
|
"num_tokens": 568806.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.692620134353637,
|
|
"epoch": 0.017433245198593718,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.1532,
|
|
"mean_token_accuracy": 0.07671754881739616,
|
|
"num_tokens": 578988.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.840510559082031,
|
|
"epoch": 0.017723799285236947,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.258,
|
|
"mean_token_accuracy": 0.0767325557768345,
|
|
"num_tokens": 588588.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.740892934799194,
|
|
"epoch": 0.018014353371880176,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2385,
|
|
"mean_token_accuracy": 0.07767370343208313,
|
|
"num_tokens": 597957.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.761815309524536,
|
|
"epoch": 0.018304907458523405,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.2168,
|
|
"mean_token_accuracy": 0.07732245922088624,
|
|
"num_tokens": 607446.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.723113679885865,
|
|
"epoch": 0.01859546154516663,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.1559,
|
|
"mean_token_accuracy": 0.07753840312361718,
|
|
"num_tokens": 617064.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.695508337020874,
|
|
"epoch": 0.01888601563180986,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2008,
|
|
"mean_token_accuracy": 0.08057244047522545,
|
|
"num_tokens": 625927.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.717827177047729,
|
|
"epoch": 0.01917656971845309,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1152,
|
|
"mean_token_accuracy": 0.07994545996189117,
|
|
"num_tokens": 635341.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.675025224685669,
|
|
"epoch": 0.01946712380509632,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1106,
|
|
"mean_token_accuracy": 0.08988085016608238,
|
|
"num_tokens": 645095.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.714554166793823,
|
|
"epoch": 0.019757677891739548,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.1558,
|
|
"mean_token_accuracy": 0.0730321068316698,
|
|
"num_tokens": 654754.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.60111026763916,
|
|
"epoch": 0.020048231978382777,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1266,
|
|
"mean_token_accuracy": 0.07690966166555882,
|
|
"num_tokens": 664589.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.6628223896026615,
|
|
"epoch": 0.020338786065026003,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1425,
|
|
"mean_token_accuracy": 0.07918459475040436,
|
|
"num_tokens": 673870.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.577814197540283,
|
|
"epoch": 0.020629340151669232,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.1137,
|
|
"mean_token_accuracy": 0.07997918874025345,
|
|
"num_tokens": 684309.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.6769345760345455,
|
|
"epoch": 0.02091989423831246,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.1629,
|
|
"mean_token_accuracy": 0.07469077445566655,
|
|
"num_tokens": 693702.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.534895896911621,
|
|
"epoch": 0.02121044832495569,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0599,
|
|
"mean_token_accuracy": 0.07970957532525062,
|
|
"num_tokens": 702951.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.588031339645386,
|
|
"epoch": 0.02150100241159892,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0677,
|
|
"mean_token_accuracy": 0.08218754455447197,
|
|
"num_tokens": 712481.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.600922870635986,
|
|
"epoch": 0.02179155649824215,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0683,
|
|
"mean_token_accuracy": 0.08380770459771156,
|
|
"num_tokens": 721579.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.572713327407837,
|
|
"epoch": 0.022082110584885375,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0774,
|
|
"mean_token_accuracy": 0.07982454895973205,
|
|
"num_tokens": 731404.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.548839807510376,
|
|
"epoch": 0.022372664671528604,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.0556,
|
|
"mean_token_accuracy": 0.07496214136481286,
|
|
"num_tokens": 740751.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.523876476287842,
|
|
"epoch": 0.022663218758171833,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0247,
|
|
"mean_token_accuracy": 0.08082472011446953,
|
|
"num_tokens": 751171.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.552808237075806,
|
|
"epoch": 0.022953772844815062,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0823,
|
|
"mean_token_accuracy": 0.07615064568817616,
|
|
"num_tokens": 760874.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.583486127853393,
|
|
"epoch": 0.02324432693145829,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.0585,
|
|
"mean_token_accuracy": 0.08326990716159344,
|
|
"num_tokens": 769652.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.488273334503174,
|
|
"epoch": 0.02353488101810152,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0421,
|
|
"mean_token_accuracy": 0.07620194889605045,
|
|
"num_tokens": 779591.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.564187002182007,
|
|
"epoch": 0.023825435104744747,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.156,
|
|
"mean_token_accuracy": 0.08098742663860321,
|
|
"num_tokens": 789582.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.506245565414429,
|
|
"epoch": 0.024115989191387976,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.0546,
|
|
"mean_token_accuracy": 0.07765479311347008,
|
|
"num_tokens": 799146.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.4926127910614015,
|
|
"epoch": 0.024406543278031205,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0246,
|
|
"mean_token_accuracy": 0.0782523088157177,
|
|
"num_tokens": 808934.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.532363748550415,
|
|
"epoch": 0.024697097364674434,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.0821,
|
|
"mean_token_accuracy": 0.07597277015447616,
|
|
"num_tokens": 819280.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.457432746887207,
|
|
"epoch": 0.024987651451317663,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.9892,
|
|
"mean_token_accuracy": 0.0840725652873516,
|
|
"num_tokens": 828818.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.463752698898316,
|
|
"epoch": 0.025278205537960893,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.9816,
|
|
"mean_token_accuracy": 0.08661384396255016,
|
|
"num_tokens": 839175.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.5449175357818605,
|
|
"epoch": 0.02556875962460412,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.0777,
|
|
"mean_token_accuracy": 0.07947314418852329,
|
|
"num_tokens": 849965.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.392349624633789,
|
|
"epoch": 0.025859313711247348,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9968,
|
|
"mean_token_accuracy": 0.08229465186595916,
|
|
"num_tokens": 859229.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.4397971630096436,
|
|
"epoch": 0.026149867797890577,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9708,
|
|
"mean_token_accuracy": 0.0816520519554615,
|
|
"num_tokens": 869199.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.399962043762207,
|
|
"epoch": 0.026440421884533806,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.9666,
|
|
"mean_token_accuracy": 0.09285714998841285,
|
|
"num_tokens": 879470.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.4366514682769775,
|
|
"epoch": 0.026730975971177035,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9274,
|
|
"mean_token_accuracy": 0.0792453158646822,
|
|
"num_tokens": 888397.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.370485734939575,
|
|
"epoch": 0.027021530057820264,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8202,
|
|
"mean_token_accuracy": 0.0855403620749712,
|
|
"num_tokens": 898321.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.4845947265625,
|
|
"epoch": 0.027312084144463494,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.0856,
|
|
"mean_token_accuracy": 0.0808610200881958,
|
|
"num_tokens": 907947.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.327203702926636,
|
|
"epoch": 0.02760263823110672,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9103,
|
|
"mean_token_accuracy": 0.0954340323805809,
|
|
"num_tokens": 916842.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.380954456329346,
|
|
"epoch": 0.02789319231774995,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.0098,
|
|
"mean_token_accuracy": 0.08165798112750053,
|
|
"num_tokens": 926431.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.412681722640992,
|
|
"epoch": 0.028183746404393178,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.9162,
|
|
"mean_token_accuracy": 0.08133741281926632,
|
|
"num_tokens": 935819.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.44426212310791,
|
|
"epoch": 0.028474300491036407,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.9418,
|
|
"mean_token_accuracy": 0.08402741849422454,
|
|
"num_tokens": 944198.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.264917373657227,
|
|
"epoch": 0.028764854577679636,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.9628,
|
|
"mean_token_accuracy": 0.08352083042263984,
|
|
"num_tokens": 954972.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.385922384262085,
|
|
"epoch": 0.029055408664322865,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.9018,
|
|
"mean_token_accuracy": 0.08520250022411346,
|
|
"num_tokens": 964532.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.475071048736572,
|
|
"epoch": 0.02934596275096609,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.9955,
|
|
"mean_token_accuracy": 0.07958225682377815,
|
|
"num_tokens": 974547.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.299204540252686,
|
|
"epoch": 0.02963651683760932,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.935,
|
|
"mean_token_accuracy": 0.08022963926196099,
|
|
"num_tokens": 984245.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.318370199203491,
|
|
"epoch": 0.02992707092425255,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7766,
|
|
"mean_token_accuracy": 0.08500204458832741,
|
|
"num_tokens": 994400.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.352757215499878,
|
|
"epoch": 0.03021762501089578,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.0024,
|
|
"mean_token_accuracy": 0.07765024341642857,
|
|
"num_tokens": 1005775.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.312537145614624,
|
|
"epoch": 0.030508179097539008,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.9055,
|
|
"mean_token_accuracy": 0.08693855553865433,
|
|
"num_tokens": 1015386.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.383286190032959,
|
|
"epoch": 0.030798733184182237,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8994,
|
|
"mean_token_accuracy": 0.09188547134399414,
|
|
"num_tokens": 1024963.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.249363946914673,
|
|
"epoch": 0.031089287270825463,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8996,
|
|
"mean_token_accuracy": 0.08531768508255481,
|
|
"num_tokens": 1034667.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.265355777740479,
|
|
"epoch": 0.03137984135746869,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8796,
|
|
"mean_token_accuracy": 0.08795020580291749,
|
|
"num_tokens": 1044171.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.295146417617798,
|
|
"epoch": 0.031670395444111925,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8538,
|
|
"mean_token_accuracy": 0.08691519349813462,
|
|
"num_tokens": 1053585.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.237406063079834,
|
|
"epoch": 0.03196094953075515,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7515,
|
|
"mean_token_accuracy": 0.09050033241510391,
|
|
"num_tokens": 1063310.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.263738679885864,
|
|
"epoch": 0.032251503617398376,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.8651,
|
|
"mean_token_accuracy": 0.08824861124157905,
|
|
"num_tokens": 1073529.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.175330972671508,
|
|
"epoch": 0.03254205770404161,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8319,
|
|
"mean_token_accuracy": 0.08951647505164147,
|
|
"num_tokens": 1083432.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.184946346282959,
|
|
"epoch": 0.032832611790684835,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.8004,
|
|
"mean_token_accuracy": 0.09656240493059158,
|
|
"num_tokens": 1092453.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.274725437164307,
|
|
"epoch": 0.03312316587732807,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.8865,
|
|
"mean_token_accuracy": 0.08661114051938057,
|
|
"num_tokens": 1102402.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.303795433044433,
|
|
"epoch": 0.03341371996397129,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8928,
|
|
"mean_token_accuracy": 0.09610759019851685,
|
|
"num_tokens": 1111907.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.228280067443848,
|
|
"epoch": 0.03370427405061452,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7846,
|
|
"mean_token_accuracy": 0.09133462607860565,
|
|
"num_tokens": 1120712.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.0720751762390135,
|
|
"epoch": 0.03399482813725775,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6691,
|
|
"mean_token_accuracy": 0.0894063800573349,
|
|
"num_tokens": 1131165.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.229758644104004,
|
|
"epoch": 0.03428538222390098,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8337,
|
|
"mean_token_accuracy": 0.08700250834226608,
|
|
"num_tokens": 1140527.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.137591791152954,
|
|
"epoch": 0.03457593631054421,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.792,
|
|
"mean_token_accuracy": 0.08842456936836243,
|
|
"num_tokens": 1149977.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.240325021743774,
|
|
"epoch": 0.034866490397187436,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.8153,
|
|
"mean_token_accuracy": 0.08972005397081376,
|
|
"num_tokens": 1159918.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.116828918457031,
|
|
"epoch": 0.03515704448383067,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.7965,
|
|
"mean_token_accuracy": 0.08587550893425941,
|
|
"num_tokens": 1169218.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.1641600131988525,
|
|
"epoch": 0.035447598570473894,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.8058,
|
|
"mean_token_accuracy": 0.09056585654616356,
|
|
"num_tokens": 1179429.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.0538177490234375,
|
|
"epoch": 0.03573815265711712,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7051,
|
|
"mean_token_accuracy": 0.0951805867254734,
|
|
"num_tokens": 1189379.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.165834856033325,
|
|
"epoch": 0.03602870674376035,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6834,
|
|
"mean_token_accuracy": 0.09452618882060052,
|
|
"num_tokens": 1198643.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.1435986995697025,
|
|
"epoch": 0.03631926083040358,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.8985,
|
|
"mean_token_accuracy": 0.08901753202080727,
|
|
"num_tokens": 1207933.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.125590705871582,
|
|
"epoch": 0.03660981491704681,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7771,
|
|
"mean_token_accuracy": 0.09473630785942078,
|
|
"num_tokens": 1217000.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.342123746871948,
|
|
"epoch": 0.03690036900369004,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.8715,
|
|
"mean_token_accuracy": 0.08738602064549923,
|
|
"num_tokens": 1227054.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.0751423835754395,
|
|
"epoch": 0.03719092309033326,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.8639,
|
|
"mean_token_accuracy": 0.08903967961668968,
|
|
"num_tokens": 1237126.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.132748985290528,
|
|
"epoch": 0.037481477176976495,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7309,
|
|
"mean_token_accuracy": 0.09907565861940384,
|
|
"num_tokens": 1247404.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.105540752410889,
|
|
"epoch": 0.03777203126361972,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.6672,
|
|
"mean_token_accuracy": 0.08641588017344475,
|
|
"num_tokens": 1257130.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.073269605636597,
|
|
"epoch": 0.038062585350262954,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.7423,
|
|
"mean_token_accuracy": 0.09811322540044784,
|
|
"num_tokens": 1266931.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.157707405090332,
|
|
"epoch": 0.03835313943690618,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.7753,
|
|
"mean_token_accuracy": 0.08842945359647274,
|
|
"num_tokens": 1277770.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.074891519546509,
|
|
"epoch": 0.03864369352354941,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.6966,
|
|
"mean_token_accuracy": 0.09733218997716904,
|
|
"num_tokens": 1287188.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.035866546630859,
|
|
"epoch": 0.03893424761019264,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.7408,
|
|
"mean_token_accuracy": 0.09134816229343415,
|
|
"num_tokens": 1297038.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.091120624542237,
|
|
"epoch": 0.03922480169683586,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6964,
|
|
"mean_token_accuracy": 0.09473009631037713,
|
|
"num_tokens": 1306860.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.030598735809326,
|
|
"epoch": 0.039515355783479096,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.6668,
|
|
"mean_token_accuracy": 0.09435953348875045,
|
|
"num_tokens": 1316585.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.1326805591583256,
|
|
"epoch": 0.03980590987012232,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.7282,
|
|
"mean_token_accuracy": 0.09551571607589722,
|
|
"num_tokens": 1325601.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.101321458816528,
|
|
"epoch": 0.040096463956765555,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7604,
|
|
"mean_token_accuracy": 0.09247554913163185,
|
|
"num_tokens": 1336305.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.1049731254577635,
|
|
"epoch": 0.04038701804340878,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6507,
|
|
"mean_token_accuracy": 0.09341847449541092,
|
|
"num_tokens": 1344820.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.997063255310058,
|
|
"epoch": 0.040677572130052006,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6331,
|
|
"mean_token_accuracy": 0.09355669766664505,
|
|
"num_tokens": 1353950.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.01454758644104,
|
|
"epoch": 0.04096812621669524,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.7545,
|
|
"mean_token_accuracy": 0.09254956245422363,
|
|
"num_tokens": 1364881.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.0095212936401365,
|
|
"epoch": 0.041258680303338464,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.7061,
|
|
"mean_token_accuracy": 0.09260506108403206,
|
|
"num_tokens": 1374018.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.11537013053894,
|
|
"epoch": 0.0415492343899817,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.6946,
|
|
"mean_token_accuracy": 0.08821133449673653,
|
|
"num_tokens": 1384319.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.958690166473389,
|
|
"epoch": 0.04183978847662492,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5713,
|
|
"mean_token_accuracy": 0.09440450817346573,
|
|
"num_tokens": 1393753.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.922836446762085,
|
|
"epoch": 0.042130342563268156,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6616,
|
|
"mean_token_accuracy": 0.09427325800061226,
|
|
"num_tokens": 1403599.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.020907402038574,
|
|
"epoch": 0.04242089664991138,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6611,
|
|
"mean_token_accuracy": 0.10043973848223686,
|
|
"num_tokens": 1412508.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.071925306320191,
|
|
"epoch": 0.04271145073655461,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.8015,
|
|
"mean_token_accuracy": 0.0910523734986782,
|
|
"num_tokens": 1422776.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.998428392410278,
|
|
"epoch": 0.04300200482319784,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.6414,
|
|
"mean_token_accuracy": 0.09633751660585403,
|
|
"num_tokens": 1432901.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.035877513885498,
|
|
"epoch": 0.043292558909841065,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.677,
|
|
"mean_token_accuracy": 0.09542910531163215,
|
|
"num_tokens": 1442916.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.878139925003052,
|
|
"epoch": 0.0435831129964843,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5395,
|
|
"mean_token_accuracy": 0.09616116657853127,
|
|
"num_tokens": 1453037.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.96289029121399,
|
|
"epoch": 0.043873667083127524,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6196,
|
|
"mean_token_accuracy": 0.10786209627985954,
|
|
"num_tokens": 1461963.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.00122447013855,
|
|
"epoch": 0.04416422116977075,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.7012,
|
|
"mean_token_accuracy": 0.09169812574982643,
|
|
"num_tokens": 1471521.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.930304098129272,
|
|
"epoch": 0.04445477525641398,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5366,
|
|
"mean_token_accuracy": 0.0987947553396225,
|
|
"num_tokens": 1481438.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.89730920791626,
|
|
"epoch": 0.04474532934305721,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5654,
|
|
"mean_token_accuracy": 0.09912522435188294,
|
|
"num_tokens": 1490522.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.994078540802002,
|
|
"epoch": 0.04503588342970044,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.7343,
|
|
"mean_token_accuracy": 0.09250347167253495,
|
|
"num_tokens": 1501034.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.894172525405883,
|
|
"epoch": 0.045326437516343666,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5391,
|
|
"mean_token_accuracy": 0.10528326034545898,
|
|
"num_tokens": 1510390.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.992980337142944,
|
|
"epoch": 0.0456169916029869,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.6468,
|
|
"mean_token_accuracy": 0.09232402816414834,
|
|
"num_tokens": 1520048.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.977211618423462,
|
|
"epoch": 0.045907545689630125,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5275,
|
|
"mean_token_accuracy": 0.10221462920308114,
|
|
"num_tokens": 1529113.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.760094785690308,
|
|
"epoch": 0.04619809977627335,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.6057,
|
|
"mean_token_accuracy": 0.09887640923261642,
|
|
"num_tokens": 1538573.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.975562715530396,
|
|
"epoch": 0.04648865386291658,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.6064,
|
|
"mean_token_accuracy": 0.10373581051826478,
|
|
"num_tokens": 1547471.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.8805656909942625,
|
|
"epoch": 0.04677920794955981,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5641,
|
|
"mean_token_accuracy": 0.10285315811634063,
|
|
"num_tokens": 1557259.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.063277673721314,
|
|
"epoch": 0.04706976203620304,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.7921,
|
|
"mean_token_accuracy": 0.09200607016682624,
|
|
"num_tokens": 1567383.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.87684121131897,
|
|
"epoch": 0.04736031612284627,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4826,
|
|
"mean_token_accuracy": 0.11064840331673623,
|
|
"num_tokens": 1577106.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.807673025131225,
|
|
"epoch": 0.04765087020948949,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5393,
|
|
"mean_token_accuracy": 0.10080247670412064,
|
|
"num_tokens": 1586100.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.877712535858154,
|
|
"epoch": 0.047941424296132726,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6279,
|
|
"mean_token_accuracy": 0.09564873427152634,
|
|
"num_tokens": 1596950.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.891899585723877,
|
|
"epoch": 0.04823197838277595,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5837,
|
|
"mean_token_accuracy": 0.09832958057522774,
|
|
"num_tokens": 1606001.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.978082180023193,
|
|
"epoch": 0.048522532469419184,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.6825,
|
|
"mean_token_accuracy": 0.0975476372987032,
|
|
"num_tokens": 1616498.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.831979036331177,
|
|
"epoch": 0.04881308655606241,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5199,
|
|
"mean_token_accuracy": 0.10347988307476044,
|
|
"num_tokens": 1625195.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.784482002258301,
|
|
"epoch": 0.04910364064270564,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4476,
|
|
"mean_token_accuracy": 0.10162880271673203,
|
|
"num_tokens": 1635176.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.806185960769653,
|
|
"epoch": 0.04939419472934887,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.553,
|
|
"mean_token_accuracy": 0.1015662670135498,
|
|
"num_tokens": 1645183.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.801709985733032,
|
|
"epoch": 0.049684748815992094,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5479,
|
|
"mean_token_accuracy": 0.10148834735155106,
|
|
"num_tokens": 1654226.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.834500074386597,
|
|
"epoch": 0.04997530290263533,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5426,
|
|
"mean_token_accuracy": 0.10362305790185929,
|
|
"num_tokens": 1664572.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.950858306884766,
|
|
"epoch": 0.05026585698927855,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.6472,
|
|
"mean_token_accuracy": 0.09981537386775016,
|
|
"num_tokens": 1674070.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.791647720336914,
|
|
"epoch": 0.050556411075921785,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4773,
|
|
"mean_token_accuracy": 0.09943379536271095,
|
|
"num_tokens": 1683473.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.777591514587402,
|
|
"epoch": 0.05084696516256501,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4869,
|
|
"mean_token_accuracy": 0.10118941962718964,
|
|
"num_tokens": 1693171.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.898639726638794,
|
|
"epoch": 0.05113751924920824,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.606,
|
|
"mean_token_accuracy": 0.09705074802041054,
|
|
"num_tokens": 1703023.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.73418025970459,
|
|
"epoch": 0.05142807333585147,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4984,
|
|
"mean_token_accuracy": 0.1019330695271492,
|
|
"num_tokens": 1712698.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.906363248825073,
|
|
"epoch": 0.051718627422494695,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.6098,
|
|
"mean_token_accuracy": 0.09838435426354408,
|
|
"num_tokens": 1721502.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.7474723815917965,
|
|
"epoch": 0.05200918150913793,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4942,
|
|
"mean_token_accuracy": 0.10594057068228721,
|
|
"num_tokens": 1730551.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.808920383453369,
|
|
"epoch": 0.052299735595781154,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5645,
|
|
"mean_token_accuracy": 0.10622440055012702,
|
|
"num_tokens": 1739368.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.827513933181763,
|
|
"epoch": 0.052590289682424386,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4176,
|
|
"mean_token_accuracy": 0.11146403327584267,
|
|
"num_tokens": 1748528.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.713736248016358,
|
|
"epoch": 0.05288084376906761,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5739,
|
|
"mean_token_accuracy": 0.09899114519357681,
|
|
"num_tokens": 1759569.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.80773286819458,
|
|
"epoch": 0.05317139785571084,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.5099,
|
|
"mean_token_accuracy": 0.10788461863994599,
|
|
"num_tokens": 1769366.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.76817569732666,
|
|
"epoch": 0.05346195194235407,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.6024,
|
|
"mean_token_accuracy": 0.09936894476413727,
|
|
"num_tokens": 1780155.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.755830335617065,
|
|
"epoch": 0.053752506028997296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4233,
|
|
"mean_token_accuracy": 0.10512633025646209,
|
|
"num_tokens": 1789436.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.823408889770508,
|
|
"epoch": 0.05404306011564053,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.5652,
|
|
"mean_token_accuracy": 0.0998048096895218,
|
|
"num_tokens": 1798836.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.751146364212036,
|
|
"epoch": 0.054333614202283755,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.444,
|
|
"mean_token_accuracy": 0.10532717406749725,
|
|
"num_tokens": 1808666.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.8108867645263675,
|
|
"epoch": 0.05462416828892699,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.5972,
|
|
"mean_token_accuracy": 0.09496863186359406,
|
|
"num_tokens": 1820001.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.751294231414795,
|
|
"epoch": 0.05491472237557021,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4693,
|
|
"mean_token_accuracy": 0.10566612035036087,
|
|
"num_tokens": 1830284.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.820448493957519,
|
|
"epoch": 0.05520527646221344,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4794,
|
|
"mean_token_accuracy": 0.10577797368168831,
|
|
"num_tokens": 1839930.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.629036235809326,
|
|
"epoch": 0.05549583054885667,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5675,
|
|
"mean_token_accuracy": 0.10090194195508957,
|
|
"num_tokens": 1850697.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.817226839065552,
|
|
"epoch": 0.0557863846354999,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.497,
|
|
"mean_token_accuracy": 0.10740380734205246,
|
|
"num_tokens": 1860196.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.774875259399414,
|
|
"epoch": 0.05607693872214313,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.47,
|
|
"mean_token_accuracy": 0.1075842596590519,
|
|
"num_tokens": 1869000.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.722468996047974,
|
|
"epoch": 0.056367492808786356,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.469,
|
|
"mean_token_accuracy": 0.10600791200995445,
|
|
"num_tokens": 1878687.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.728367662429809,
|
|
"epoch": 0.05665804689542958,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3467,
|
|
"mean_token_accuracy": 0.10569515079259872,
|
|
"num_tokens": 1886914.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.671978425979614,
|
|
"epoch": 0.056948600982072814,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5321,
|
|
"mean_token_accuracy": 0.10422437414526939,
|
|
"num_tokens": 1897392.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.805356025695801,
|
|
"epoch": 0.05723915506871604,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.488,
|
|
"mean_token_accuracy": 0.10600305423140526,
|
|
"num_tokens": 1906215.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.8313037872314455,
|
|
"epoch": 0.05752970915535927,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.5017,
|
|
"mean_token_accuracy": 0.10730748698115349,
|
|
"num_tokens": 1915376.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.659111022949219,
|
|
"epoch": 0.0578202632420025,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.465,
|
|
"mean_token_accuracy": 0.10440039038658142,
|
|
"num_tokens": 1925558.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.676358318328857,
|
|
"epoch": 0.05811081732864573,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.4301,
|
|
"mean_token_accuracy": 0.10430914014577866,
|
|
"num_tokens": 1935176.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.770152616500854,
|
|
"epoch": 0.05840137141528896,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.4924,
|
|
"mean_token_accuracy": 0.10445862039923667,
|
|
"num_tokens": 1945135.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.597527885437012,
|
|
"epoch": 0.05869192550193218,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.4838,
|
|
"mean_token_accuracy": 0.10765932872891426,
|
|
"num_tokens": 1955585.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.867468976974488,
|
|
"epoch": 0.058982479588575415,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.4687,
|
|
"mean_token_accuracy": 0.10314074084162712,
|
|
"num_tokens": 1964750.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.624800300598144,
|
|
"epoch": 0.05927303367521864,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.5436,
|
|
"mean_token_accuracy": 0.10335941463708878,
|
|
"num_tokens": 1974697.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.7362236976623535,
|
|
"epoch": 0.05956358776186187,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.4934,
|
|
"mean_token_accuracy": 0.10554013177752494,
|
|
"num_tokens": 1983384.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.754078722000122,
|
|
"epoch": 0.0598541418485051,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.5229,
|
|
"mean_token_accuracy": 0.11017107889056206,
|
|
"num_tokens": 1993790.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.697407197952271,
|
|
"epoch": 0.060144695935148325,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.4027,
|
|
"mean_token_accuracy": 0.10623413920402527,
|
|
"num_tokens": 2003296.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.585323095321655,
|
|
"epoch": 0.06043525002179156,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.11027837991714477,
|
|
"num_tokens": 2012481.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.705205965042114,
|
|
"epoch": 0.06072580410843478,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.4994,
|
|
"mean_token_accuracy": 0.10490612536668778,
|
|
"num_tokens": 2022382.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.638956928253174,
|
|
"epoch": 0.061016358195078016,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.4097,
|
|
"mean_token_accuracy": 0.11042128577828407,
|
|
"num_tokens": 2032290.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.6624797821044925,
|
|
"epoch": 0.06130691228172124,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.3594,
|
|
"mean_token_accuracy": 0.10972521901130676,
|
|
"num_tokens": 2041094.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.609392881393433,
|
|
"epoch": 0.061597466368364474,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.5125,
|
|
"mean_token_accuracy": 0.10544388592243195,
|
|
"num_tokens": 2051169.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.6946526050567625,
|
|
"epoch": 0.0618880204550077,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.3966,
|
|
"mean_token_accuracy": 0.1103939101099968,
|
|
"num_tokens": 2061310.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.619559907913208,
|
|
"epoch": 0.062178574541650926,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.3829,
|
|
"mean_token_accuracy": 0.10736953839659691,
|
|
"num_tokens": 2071537.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.6101906299591064,
|
|
"epoch": 0.06246912862829416,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.4507,
|
|
"mean_token_accuracy": 0.10780780464410782,
|
|
"num_tokens": 2082540.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.621304225921631,
|
|
"epoch": 0.06275968271493738,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.4413,
|
|
"mean_token_accuracy": 0.11023736447095871,
|
|
"num_tokens": 2091313.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.691177225112915,
|
|
"epoch": 0.06305023680158062,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.4862,
|
|
"mean_token_accuracy": 0.10716225057840348,
|
|
"num_tokens": 2100826.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.671515083312988,
|
|
"epoch": 0.06334079088822385,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.4177,
|
|
"mean_token_accuracy": 0.10876795202493668,
|
|
"num_tokens": 2109798.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.640522909164429,
|
|
"epoch": 0.06363134497486707,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.3138,
|
|
"mean_token_accuracy": 0.11188038140535354,
|
|
"num_tokens": 2119641.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.565330696105957,
|
|
"epoch": 0.0639218990615103,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.3525,
|
|
"mean_token_accuracy": 0.11164129376411439,
|
|
"num_tokens": 2128724.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.521946573257447,
|
|
"epoch": 0.06421245314815353,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.2737,
|
|
"mean_token_accuracy": 0.11544388085603714,
|
|
"num_tokens": 2137251.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.6156073093414305,
|
|
"epoch": 0.06450300723479675,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.3821,
|
|
"mean_token_accuracy": 0.118138437718153,
|
|
"num_tokens": 2146978.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.669202089309692,
|
|
"epoch": 0.06479356132143999,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.4636,
|
|
"mean_token_accuracy": 0.10260412320494652,
|
|
"num_tokens": 2157153.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.475356149673462,
|
|
"epoch": 0.06508411540808322,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.3599,
|
|
"mean_token_accuracy": 0.11348235085606576,
|
|
"num_tokens": 2166946.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.657857656478882,
|
|
"epoch": 0.06537466949472644,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.3772,
|
|
"mean_token_accuracy": 0.11038358807563782,
|
|
"num_tokens": 2177249.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.466132879257202,
|
|
"epoch": 0.06566522358136967,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.3098,
|
|
"mean_token_accuracy": 0.11056527942419052,
|
|
"num_tokens": 2187240.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.6453643321990965,
|
|
"epoch": 0.0659557776680129,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.3145,
|
|
"mean_token_accuracy": 0.11672020331025124,
|
|
"num_tokens": 2196199.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.595391893386841,
|
|
"epoch": 0.06624633175465613,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.2689,
|
|
"mean_token_accuracy": 0.11631305515766144,
|
|
"num_tokens": 2205726.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.587292861938477,
|
|
"epoch": 0.06653688584129935,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.4457,
|
|
"mean_token_accuracy": 0.10664665251970291,
|
|
"num_tokens": 2216552.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.597306776046753,
|
|
"epoch": 0.06682743992794259,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.3034,
|
|
"mean_token_accuracy": 0.11342373788356781,
|
|
"num_tokens": 2225935.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.644480466842651,
|
|
"epoch": 0.06711799401458582,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.4604,
|
|
"mean_token_accuracy": 0.11499964445829391,
|
|
"num_tokens": 2235907.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.599408388137817,
|
|
"epoch": 0.06740854810122904,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.3881,
|
|
"mean_token_accuracy": 0.10959212481975555,
|
|
"num_tokens": 2244473.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.641416931152344,
|
|
"epoch": 0.06769910218787227,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.4123,
|
|
"mean_token_accuracy": 0.1027280792593956,
|
|
"num_tokens": 2253834.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.5416028022766115,
|
|
"epoch": 0.0679896562745155,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.3413,
|
|
"mean_token_accuracy": 0.11207354813814163,
|
|
"num_tokens": 2263243.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.6198502540588375,
|
|
"epoch": 0.06828021036115874,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.3035,
|
|
"mean_token_accuracy": 0.11463942378759384,
|
|
"num_tokens": 2273008.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.430229234695434,
|
|
"epoch": 0.06857076444780195,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.3457,
|
|
"mean_token_accuracy": 0.1155870608985424,
|
|
"num_tokens": 2282522.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.601986408233643,
|
|
"epoch": 0.06886131853444519,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.2529,
|
|
"mean_token_accuracy": 0.12027783617377281,
|
|
"num_tokens": 2292114.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.400939083099365,
|
|
"epoch": 0.06915187262108842,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.1978,
|
|
"mean_token_accuracy": 0.12538810446858406,
|
|
"num_tokens": 2301592.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.51713194847107,
|
|
"epoch": 0.06944242670773164,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.3201,
|
|
"mean_token_accuracy": 0.11454231590032578,
|
|
"num_tokens": 2310940.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.463452672958374,
|
|
"epoch": 0.06973298079437487,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.1151392012834549,
|
|
"num_tokens": 2320535.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.596390962600708,
|
|
"epoch": 0.0700235348810181,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.2563,
|
|
"mean_token_accuracy": 0.12523823976516724,
|
|
"num_tokens": 2330431.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.3869446277618405,
|
|
"epoch": 0.07031408896766134,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.2036,
|
|
"mean_token_accuracy": 0.12079060897231102,
|
|
"num_tokens": 2339323.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.510322713851929,
|
|
"epoch": 0.07060464305430456,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.2992,
|
|
"mean_token_accuracy": 0.11487918049097061,
|
|
"num_tokens": 2349750.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.503530073165893,
|
|
"epoch": 0.07089519714094779,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.2558,
|
|
"mean_token_accuracy": 0.1210679478943348,
|
|
"num_tokens": 2359147.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.520279359817505,
|
|
"epoch": 0.07118575122759102,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.2786,
|
|
"mean_token_accuracy": 0.11829182729125023,
|
|
"num_tokens": 2368762.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.391372203826904,
|
|
"epoch": 0.07147630531423424,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.1568,
|
|
"mean_token_accuracy": 0.12013374790549278,
|
|
"num_tokens": 2378060.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.465664291381836,
|
|
"epoch": 0.07176685940087747,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.2575,
|
|
"mean_token_accuracy": 0.12200002744793892,
|
|
"num_tokens": 2386997.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.6493157863616945,
|
|
"epoch": 0.0720574134875207,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.4026,
|
|
"mean_token_accuracy": 0.11086667999625206,
|
|
"num_tokens": 2396980.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.435001850128174,
|
|
"epoch": 0.07234796757416392,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.2556,
|
|
"mean_token_accuracy": 0.1203616626560688,
|
|
"num_tokens": 2406450.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.519892168045044,
|
|
"epoch": 0.07263852166080716,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.2632,
|
|
"mean_token_accuracy": 0.11274134442210197,
|
|
"num_tokens": 2415985.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.457003879547119,
|
|
"epoch": 0.07292907574745039,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.2928,
|
|
"mean_token_accuracy": 0.11486212983727455,
|
|
"num_tokens": 2425676.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.508018493652344,
|
|
"epoch": 0.07321962983409362,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.2922,
|
|
"mean_token_accuracy": 0.10985862240195274,
|
|
"num_tokens": 2435918.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.5097509860992435,
|
|
"epoch": 0.07351018392073684,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.3562,
|
|
"mean_token_accuracy": 0.11788229197263718,
|
|
"num_tokens": 2445267.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.599736261367798,
|
|
"epoch": 0.07380073800738007,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.3224,
|
|
"mean_token_accuracy": 0.11191006749868393,
|
|
"num_tokens": 2455885.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.489064168930054,
|
|
"epoch": 0.0740912920940233,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.3483,
|
|
"mean_token_accuracy": 0.11579938605427742,
|
|
"num_tokens": 2465604.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.5532605171203615,
|
|
"epoch": 0.07438184618066652,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.4458,
|
|
"mean_token_accuracy": 0.10784725919365883,
|
|
"num_tokens": 2475168.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.473039054870606,
|
|
"epoch": 0.07467240026730976,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.2243,
|
|
"mean_token_accuracy": 0.12185250818729401,
|
|
"num_tokens": 2484185.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.583461809158325,
|
|
"epoch": 0.07496295435395299,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.3387,
|
|
"mean_token_accuracy": 0.11904568299651146,
|
|
"num_tokens": 2492992.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.418948078155518,
|
|
"epoch": 0.07525350844059622,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.2065,
|
|
"mean_token_accuracy": 0.119369375705719,
|
|
"num_tokens": 2502541.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.468045377731324,
|
|
"epoch": 0.07554406252723944,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.2364,
|
|
"mean_token_accuracy": 0.11862708181142807,
|
|
"num_tokens": 2511947.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.385909509658814,
|
|
"epoch": 0.07583461661388267,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.2525,
|
|
"mean_token_accuracy": 0.12124920263886452,
|
|
"num_tokens": 2521727.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.531244993209839,
|
|
"epoch": 0.07612517070052591,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.2694,
|
|
"mean_token_accuracy": 0.1218131199479103,
|
|
"num_tokens": 2530680.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.422513055801391,
|
|
"epoch": 0.07641572478716913,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.1542,
|
|
"mean_token_accuracy": 0.12266649156808854,
|
|
"num_tokens": 2539414.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.491461181640625,
|
|
"epoch": 0.07670627887381236,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.3072,
|
|
"mean_token_accuracy": 0.1141884945333004,
|
|
"num_tokens": 2549609.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.4648158073425295,
|
|
"epoch": 0.07699683296045559,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.2773,
|
|
"mean_token_accuracy": 0.11799687221646309,
|
|
"num_tokens": 2560081.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.400183534622192,
|
|
"epoch": 0.07728738704709882,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.2328,
|
|
"mean_token_accuracy": 0.11631238982081413,
|
|
"num_tokens": 2569995.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.37140007019043,
|
|
"epoch": 0.07757794113374204,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.1738,
|
|
"mean_token_accuracy": 0.1213558554649353,
|
|
"num_tokens": 2579502.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.476347970962524,
|
|
"epoch": 0.07786849522038528,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.2553,
|
|
"mean_token_accuracy": 0.12006450816988945,
|
|
"num_tokens": 2589515.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.434140920639038,
|
|
"epoch": 0.07815904930702851,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.1897,
|
|
"mean_token_accuracy": 0.12782623916864394,
|
|
"num_tokens": 2598822.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.443134021759033,
|
|
"epoch": 0.07844960339367173,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.315,
|
|
"mean_token_accuracy": 0.11421679928898812,
|
|
"num_tokens": 2609169.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.484804105758667,
|
|
"epoch": 0.07874015748031496,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.2716,
|
|
"mean_token_accuracy": 0.11644304916262627,
|
|
"num_tokens": 2618683.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.466926431655883,
|
|
"epoch": 0.07903071156695819,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.2371,
|
|
"mean_token_accuracy": 0.12091248780488968,
|
|
"num_tokens": 2627976.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.505730533599854,
|
|
"epoch": 0.07932126565360141,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.294,
|
|
"mean_token_accuracy": 0.11860666498541832,
|
|
"num_tokens": 2637235.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.393795537948608,
|
|
"epoch": 0.07961181974024464,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.2951,
|
|
"mean_token_accuracy": 0.11754858568310737,
|
|
"num_tokens": 2646698.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.505375099182129,
|
|
"epoch": 0.07990237382688788,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.2788,
|
|
"mean_token_accuracy": 0.12090856656432152,
|
|
"num_tokens": 2656357.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.434703159332275,
|
|
"epoch": 0.08019292791353111,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.1776,
|
|
"mean_token_accuracy": 0.11764631941914558,
|
|
"num_tokens": 2665561.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.437805318832398,
|
|
"epoch": 0.08048348200017433,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.1909,
|
|
"mean_token_accuracy": 0.12296778410673141,
|
|
"num_tokens": 2674600.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.368801641464233,
|
|
"epoch": 0.08077403608681756,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.1838,
|
|
"mean_token_accuracy": 0.12880179584026336,
|
|
"num_tokens": 2684944.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.461294555664063,
|
|
"epoch": 0.0810645901734608,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.1342,
|
|
"mean_token_accuracy": 0.12006727010011672,
|
|
"num_tokens": 2694467.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.442787504196167,
|
|
"epoch": 0.08135514426010401,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.2769,
|
|
"mean_token_accuracy": 0.11950750723481178,
|
|
"num_tokens": 2704499.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.378614997863769,
|
|
"epoch": 0.08164569834674724,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.1896,
|
|
"mean_token_accuracy": 0.11940810978412628,
|
|
"num_tokens": 2715424.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.41011266708374,
|
|
"epoch": 0.08193625243339048,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.1788,
|
|
"mean_token_accuracy": 0.12036006227135658,
|
|
"num_tokens": 2724805.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.3130451202392575,
|
|
"epoch": 0.08222680652003371,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.1481,
|
|
"mean_token_accuracy": 0.12863539010286332,
|
|
"num_tokens": 2734216.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.424062490463257,
|
|
"epoch": 0.08251736060667693,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.2572,
|
|
"mean_token_accuracy": 0.11476619765162469,
|
|
"num_tokens": 2744146.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.422879314422607,
|
|
"epoch": 0.08280791469332016,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.2419,
|
|
"mean_token_accuracy": 0.12102322354912758,
|
|
"num_tokens": 2754000.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.637474250793457,
|
|
"epoch": 0.0830984687799634,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.288,
|
|
"mean_token_accuracy": 0.11775907129049301,
|
|
"num_tokens": 2763185.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.372423696517944,
|
|
"epoch": 0.08338902286660661,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.2381,
|
|
"mean_token_accuracy": 0.11884959116578102,
|
|
"num_tokens": 2773324.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.485676908493042,
|
|
"epoch": 0.08367957695324985,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.2207,
|
|
"mean_token_accuracy": 0.11649533435702324,
|
|
"num_tokens": 2783184.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.396980142593383,
|
|
"epoch": 0.08397013103989308,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.2223,
|
|
"mean_token_accuracy": 0.11544240266084671,
|
|
"num_tokens": 2793278.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.366146802902222,
|
|
"epoch": 0.08426068512653631,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.1262,
|
|
"mean_token_accuracy": 0.12203074395656585,
|
|
"num_tokens": 2802164.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.457718706130981,
|
|
"epoch": 0.08455123921317953,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.1891,
|
|
"mean_token_accuracy": 0.11854342371225357,
|
|
"num_tokens": 2811537.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.381798458099365,
|
|
"epoch": 0.08484179329982276,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.2965,
|
|
"mean_token_accuracy": 0.11684540212154389,
|
|
"num_tokens": 2821861.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.412109518051148,
|
|
"epoch": 0.085132347386466,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.1718,
|
|
"mean_token_accuracy": 0.12374548763036727,
|
|
"num_tokens": 2831916.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.33392972946167,
|
|
"epoch": 0.08542290147310921,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.1604,
|
|
"mean_token_accuracy": 0.12683377638459206,
|
|
"num_tokens": 2840979.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.418259906768799,
|
|
"epoch": 0.08571345555975245,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.1782,
|
|
"mean_token_accuracy": 0.11932171955704689,
|
|
"num_tokens": 2851332.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.265405559539795,
|
|
"epoch": 0.08600400964639568,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.073,
|
|
"mean_token_accuracy": 0.12999156266450881,
|
|
"num_tokens": 2860134.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.458755302429199,
|
|
"epoch": 0.08629456373303891,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.2126,
|
|
"mean_token_accuracy": 0.11848914325237274,
|
|
"num_tokens": 2869218.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.345139837265014,
|
|
"epoch": 0.08658511781968213,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.1426,
|
|
"mean_token_accuracy": 0.12274593263864517,
|
|
"num_tokens": 2878529.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.426075124740601,
|
|
"epoch": 0.08687567190632536,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.1579,
|
|
"mean_token_accuracy": 0.11727055683732032,
|
|
"num_tokens": 2888149.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.3561450958251955,
|
|
"epoch": 0.0871662259929686,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.1347,
|
|
"mean_token_accuracy": 0.12332948073744773,
|
|
"num_tokens": 2897790.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.316312408447265,
|
|
"epoch": 0.08745678007961181,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.2274,
|
|
"mean_token_accuracy": 0.12053183913230896,
|
|
"num_tokens": 2908168.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.390694427490234,
|
|
"epoch": 0.08774733416625505,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.116,
|
|
"mean_token_accuracy": 0.12294506877660752,
|
|
"num_tokens": 2918734.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.296877431869507,
|
|
"epoch": 0.08803788825289828,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.0839,
|
|
"mean_token_accuracy": 0.1293163001537323,
|
|
"num_tokens": 2927599.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.2974005222320555,
|
|
"epoch": 0.0883284423395415,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.1189,
|
|
"mean_token_accuracy": 0.12587246671319008,
|
|
"num_tokens": 2936707.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.4475304126739506,
|
|
"epoch": 0.08861899642618473,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.187,
|
|
"mean_token_accuracy": 0.12100831568241119,
|
|
"num_tokens": 2945607.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.347147464752197,
|
|
"epoch": 0.08890955051282796,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.0989,
|
|
"mean_token_accuracy": 0.12438113316893577,
|
|
"num_tokens": 2954836.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.321421432495117,
|
|
"epoch": 0.0892001045994712,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.1061,
|
|
"mean_token_accuracy": 0.12466374784708023,
|
|
"num_tokens": 2964338.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.353213739395142,
|
|
"epoch": 0.08949065868611442,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.1634,
|
|
"mean_token_accuracy": 0.12038285210728646,
|
|
"num_tokens": 2973943.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.339908075332642,
|
|
"epoch": 0.08978121277275765,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.0839,
|
|
"mean_token_accuracy": 0.12774784490466118,
|
|
"num_tokens": 2983315.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.300918197631836,
|
|
"epoch": 0.09007176685940088,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.0278,
|
|
"mean_token_accuracy": 0.12856598794460297,
|
|
"num_tokens": 2992039.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.178817892074585,
|
|
"epoch": 0.0903623209460441,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.0594,
|
|
"mean_token_accuracy": 0.12656542137265206,
|
|
"num_tokens": 3001428.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.426393222808838,
|
|
"epoch": 0.09065287503268733,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.1823,
|
|
"mean_token_accuracy": 0.12130758315324783,
|
|
"num_tokens": 3012060.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.289572381973267,
|
|
"epoch": 0.09094342911933057,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.133,
|
|
"mean_token_accuracy": 0.11978519856929778,
|
|
"num_tokens": 3021371.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.355306005477905,
|
|
"epoch": 0.0912339832059738,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 6.1783,
|
|
"mean_token_accuracy": 0.12279156744480133,
|
|
"num_tokens": 3030934.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.265284681320191,
|
|
"epoch": 0.09152453729261702,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.1286,
|
|
"mean_token_accuracy": 0.12520743757486344,
|
|
"num_tokens": 3041182.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.356278705596924,
|
|
"epoch": 0.09181509137926025,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 6.0982,
|
|
"mean_token_accuracy": 0.12530012279748917,
|
|
"num_tokens": 3050086.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.332121706008911,
|
|
"epoch": 0.09210564546590348,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.2171,
|
|
"mean_token_accuracy": 0.12314857169985771,
|
|
"num_tokens": 3061403.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.35912766456604,
|
|
"epoch": 0.0923961995525467,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.1708,
|
|
"mean_token_accuracy": 0.12304715439677238,
|
|
"num_tokens": 3070942.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.291305208206177,
|
|
"epoch": 0.09268675363918993,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.0504,
|
|
"mean_token_accuracy": 0.12217177525162697,
|
|
"num_tokens": 3079526.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.269596576690674,
|
|
"epoch": 0.09297730772583317,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 6.0384,
|
|
"mean_token_accuracy": 0.12933970913290976,
|
|
"num_tokens": 3090977.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.290025997161865,
|
|
"epoch": 0.0932678618124764,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 6.1565,
|
|
"mean_token_accuracy": 0.1253731794655323,
|
|
"num_tokens": 3100830.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.346334457397461,
|
|
"epoch": 0.09355841589911962,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.1792,
|
|
"mean_token_accuracy": 0.12612521946430205,
|
|
"num_tokens": 3110285.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.386321640014648,
|
|
"epoch": 0.09384896998576285,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.0621,
|
|
"mean_token_accuracy": 0.12808025181293486,
|
|
"num_tokens": 3119690.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.274943828582764,
|
|
"epoch": 0.09413952407240608,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.1145,
|
|
"mean_token_accuracy": 0.12201056703925132,
|
|
"num_tokens": 3129656.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.2934671401977536,
|
|
"epoch": 0.0944300781590493,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.0519,
|
|
"mean_token_accuracy": 0.12800228744745254,
|
|
"num_tokens": 3140083.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.261609125137329,
|
|
"epoch": 0.09472063224569253,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.0483,
|
|
"mean_token_accuracy": 0.13036949634552003,
|
|
"num_tokens": 3149747.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.2602826118469235,
|
|
"epoch": 0.09501118633233577,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.019,
|
|
"mean_token_accuracy": 0.12770563066005708,
|
|
"num_tokens": 3160367.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.249707126617432,
|
|
"epoch": 0.09530174041897899,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.0068,
|
|
"mean_token_accuracy": 0.12763984724879265,
|
|
"num_tokens": 3169375.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.236002111434937,
|
|
"epoch": 0.09559229450562222,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 6.0794,
|
|
"mean_token_accuracy": 0.13096466660499573,
|
|
"num_tokens": 3178907.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.366798305511475,
|
|
"epoch": 0.09588284859226545,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.0808,
|
|
"mean_token_accuracy": 0.12382574900984764,
|
|
"num_tokens": 3189336.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.227943420410156,
|
|
"epoch": 0.09617340267890868,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 6.1707,
|
|
"mean_token_accuracy": 0.12618450224399566,
|
|
"num_tokens": 3198787.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.291205787658692,
|
|
"epoch": 0.0964639567655519,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.0351,
|
|
"mean_token_accuracy": 0.1262456052005291,
|
|
"num_tokens": 3208161.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.287449550628662,
|
|
"epoch": 0.09675451085219514,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.095,
|
|
"mean_token_accuracy": 0.1295604422688484,
|
|
"num_tokens": 3217589.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.171545219421387,
|
|
"epoch": 0.09704506493883837,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.0335,
|
|
"mean_token_accuracy": 0.12443587705492973,
|
|
"num_tokens": 3227159.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.260297155380249,
|
|
"epoch": 0.09733561902548159,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.0323,
|
|
"mean_token_accuracy": 0.13212064653635025,
|
|
"num_tokens": 3236765.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.34990553855896,
|
|
"epoch": 0.09762617311212482,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 6.1196,
|
|
"mean_token_accuracy": 0.12271321415901185,
|
|
"num_tokens": 3247136.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.242118835449219,
|
|
"epoch": 0.09791672719876805,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 6.1049,
|
|
"mean_token_accuracy": 0.12386861220002174,
|
|
"num_tokens": 3256880.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.314453554153443,
|
|
"epoch": 0.09820728128541129,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.138,
|
|
"mean_token_accuracy": 0.12304992526769638,
|
|
"num_tokens": 3267104.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.257071495056152,
|
|
"epoch": 0.0984978353720545,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.0882,
|
|
"mean_token_accuracy": 0.12182446792721749,
|
|
"num_tokens": 3276165.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.447886228561401,
|
|
"epoch": 0.09878838945869774,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.2112,
|
|
"mean_token_accuracy": 0.12100318372249604,
|
|
"num_tokens": 3285600.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.263007783889771,
|
|
"epoch": 0.09907894354534097,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.1135,
|
|
"mean_token_accuracy": 0.12620161846280098,
|
|
"num_tokens": 3295654.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.253702402114868,
|
|
"epoch": 0.09936949763198419,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 6.0644,
|
|
"mean_token_accuracy": 0.134315574914217,
|
|
"num_tokens": 3304743.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.289787006378174,
|
|
"epoch": 0.09966005171862742,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.0324,
|
|
"mean_token_accuracy": 0.12361097186803818,
|
|
"num_tokens": 3313951.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.332495594024659,
|
|
"epoch": 0.09995060580527065,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 6.0995,
|
|
"mean_token_accuracy": 0.1251884751021862,
|
|
"num_tokens": 3324508.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.194294214248657,
|
|
"epoch": 0.10024115989191389,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 6.054,
|
|
"mean_token_accuracy": 0.12405704930424691,
|
|
"num_tokens": 3335373.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.333467721939087,
|
|
"epoch": 0.1005317139785571,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.0678,
|
|
"mean_token_accuracy": 0.12813965305685998,
|
|
"num_tokens": 3345242.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.268236112594605,
|
|
"epoch": 0.10082226806520034,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.0504,
|
|
"mean_token_accuracy": 0.13582077920436858,
|
|
"num_tokens": 3354783.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.266420888900757,
|
|
"epoch": 0.10111282215184357,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.0742,
|
|
"mean_token_accuracy": 0.13280235901474952,
|
|
"num_tokens": 3363705.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.345703363418579,
|
|
"epoch": 0.10140337623848679,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 6.1009,
|
|
"mean_token_accuracy": 0.12972408011555672,
|
|
"num_tokens": 3372331.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.221278953552246,
|
|
"epoch": 0.10169393032513002,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.0122,
|
|
"mean_token_accuracy": 0.13363172858953476,
|
|
"num_tokens": 3380986.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.314662122726441,
|
|
"epoch": 0.10198448441177325,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.0681,
|
|
"mean_token_accuracy": 0.12935666590929032,
|
|
"num_tokens": 3390852.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.2016339778900145,
|
|
"epoch": 0.10227503849841647,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.0835,
|
|
"mean_token_accuracy": 0.1321997858583927,
|
|
"num_tokens": 3399916.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.3428630352020265,
|
|
"epoch": 0.1025655925850597,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.0995,
|
|
"mean_token_accuracy": 0.13536889478564262,
|
|
"num_tokens": 3409913.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.2461179256439205,
|
|
"epoch": 0.10285614667170294,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.1217,
|
|
"mean_token_accuracy": 0.12532801926136017,
|
|
"num_tokens": 3420817.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.281768560409546,
|
|
"epoch": 0.10314670075834617,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.0642,
|
|
"mean_token_accuracy": 0.13035471364855766,
|
|
"num_tokens": 3430873.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.25596866607666,
|
|
"epoch": 0.10343725484498939,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 5.9967,
|
|
"mean_token_accuracy": 0.13541611135005951,
|
|
"num_tokens": 3440078.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.170544290542603,
|
|
"epoch": 0.10372780893163262,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 5.9563,
|
|
"mean_token_accuracy": 0.1369057409465313,
|
|
"num_tokens": 3449029.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.206426477432251,
|
|
"epoch": 0.10401836301827586,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.0783,
|
|
"mean_token_accuracy": 0.13255516290664673,
|
|
"num_tokens": 3458880.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.415725946426392,
|
|
"epoch": 0.10430891710491907,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 6.1575,
|
|
"mean_token_accuracy": 0.1243210181593895,
|
|
"num_tokens": 3468132.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.199592542648316,
|
|
"epoch": 0.10459947119156231,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.0499,
|
|
"mean_token_accuracy": 0.1273614466190338,
|
|
"num_tokens": 3477429.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.205861282348633,
|
|
"epoch": 0.10489002527820554,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.0663,
|
|
"mean_token_accuracy": 0.12217539176344872,
|
|
"num_tokens": 3487526.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.246215867996216,
|
|
"epoch": 0.10518057936484877,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 6.0049,
|
|
"mean_token_accuracy": 0.14162934869527816,
|
|
"num_tokens": 3496280.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.257954835891724,
|
|
"epoch": 0.10547113345149199,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 6.0072,
|
|
"mean_token_accuracy": 0.13223105296492577,
|
|
"num_tokens": 3505871.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.254146718978882,
|
|
"epoch": 0.10576168753813522,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 5.9791,
|
|
"mean_token_accuracy": 0.13909292891621589,
|
|
"num_tokens": 3515846.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.229612159729004,
|
|
"epoch": 0.10605224162477846,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 6.0716,
|
|
"mean_token_accuracy": 0.12691670581698417,
|
|
"num_tokens": 3525358.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.216009950637817,
|
|
"epoch": 0.10634279571142168,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 5.985,
|
|
"mean_token_accuracy": 0.13652188181877137,
|
|
"num_tokens": 3534459.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.155793523788452,
|
|
"epoch": 0.10663334979806491,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 5.9546,
|
|
"mean_token_accuracy": 0.1332765720784664,
|
|
"num_tokens": 3545268.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.335662126541138,
|
|
"epoch": 0.10692390388470814,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.0853,
|
|
"mean_token_accuracy": 0.13031049072742462,
|
|
"num_tokens": 3554281.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.235852289199829,
|
|
"epoch": 0.10721445797135137,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.0309,
|
|
"mean_token_accuracy": 0.13206790015101433,
|
|
"num_tokens": 3563505.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.190238428115845,
|
|
"epoch": 0.10750501205799459,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 5.9761,
|
|
"mean_token_accuracy": 0.14046704694628714,
|
|
"num_tokens": 3573288.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.100773715972901,
|
|
"epoch": 0.10779556614463782,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 5.898,
|
|
"mean_token_accuracy": 0.14386857226490973,
|
|
"num_tokens": 3582156.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.168911600112915,
|
|
"epoch": 0.10808612023128106,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.0786,
|
|
"mean_token_accuracy": 0.13001196533441545,
|
|
"num_tokens": 3592708.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.220205640792846,
|
|
"epoch": 0.10837667431792428,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 5.9618,
|
|
"mean_token_accuracy": 0.13206626623868942,
|
|
"num_tokens": 3602983.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.231088352203369,
|
|
"epoch": 0.10866722840456751,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 6.031,
|
|
"mean_token_accuracy": 0.13375057205557822,
|
|
"num_tokens": 3612779.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.18984317779541,
|
|
"epoch": 0.10895778249121074,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.0085,
|
|
"mean_token_accuracy": 0.1313602216541767,
|
|
"num_tokens": 3622850.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.1243922233581545,
|
|
"epoch": 0.10924833657785397,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.0192,
|
|
"mean_token_accuracy": 0.1339510276913643,
|
|
"num_tokens": 3632673.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.293065547943115,
|
|
"epoch": 0.1095388906644972,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 6.0195,
|
|
"mean_token_accuracy": 0.1283419005572796,
|
|
"num_tokens": 3642571.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.16258282661438,
|
|
"epoch": 0.10982944475114043,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.0009,
|
|
"mean_token_accuracy": 0.13664330318570136,
|
|
"num_tokens": 3651864.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.2043415069580075,
|
|
"epoch": 0.11011999883778366,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 5.9716,
|
|
"mean_token_accuracy": 0.1314692884683609,
|
|
"num_tokens": 3662410.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.146759462356568,
|
|
"epoch": 0.11041055292442688,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 5.9287,
|
|
"mean_token_accuracy": 0.13767404705286027,
|
|
"num_tokens": 3672340.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.102005672454834,
|
|
"epoch": 0.11070110701107011,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 6.0753,
|
|
"mean_token_accuracy": 0.1314219541847706,
|
|
"num_tokens": 3682745.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.236651849746704,
|
|
"epoch": 0.11099166109771334,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 5.9814,
|
|
"mean_token_accuracy": 0.13747138530015945,
|
|
"num_tokens": 3692075.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.1932165145874025,
|
|
"epoch": 0.11128221518435656,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 5.9772,
|
|
"mean_token_accuracy": 0.13977260813117026,
|
|
"num_tokens": 3701639.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.29685697555542,
|
|
"epoch": 0.1115727692709998,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.0266,
|
|
"mean_token_accuracy": 0.13467289954423906,
|
|
"num_tokens": 3711103.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.2153466701507565,
|
|
"epoch": 0.11186332335764303,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.0563,
|
|
"mean_token_accuracy": 0.1295161299407482,
|
|
"num_tokens": 3720229.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.287035751342773,
|
|
"epoch": 0.11215387744428626,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.0289,
|
|
"mean_token_accuracy": 0.1336451180279255,
|
|
"num_tokens": 3730616.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.164524412155151,
|
|
"epoch": 0.11244443153092948,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 5.8945,
|
|
"mean_token_accuracy": 0.13720172494649888,
|
|
"num_tokens": 3739829.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.15112156867981,
|
|
"epoch": 0.11273498561757271,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.0658,
|
|
"mean_token_accuracy": 0.12816951870918275,
|
|
"num_tokens": 3749830.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.2455854415893555,
|
|
"epoch": 0.11302553970421594,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 5.9084,
|
|
"mean_token_accuracy": 0.14013779759407044,
|
|
"num_tokens": 3759583.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.15108060836792,
|
|
"epoch": 0.11331609379085916,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 5.9655,
|
|
"mean_token_accuracy": 0.13013281747698785,
|
|
"num_tokens": 3769219.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.253466367721558,
|
|
"epoch": 0.1136066478775024,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 5.9664,
|
|
"mean_token_accuracy": 0.1329497739672661,
|
|
"num_tokens": 3778913.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.288890600204468,
|
|
"epoch": 0.11389720196414563,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.053,
|
|
"mean_token_accuracy": 0.133541439473629,
|
|
"num_tokens": 3788544.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.138053035736084,
|
|
"epoch": 0.11418775605078886,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 5.8634,
|
|
"mean_token_accuracy": 0.1413568802177906,
|
|
"num_tokens": 3797921.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.105274248123169,
|
|
"epoch": 0.11447831013743208,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.9334,
|
|
"mean_token_accuracy": 0.13669840842485428,
|
|
"num_tokens": 3807442.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.157075214385986,
|
|
"epoch": 0.11476886422407531,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 5.9183,
|
|
"mean_token_accuracy": 0.1388249270617962,
|
|
"num_tokens": 3817875.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.148885679244995,
|
|
"epoch": 0.11505941831071854,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 6.0358,
|
|
"mean_token_accuracy": 0.1268165521323681,
|
|
"num_tokens": 3826925.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.279354047775269,
|
|
"epoch": 0.11534997239736176,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.0141,
|
|
"mean_token_accuracy": 0.1367575228214264,
|
|
"num_tokens": 3836047.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.161008596420288,
|
|
"epoch": 0.115640526484005,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 5.969,
|
|
"mean_token_accuracy": 0.1333487443625927,
|
|
"num_tokens": 3845874.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.21445050239563,
|
|
"epoch": 0.11593108057064823,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 5.9627,
|
|
"mean_token_accuracy": 0.13459742665290833,
|
|
"num_tokens": 3855967.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.1095618724823,
|
|
"epoch": 0.11622163465729146,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 5.9908,
|
|
"mean_token_accuracy": 0.13831030651926995,
|
|
"num_tokens": 3866467.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.203917360305786,
|
|
"epoch": 0.11651218874393468,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 5.9744,
|
|
"mean_token_accuracy": 0.13815115690231322,
|
|
"num_tokens": 3875916.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.18111400604248,
|
|
"epoch": 0.11680274283057791,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 5.9594,
|
|
"mean_token_accuracy": 0.1330759234726429,
|
|
"num_tokens": 3885987.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.2428779125213625,
|
|
"epoch": 0.11709329691722115,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 6.0059,
|
|
"mean_token_accuracy": 0.13700252026319504,
|
|
"num_tokens": 3895809.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.195304489135742,
|
|
"epoch": 0.11738385100386436,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 6.0109,
|
|
"mean_token_accuracy": 0.132937653362751,
|
|
"num_tokens": 3905491.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.115702676773071,
|
|
"epoch": 0.1176744050905076,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 5.8753,
|
|
"mean_token_accuracy": 0.14159394055604935,
|
|
"num_tokens": 3915073.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.103581476211548,
|
|
"epoch": 0.11796495917715083,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 5.9516,
|
|
"mean_token_accuracy": 0.1373932972550392,
|
|
"num_tokens": 3924786.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.296129131317139,
|
|
"epoch": 0.11825551326379405,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 6.1096,
|
|
"mean_token_accuracy": 0.12966496869921684,
|
|
"num_tokens": 3934745.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.2405133724212645,
|
|
"epoch": 0.11854606735043728,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 5.928,
|
|
"mean_token_accuracy": 0.1365241065621376,
|
|
"num_tokens": 3945328.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.151012706756592,
|
|
"epoch": 0.11883662143708051,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 6.0549,
|
|
"mean_token_accuracy": 0.12878239378333092,
|
|
"num_tokens": 3955894.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.219704437255859,
|
|
"epoch": 0.11912717552372375,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 5.9845,
|
|
"mean_token_accuracy": 0.13799333572387695,
|
|
"num_tokens": 3964895.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.154746055603027,
|
|
"epoch": 0.11941772961036697,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 6.0396,
|
|
"mean_token_accuracy": 0.12946364358067514,
|
|
"num_tokens": 3974681.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.248775148391724,
|
|
"epoch": 0.1197082836970102,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 6.0252,
|
|
"mean_token_accuracy": 0.13851658627390862,
|
|
"num_tokens": 3984402.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.127353715896606,
|
|
"epoch": 0.11999883778365343,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 6.0268,
|
|
"mean_token_accuracy": 0.1265586420893669,
|
|
"num_tokens": 3995279.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.17973141670227,
|
|
"epoch": 0.12028939187029665,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 5.8617,
|
|
"mean_token_accuracy": 0.15186458677053452,
|
|
"num_tokens": 4003525.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.115002965927124,
|
|
"epoch": 0.12057994595693988,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.13773723766207696,
|
|
"num_tokens": 4012373.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.103297424316406,
|
|
"epoch": 0.12087050004358312,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 5.9031,
|
|
"mean_token_accuracy": 0.13564639389514924,
|
|
"num_tokens": 4021972.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.180972766876221,
|
|
"epoch": 0.12116105413022635,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 5.9966,
|
|
"mean_token_accuracy": 0.134269118309021,
|
|
"num_tokens": 4032384.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.158924150466919,
|
|
"epoch": 0.12145160821686957,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 5.8791,
|
|
"mean_token_accuracy": 0.13695783466100692,
|
|
"num_tokens": 4041974.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.068432807922363,
|
|
"epoch": 0.1217421623035128,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 5.9396,
|
|
"mean_token_accuracy": 0.13688302487134935,
|
|
"num_tokens": 4051694.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.212389612197876,
|
|
"epoch": 0.12203271639015603,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 6.002,
|
|
"mean_token_accuracy": 0.13414775878190993,
|
|
"num_tokens": 4062828.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.241210889816284,
|
|
"epoch": 0.12232327047679925,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 5.9742,
|
|
"mean_token_accuracy": 0.13490709364414216,
|
|
"num_tokens": 4072012.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.0934614658355715,
|
|
"epoch": 0.12261382456344248,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 5.9166,
|
|
"mean_token_accuracy": 0.14179718866944313,
|
|
"num_tokens": 4081587.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.224443626403809,
|
|
"epoch": 0.12290437865008572,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 5.9655,
|
|
"mean_token_accuracy": 0.13849997371435166,
|
|
"num_tokens": 4091374.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.211995363235474,
|
|
"epoch": 0.12319493273672895,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 5.9553,
|
|
"mean_token_accuracy": 0.1392621487379074,
|
|
"num_tokens": 4100532.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.090556001663208,
|
|
"epoch": 0.12348548682337217,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 5.9123,
|
|
"mean_token_accuracy": 0.13530985191464423,
|
|
"num_tokens": 4110194.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.171910429000855,
|
|
"epoch": 0.1237760409100154,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 5.9322,
|
|
"mean_token_accuracy": 0.136457958817482,
|
|
"num_tokens": 4119816.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.220505046844482,
|
|
"epoch": 0.12406659499665863,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 6.0141,
|
|
"mean_token_accuracy": 0.13071410208940507,
|
|
"num_tokens": 4130072.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.171974468231201,
|
|
"epoch": 0.12435714908330185,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 5.9574,
|
|
"mean_token_accuracy": 0.13431628346443175,
|
|
"num_tokens": 4139101.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.139108896255493,
|
|
"epoch": 0.12464770316994508,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 5.8817,
|
|
"mean_token_accuracy": 0.13240241631865501,
|
|
"num_tokens": 4149500.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.066398859024048,
|
|
"epoch": 0.12493825725658832,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 5.8244,
|
|
"mean_token_accuracy": 0.14197371304035186,
|
|
"num_tokens": 4158642.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.045716571807861,
|
|
"epoch": 0.12522881134323155,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 5.8963,
|
|
"mean_token_accuracy": 0.13848227709531785,
|
|
"num_tokens": 4167911.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.142486429214477,
|
|
"epoch": 0.12551936542987477,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 5.8062,
|
|
"mean_token_accuracy": 0.14425584971904754,
|
|
"num_tokens": 4177589.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.1435057640075685,
|
|
"epoch": 0.125809919516518,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 5.9855,
|
|
"mean_token_accuracy": 0.13806044012308122,
|
|
"num_tokens": 4187332.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.165029859542846,
|
|
"epoch": 0.12610047360316123,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 5.9359,
|
|
"mean_token_accuracy": 0.1438089445233345,
|
|
"num_tokens": 4196686.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.102525806427002,
|
|
"epoch": 0.12639102768980445,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 5.8595,
|
|
"mean_token_accuracy": 0.14088326916098595,
|
|
"num_tokens": 4205908.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.056342458724975,
|
|
"epoch": 0.1266815817764477,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 5.8166,
|
|
"mean_token_accuracy": 0.14242705181241036,
|
|
"num_tokens": 4215992.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.085980653762817,
|
|
"epoch": 0.12697213586309092,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 5.8801,
|
|
"mean_token_accuracy": 0.14103155285120011,
|
|
"num_tokens": 4227094.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.166931819915772,
|
|
"epoch": 0.12726268994973414,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 5.9574,
|
|
"mean_token_accuracy": 0.14575981721282005,
|
|
"num_tokens": 4236432.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.066125965118408,
|
|
"epoch": 0.12755324403637738,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 5.7683,
|
|
"mean_token_accuracy": 0.14632787331938743,
|
|
"num_tokens": 4245305.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.0759042263031,
|
|
"epoch": 0.1278437981230206,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 5.9274,
|
|
"mean_token_accuracy": 0.13991366624832152,
|
|
"num_tokens": 4255445.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.161522722244262,
|
|
"epoch": 0.12813435220966382,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 5.9095,
|
|
"mean_token_accuracy": 0.14544984251260756,
|
|
"num_tokens": 4265114.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.1537879467010494,
|
|
"epoch": 0.12842490629630707,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 5.9025,
|
|
"mean_token_accuracy": 0.13503851667046546,
|
|
"num_tokens": 4274787.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.1523271083831785,
|
|
"epoch": 0.1287154603829503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 5.9789,
|
|
"mean_token_accuracy": 0.13642423674464227,
|
|
"num_tokens": 4284509.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.217884969711304,
|
|
"epoch": 0.1290060144695935,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 5.9438,
|
|
"mean_token_accuracy": 0.1450071580708027,
|
|
"num_tokens": 4294128.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.103723478317261,
|
|
"epoch": 0.12929656855623675,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 5.9493,
|
|
"mean_token_accuracy": 0.13352228179574013,
|
|
"num_tokens": 4303779.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.194416570663452,
|
|
"epoch": 0.12958712264287997,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 5.9859,
|
|
"mean_token_accuracy": 0.14135119765996934,
|
|
"num_tokens": 4314520.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.184441709518433,
|
|
"epoch": 0.1298776767295232,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 5.9214,
|
|
"mean_token_accuracy": 0.13849867284297943,
|
|
"num_tokens": 4324477.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.059356164932251,
|
|
"epoch": 0.13016823081616644,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 5.8373,
|
|
"mean_token_accuracy": 0.14615851789712905,
|
|
"num_tokens": 4334009.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.182690954208374,
|
|
"epoch": 0.13045878490280965,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 5.9892,
|
|
"mean_token_accuracy": 0.13937190547585487,
|
|
"num_tokens": 4343748.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.105481243133545,
|
|
"epoch": 0.13074933898945287,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 5.9097,
|
|
"mean_token_accuracy": 0.13728254288434982,
|
|
"num_tokens": 4352957.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.125918388366699,
|
|
"epoch": 0.13103989307609612,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 5.9005,
|
|
"mean_token_accuracy": 0.1423856124281883,
|
|
"num_tokens": 4362505.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.138764905929565,
|
|
"epoch": 0.13133044716273934,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 5.9093,
|
|
"mean_token_accuracy": 0.14300098568201064,
|
|
"num_tokens": 4371664.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.16502833366394,
|
|
"epoch": 0.13162100124938259,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 6.0271,
|
|
"mean_token_accuracy": 0.13194756507873534,
|
|
"num_tokens": 4382600.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.100395345687867,
|
|
"epoch": 0.1319115553360258,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 5.927,
|
|
"mean_token_accuracy": 0.13698131814599038,
|
|
"num_tokens": 4393672.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.156081247329712,
|
|
"epoch": 0.13220210942266902,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 5.8873,
|
|
"mean_token_accuracy": 0.13874078989028932,
|
|
"num_tokens": 4403156.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.1929707527160645,
|
|
"epoch": 0.13249266350931227,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 5.9783,
|
|
"mean_token_accuracy": 0.13139391839504241,
|
|
"num_tokens": 4412742.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.086524152755738,
|
|
"epoch": 0.1327832175959555,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 5.8274,
|
|
"mean_token_accuracy": 0.15008396059274673,
|
|
"num_tokens": 4422232.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.099150848388672,
|
|
"epoch": 0.1330737716825987,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 5.8815,
|
|
"mean_token_accuracy": 0.13752613663673402,
|
|
"num_tokens": 4432405.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.194040107727051,
|
|
"epoch": 0.13336432576924195,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 5.9082,
|
|
"mean_token_accuracy": 0.1354072481393814,
|
|
"num_tokens": 4441504.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.136425590515136,
|
|
"epoch": 0.13365487985588517,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 5.9503,
|
|
"mean_token_accuracy": 0.1408660188317299,
|
|
"num_tokens": 4452478.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.132717275619507,
|
|
"epoch": 0.1339454339425284,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 5.8967,
|
|
"mean_token_accuracy": 0.13943730369210244,
|
|
"num_tokens": 4462267.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.183125257492065,
|
|
"epoch": 0.13423598802917164,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 5.9465,
|
|
"mean_token_accuracy": 0.14744184017181397,
|
|
"num_tokens": 4472580.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.118056774139404,
|
|
"epoch": 0.13452654211581486,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 5.9433,
|
|
"mean_token_accuracy": 0.13671476170420646,
|
|
"num_tokens": 4481768.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.0453300952911375,
|
|
"epoch": 0.13481709620245808,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 5.8219,
|
|
"mean_token_accuracy": 0.14138804078102113,
|
|
"num_tokens": 4491269.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.065739154815674,
|
|
"epoch": 0.13510765028910132,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 5.7896,
|
|
"mean_token_accuracy": 0.14193628579378129,
|
|
"num_tokens": 4501357.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.165984678268432,
|
|
"epoch": 0.13539820437574454,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 5.829,
|
|
"mean_token_accuracy": 0.14373186007142066,
|
|
"num_tokens": 4510972.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.087504100799561,
|
|
"epoch": 0.13568875846238776,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 5.8799,
|
|
"mean_token_accuracy": 0.1405518189072609,
|
|
"num_tokens": 4520887.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.0864016056060795,
|
|
"epoch": 0.135979312549031,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 5.8226,
|
|
"mean_token_accuracy": 0.14198774620890617,
|
|
"num_tokens": 4529971.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.121823167800903,
|
|
"epoch": 0.13626986663567422,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 5.9209,
|
|
"mean_token_accuracy": 0.1370498724281788,
|
|
"num_tokens": 4540375.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.068475914001465,
|
|
"epoch": 0.13656042072231747,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 5.8503,
|
|
"mean_token_accuracy": 0.13427165821194648,
|
|
"num_tokens": 4549554.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.043267726898193,
|
|
"epoch": 0.1368509748089607,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 5.7866,
|
|
"mean_token_accuracy": 0.14580736979842185,
|
|
"num_tokens": 4558697.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.053312063217163,
|
|
"epoch": 0.1371415288956039,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 5.8247,
|
|
"mean_token_accuracy": 0.14898887798190116,
|
|
"num_tokens": 4567592.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.1225522518157955,
|
|
"epoch": 0.13743208298224716,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 5.8661,
|
|
"mean_token_accuracy": 0.14503989070653917,
|
|
"num_tokens": 4576594.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.086222171783447,
|
|
"epoch": 0.13772263706889037,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 5.8143,
|
|
"mean_token_accuracy": 0.14734100848436354,
|
|
"num_tokens": 4585317.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.101655912399292,
|
|
"epoch": 0.1380131911555336,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 5.8265,
|
|
"mean_token_accuracy": 0.15099092870950698,
|
|
"num_tokens": 4593621.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.059905099868774,
|
|
"epoch": 0.13830374524217684,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 5.8365,
|
|
"mean_token_accuracy": 0.13829366117715836,
|
|
"num_tokens": 4603537.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.074821090698242,
|
|
"epoch": 0.13859429932882006,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 5.8688,
|
|
"mean_token_accuracy": 0.14191555976867676,
|
|
"num_tokens": 4613085.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.087924957275391,
|
|
"epoch": 0.13888485341546328,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 5.8978,
|
|
"mean_token_accuracy": 0.13870447725057602,
|
|
"num_tokens": 4622750.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.221824693679809,
|
|
"epoch": 0.13917540750210652,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 5.9973,
|
|
"mean_token_accuracy": 0.13631478250026702,
|
|
"num_tokens": 4632787.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.084641647338867,
|
|
"epoch": 0.13946596158874974,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 5.8022,
|
|
"mean_token_accuracy": 0.14561834558844566,
|
|
"num_tokens": 4642277.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.126396894454956,
|
|
"epoch": 0.13975651567539296,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 5.8342,
|
|
"mean_token_accuracy": 0.14577654898166656,
|
|
"num_tokens": 4651764.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.0264387130737305,
|
|
"epoch": 0.1400470697620362,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 5.8166,
|
|
"mean_token_accuracy": 0.14705195873975754,
|
|
"num_tokens": 4661938.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.151280355453491,
|
|
"epoch": 0.14033762384867943,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 5.8146,
|
|
"mean_token_accuracy": 0.14316702708601953,
|
|
"num_tokens": 4670960.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.228445291519165,
|
|
"epoch": 0.14062817793532267,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.1354992315173149,
|
|
"num_tokens": 4680777.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.084039545059204,
|
|
"epoch": 0.1409187320219659,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 5.7482,
|
|
"mean_token_accuracy": 0.15117157846689225,
|
|
"num_tokens": 4690390.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.010620021820069,
|
|
"epoch": 0.1412092861086091,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 5.9686,
|
|
"mean_token_accuracy": 0.14216312393546104,
|
|
"num_tokens": 4701133.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.182425451278687,
|
|
"epoch": 0.14149984019525236,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 5.9243,
|
|
"mean_token_accuracy": 0.14200448989868164,
|
|
"num_tokens": 4711962.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.141026830673217,
|
|
"epoch": 0.14179039428189558,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 5.8835,
|
|
"mean_token_accuracy": 0.14305603951215745,
|
|
"num_tokens": 4721748.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.0170793533325195,
|
|
"epoch": 0.1420809483685388,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 5.766,
|
|
"mean_token_accuracy": 0.14974772036075593,
|
|
"num_tokens": 4731178.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.9737021923065186,
|
|
"epoch": 0.14237150245518204,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 5.7396,
|
|
"mean_token_accuracy": 0.148803973197937,
|
|
"num_tokens": 4739590.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.12122106552124,
|
|
"epoch": 0.14266205654182526,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 5.9511,
|
|
"mean_token_accuracy": 0.14803745746612548,
|
|
"num_tokens": 4749840.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.207825994491577,
|
|
"epoch": 0.14295261062846848,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 5.9567,
|
|
"mean_token_accuracy": 0.13596878871321677,
|
|
"num_tokens": 4759347.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.198162078857422,
|
|
"epoch": 0.14324316471511173,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 5.8925,
|
|
"mean_token_accuracy": 0.14506246596574784,
|
|
"num_tokens": 4769911.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.0526893615722654,
|
|
"epoch": 0.14353371880175494,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 5.8162,
|
|
"mean_token_accuracy": 0.1422274589538574,
|
|
"num_tokens": 4780775.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.0685014724731445,
|
|
"epoch": 0.14382427288839816,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 5.8628,
|
|
"mean_token_accuracy": 0.1423807591199875,
|
|
"num_tokens": 4790845.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.090683555603027,
|
|
"epoch": 0.1441148269750414,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 5.8106,
|
|
"mean_token_accuracy": 0.1501375898718834,
|
|
"num_tokens": 4801214.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.07487530708313,
|
|
"epoch": 0.14440538106168463,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 5.7633,
|
|
"mean_token_accuracy": 0.1493755668401718,
|
|
"num_tokens": 4810720.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.07555193901062,
|
|
"epoch": 0.14469593514832785,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 5.7415,
|
|
"mean_token_accuracy": 0.14737216681241988,
|
|
"num_tokens": 4820078.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.0236917495727536,
|
|
"epoch": 0.1449864892349711,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 5.8242,
|
|
"mean_token_accuracy": 0.14522194787859916,
|
|
"num_tokens": 4830561.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.035146951675415,
|
|
"epoch": 0.1452770433216143,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 5.8653,
|
|
"mean_token_accuracy": 0.1423221454024315,
|
|
"num_tokens": 4839849.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.047043657302856,
|
|
"epoch": 0.14556759740825756,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002740575249600342,
|
|
"loss": 5.809,
|
|
"mean_token_accuracy": 0.14349082857370377,
|
|
"num_tokens": 4850113.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.110679817199707,
|
|
"epoch": 0.14585815149490078,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027287945635254263,
|
|
"loss": 5.7927,
|
|
"mean_token_accuracy": 0.13830389603972434,
|
|
"num_tokens": 4859572.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.028803539276123,
|
|
"epoch": 0.146148705581544,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00027170144588082635,
|
|
"loss": 5.8538,
|
|
"mean_token_accuracy": 0.14364985525608062,
|
|
"num_tokens": 4870238.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.05920820236206,
|
|
"epoch": 0.14643925966818724,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00027052352584063763,
|
|
"loss": 5.7673,
|
|
"mean_token_accuracy": 0.14797088503837585,
|
|
"num_tokens": 4879743.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.063130807876587,
|
|
"epoch": 0.14672981375483046,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 5.8141,
|
|
"mean_token_accuracy": 0.1519768014550209,
|
|
"num_tokens": 4888874.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.081371688842774,
|
|
"epoch": 0.14702036784147368,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00026816808622456937,
|
|
"loss": 5.8422,
|
|
"mean_token_accuracy": 0.14229626208543777,
|
|
"num_tokens": 4898244.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.169855117797852,
|
|
"epoch": 0.14731092192811693,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002669906312242569,
|
|
"loss": 5.8431,
|
|
"mean_token_accuracy": 0.1464947611093521,
|
|
"num_tokens": 4907978.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.089100074768067,
|
|
"epoch": 0.14760147601476015,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00026581339580483525,
|
|
"loss": 5.8544,
|
|
"mean_token_accuracy": 0.14092473834753036,
|
|
"num_tokens": 4917647.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.051334381103516,
|
|
"epoch": 0.14789203010140337,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002646364122408082,
|
|
"loss": 5.8866,
|
|
"mean_token_accuracy": 0.1406030498445034,
|
|
"num_tokens": 4927322.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.089036989212036,
|
|
"epoch": 0.1481825841880466,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002634597127997749,
|
|
"loss": 5.8912,
|
|
"mean_token_accuracy": 0.14071550071239472,
|
|
"num_tokens": 4938095.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.179394674301148,
|
|
"epoch": 0.14847313827468983,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002622833297415445,
|
|
"loss": 5.8789,
|
|
"mean_token_accuracy": 0.14479369372129441,
|
|
"num_tokens": 4947768.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.177506828308106,
|
|
"epoch": 0.14876369236133305,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002611072953172531,
|
|
"loss": 5.8719,
|
|
"mean_token_accuracy": 0.14385495483875274,
|
|
"num_tokens": 4957776.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.029763984680176,
|
|
"epoch": 0.1490542464479763,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00025993164176847845,
|
|
"loss": 5.8281,
|
|
"mean_token_accuracy": 0.15120696425437927,
|
|
"num_tokens": 4966992.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.071042394638061,
|
|
"epoch": 0.14934480053461952,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002587564013263564,
|
|
"loss": 5.8395,
|
|
"mean_token_accuracy": 0.14219800606369973,
|
|
"num_tokens": 4976216.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.099312973022461,
|
|
"epoch": 0.14963535462126276,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002575816062106974,
|
|
"loss": 5.7508,
|
|
"mean_token_accuracy": 0.14777441769838334,
|
|
"num_tokens": 4985143.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.13524899482727,
|
|
"epoch": 0.14992590870790598,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00025640728862910293,
|
|
"loss": 5.9437,
|
|
"mean_token_accuracy": 0.13997391015291213,
|
|
"num_tokens": 4995058.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.081535530090332,
|
|
"epoch": 0.1502164627945492,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 5.7767,
|
|
"mean_token_accuracy": 0.14847566336393356,
|
|
"num_tokens": 5003930.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.133546257019043,
|
|
"epoch": 0.15050701688119245,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00025406021483217225,
|
|
"loss": 5.8917,
|
|
"mean_token_accuracy": 0.14307338669896125,
|
|
"num_tokens": 5013907.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.05073037147522,
|
|
"epoch": 0.15079757096783566,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00025288752296304963,
|
|
"loss": 5.7465,
|
|
"mean_token_accuracy": 0.1435894712805748,
|
|
"num_tokens": 5024028.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.054743099212646,
|
|
"epoch": 0.15108812505447888,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000251715437318655,
|
|
"loss": 5.8135,
|
|
"mean_token_accuracy": 0.14531584978103637,
|
|
"num_tokens": 5033555.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.053986740112305,
|
|
"epoch": 0.15137867914112213,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002505439900323084,
|
|
"loss": 5.862,
|
|
"mean_token_accuracy": 0.14667272865772246,
|
|
"num_tokens": 5043180.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.088306331634522,
|
|
"epoch": 0.15166923322776535,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00024937321321982894,
|
|
"loss": 5.7691,
|
|
"mean_token_accuracy": 0.14470289498567582,
|
|
"num_tokens": 5052220.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.0878173351287845,
|
|
"epoch": 0.15195978731440857,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00024820313897865433,
|
|
"loss": 5.7726,
|
|
"mean_token_accuracy": 0.150884909927845,
|
|
"num_tokens": 5061544.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.131782722473145,
|
|
"epoch": 0.15225034140105181,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00024703379938696105,
|
|
"loss": 5.9184,
|
|
"mean_token_accuracy": 0.1400494635105133,
|
|
"num_tokens": 5070611.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.153847122192383,
|
|
"epoch": 0.15254089548769503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00024586522650278447,
|
|
"loss": 5.874,
|
|
"mean_token_accuracy": 0.1386608324944973,
|
|
"num_tokens": 5080750.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.137750577926636,
|
|
"epoch": 0.15283144957433825,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00024469745236314064,
|
|
"loss": 5.8592,
|
|
"mean_token_accuracy": 0.13961437940597535,
|
|
"num_tokens": 5090067.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.123914098739624,
|
|
"epoch": 0.1531220036609815,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00024353050898314767,
|
|
"loss": 5.8592,
|
|
"mean_token_accuracy": 0.14175378978252412,
|
|
"num_tokens": 5100053.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.1782163143157955,
|
|
"epoch": 0.15341255774762472,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00024236442835514743,
|
|
"loss": 5.8117,
|
|
"mean_token_accuracy": 0.1458034932613373,
|
|
"num_tokens": 5109296.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.071597719192505,
|
|
"epoch": 0.15370311183426794,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 5.8673,
|
|
"mean_token_accuracy": 0.14649384766817092,
|
|
"num_tokens": 5118744.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.130202531814575,
|
|
"epoch": 0.15399366592091118,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00024003498320535462,
|
|
"loss": 5.8775,
|
|
"mean_token_accuracy": 0.1437153235077858,
|
|
"num_tokens": 5127763.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.148328161239624,
|
|
"epoch": 0.1542842200075544,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023887168254647727,
|
|
"loss": 5.9019,
|
|
"mean_token_accuracy": 0.14456916153430938,
|
|
"num_tokens": 5138067.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.150043106079101,
|
|
"epoch": 0.15457477409419765,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00023770937236367308,
|
|
"loss": 5.8983,
|
|
"mean_token_accuracy": 0.1399595282971859,
|
|
"num_tokens": 5148280.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.082090711593628,
|
|
"epoch": 0.15486532818084087,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00023654808452226278,
|
|
"loss": 5.7799,
|
|
"mean_token_accuracy": 0.15123223662376403,
|
|
"num_tokens": 5158182.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.0570969581604,
|
|
"epoch": 0.15515588226748409,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00023538785085953912,
|
|
"loss": 5.7383,
|
|
"mean_token_accuracy": 0.14949096888303756,
|
|
"num_tokens": 5167524.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.062791872024536,
|
|
"epoch": 0.15544643635412733,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00023422870318389404,
|
|
"loss": 5.7904,
|
|
"mean_token_accuracy": 0.13950854763388634,
|
|
"num_tokens": 5177581.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.112806892395019,
|
|
"epoch": 0.15573699044077055,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002330706732739468,
|
|
"loss": 5.783,
|
|
"mean_token_accuracy": 0.14393220096826553,
|
|
"num_tokens": 5187156.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.065037822723388,
|
|
"epoch": 0.15602754452741377,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00023191379287767211,
|
|
"loss": 5.8843,
|
|
"mean_token_accuracy": 0.14143779054284095,
|
|
"num_tokens": 5198015.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.149574279785156,
|
|
"epoch": 0.15631809861405702,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002307580937115305,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.14658329337835313,
|
|
"num_tokens": 5207961.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.041101455688477,
|
|
"epoch": 0.15660865270070023,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00022960360745959846,
|
|
"loss": 5.8328,
|
|
"mean_token_accuracy": 0.14369555339217185,
|
|
"num_tokens": 5217318.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.05631422996521,
|
|
"epoch": 0.15689920678734345,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00022845036577269972,
|
|
"loss": 5.6925,
|
|
"mean_token_accuracy": 0.1581657573580742,
|
|
"num_tokens": 5226393.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.980499124526977,
|
|
"epoch": 0.1571897608739867,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 5.6844,
|
|
"mean_token_accuracy": 0.14944447427988053,
|
|
"num_tokens": 5236003.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.103996896743775,
|
|
"epoch": 0.15748031496062992,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002261477425258287,
|
|
"loss": 5.853,
|
|
"mean_token_accuracy": 0.1508561223745346,
|
|
"num_tokens": 5246472.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.170705938339234,
|
|
"epoch": 0.15777086904727314,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002249984240934358,
|
|
"loss": 5.931,
|
|
"mean_token_accuracy": 0.1417808599770069,
|
|
"num_tokens": 5256921.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.98744764328003,
|
|
"epoch": 0.15806142313391638,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00022385047647950464,
|
|
"loss": 5.7333,
|
|
"mean_token_accuracy": 0.15196397304534912,
|
|
"num_tokens": 5266832.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.991374492645264,
|
|
"epoch": 0.1583519772205596,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002227039311555986,
|
|
"loss": 5.7585,
|
|
"mean_token_accuracy": 0.1483888141810894,
|
|
"num_tokens": 5276386.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.100140237808228,
|
|
"epoch": 0.15864253130720282,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002215588195548372,
|
|
"loss": 5.7618,
|
|
"mean_token_accuracy": 0.14937272816896438,
|
|
"num_tokens": 5285959.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.154450845718384,
|
|
"epoch": 0.15893308539384607,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00022041517307103337,
|
|
"loss": 5.7947,
|
|
"mean_token_accuracy": 0.14946697056293487,
|
|
"num_tokens": 5295457.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.096340894699097,
|
|
"epoch": 0.1592236394804893,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0002192730230578331,
|
|
"loss": 5.8109,
|
|
"mean_token_accuracy": 0.14488886743783952,
|
|
"num_tokens": 5306092.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.052392101287841,
|
|
"epoch": 0.15951419356713253,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002181324008278559,
|
|
"loss": 5.8358,
|
|
"mean_token_accuracy": 0.14705842584371567,
|
|
"num_tokens": 5314960.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.998585510253906,
|
|
"epoch": 0.15980474765377575,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00021699333765183655,
|
|
"loss": 5.7745,
|
|
"mean_token_accuracy": 0.15136635154485703,
|
|
"num_tokens": 5324390.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.0869420051574705,
|
|
"epoch": 0.16009530174041897,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002158558647577673,
|
|
"loss": 5.812,
|
|
"mean_token_accuracy": 0.14605457559227944,
|
|
"num_tokens": 5334650.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.197345113754272,
|
|
"epoch": 0.16038585582706222,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00021472001333004215,
|
|
"loss": 5.8713,
|
|
"mean_token_accuracy": 0.1437445230782032,
|
|
"num_tokens": 5343958.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.1040606021881105,
|
|
"epoch": 0.16067640991370544,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 5.8254,
|
|
"mean_token_accuracy": 0.14942396581172943,
|
|
"num_tokens": 5353428.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.14132776260376,
|
|
"epoch": 0.16096696400034866,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002124532993880799,
|
|
"loss": 5.8757,
|
|
"mean_token_accuracy": 0.14288587495684624,
|
|
"num_tokens": 5364132.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.148387336730957,
|
|
"epoch": 0.1612575180869919,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00021132249901695044,
|
|
"loss": 5.853,
|
|
"mean_token_accuracy": 0.14395386576652527,
|
|
"num_tokens": 5374066.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.0949657440185545,
|
|
"epoch": 0.16154807217363512,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00021019344439667705,
|
|
"loss": 5.8179,
|
|
"mean_token_accuracy": 0.1493311658501625,
|
|
"num_tokens": 5383479.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.99338116645813,
|
|
"epoch": 0.16183862626027834,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00020906616648086213,
|
|
"loss": 5.683,
|
|
"mean_token_accuracy": 0.15754484832286836,
|
|
"num_tokens": 5392894.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.032156229019165,
|
|
"epoch": 0.1621291803469216,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00020794069617439942,
|
|
"loss": 5.8144,
|
|
"mean_token_accuracy": 0.1486166849732399,
|
|
"num_tokens": 5402886.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.081609296798706,
|
|
"epoch": 0.1624197344335648,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00020681706433262593,
|
|
"loss": 5.7009,
|
|
"mean_token_accuracy": 0.1564814940094948,
|
|
"num_tokens": 5411656.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.125734090805054,
|
|
"epoch": 0.16271028852020802,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00020569530176047602,
|
|
"loss": 5.8456,
|
|
"mean_token_accuracy": 0.14672704488039018,
|
|
"num_tokens": 5421650.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.053952217102051,
|
|
"epoch": 0.16300084260685127,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002045754392116374,
|
|
"loss": 5.7943,
|
|
"mean_token_accuracy": 0.153436142206192,
|
|
"num_tokens": 5431154.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.064789199829102,
|
|
"epoch": 0.1632913966934945,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00020345750738770757,
|
|
"loss": 5.8505,
|
|
"mean_token_accuracy": 0.14092413783073426,
|
|
"num_tokens": 5441464.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.110472869873047,
|
|
"epoch": 0.16358195078013774,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00020234153693735214,
|
|
"loss": 5.8195,
|
|
"mean_token_accuracy": 0.14966847896575927,
|
|
"num_tokens": 5452077.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.1211137771606445,
|
|
"epoch": 0.16387250486678095,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002012275584554647,
|
|
"loss": 5.8616,
|
|
"mean_token_accuracy": 0.14003771468997,
|
|
"num_tokens": 5461792.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.047807073593139,
|
|
"epoch": 0.16416305895342417,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 5.7094,
|
|
"mean_token_accuracy": 0.15667359828948973,
|
|
"num_tokens": 5471637.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.167580080032349,
|
|
"epoch": 0.16445361304006742,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00019900569950277692,
|
|
"loss": 5.9432,
|
|
"mean_token_accuracy": 0.14090102761983872,
|
|
"num_tokens": 5482341.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.063886308670044,
|
|
"epoch": 0.16474416712671064,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00019789787994536228,
|
|
"loss": 5.7621,
|
|
"mean_token_accuracy": 0.15223144590854645,
|
|
"num_tokens": 5492335.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.057334852218628,
|
|
"epoch": 0.16503472121335386,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00019679217418151667,
|
|
"loss": 5.7486,
|
|
"mean_token_accuracy": 0.15512095093727113,
|
|
"num_tokens": 5501879.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.074555587768555,
|
|
"epoch": 0.1653252752999971,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00019568861252472236,
|
|
"loss": 5.7906,
|
|
"mean_token_accuracy": 0.15294522792100906,
|
|
"num_tokens": 5512419.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.052340698242188,
|
|
"epoch": 0.16561582938664032,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019458722522967952,
|
|
"loss": 5.6966,
|
|
"mean_token_accuracy": 0.15103003978729249,
|
|
"num_tokens": 5521017.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.01881628036499,
|
|
"epoch": 0.16590638347328354,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00019348804249147723,
|
|
"loss": 5.7061,
|
|
"mean_token_accuracy": 0.14818918108940124,
|
|
"num_tokens": 5530916.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.014477682113648,
|
|
"epoch": 0.1661969375599268,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001923910944447655,
|
|
"loss": 5.6511,
|
|
"mean_token_accuracy": 0.16009259968996048,
|
|
"num_tokens": 5540228.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.070650434494018,
|
|
"epoch": 0.16648749164657,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00019129641116292928,
|
|
"loss": 5.7298,
|
|
"mean_token_accuracy": 0.15569742172956466,
|
|
"num_tokens": 5549921.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.1058845043182375,
|
|
"epoch": 0.16677804573321323,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00019020402265726343,
|
|
"loss": 5.8318,
|
|
"mean_token_accuracy": 0.143881855905056,
|
|
"num_tokens": 5560308.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.0743451595306395,
|
|
"epoch": 0.16706859981985647,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001891139588761509,
|
|
"loss": 5.7595,
|
|
"mean_token_accuracy": 0.1477736845612526,
|
|
"num_tokens": 5569026.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.015086269378662,
|
|
"epoch": 0.1673591539064997,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00018802624970424076,
|
|
"loss": 5.725,
|
|
"mean_token_accuracy": 0.15312366485595702,
|
|
"num_tokens": 5578812.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.130430459976196,
|
|
"epoch": 0.1676497079931429,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 5.831,
|
|
"mean_token_accuracy": 0.14988720864057542,
|
|
"num_tokens": 5588763.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.03325777053833,
|
|
"epoch": 0.16794026207978616,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00018585801440304306,
|
|
"loss": 5.6702,
|
|
"mean_token_accuracy": 0.15091593116521834,
|
|
"num_tokens": 5597719.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.071933698654175,
|
|
"epoch": 0.16823081616642938,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00018477754771702165,
|
|
"loss": 5.7461,
|
|
"mean_token_accuracy": 0.14670687392354012,
|
|
"num_tokens": 5607376.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.061179494857788,
|
|
"epoch": 0.16852137025307262,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00018369955452510506,
|
|
"loss": 5.757,
|
|
"mean_token_accuracy": 0.14725697934627532,
|
|
"num_tokens": 5617227.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.983553457260132,
|
|
"epoch": 0.16881192433971584,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001826240643810212,
|
|
"loss": 5.6661,
|
|
"mean_token_accuracy": 0.1538853704929352,
|
|
"num_tokens": 5626241.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.049111032485962,
|
|
"epoch": 0.16910247842635906,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001815511067698758,
|
|
"loss": 5.8158,
|
|
"mean_token_accuracy": 0.14741248339414598,
|
|
"num_tokens": 5636969.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.976936292648316,
|
|
"epoch": 0.1693930325130023,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001804807111073436,
|
|
"loss": 5.6947,
|
|
"mean_token_accuracy": 0.14919188469648362,
|
|
"num_tokens": 5646430.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.0697746753692625,
|
|
"epoch": 0.16968358659964553,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001794129067388625,
|
|
"loss": 5.7424,
|
|
"mean_token_accuracy": 0.15479477643966674,
|
|
"num_tokens": 5656049.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.103779983520508,
|
|
"epoch": 0.16997414068628874,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017834772293882868,
|
|
"loss": 5.7761,
|
|
"mean_token_accuracy": 0.14724364280700683,
|
|
"num_tokens": 5665657.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.004815101623535,
|
|
"epoch": 0.170264694772932,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000177285188909794,
|
|
"loss": 5.6713,
|
|
"mean_token_accuracy": 0.14558330550789833,
|
|
"num_tokens": 5675306.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.077218818664551,
|
|
"epoch": 0.1705552488595752,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001762253337816656,
|
|
"loss": 5.8962,
|
|
"mean_token_accuracy": 0.14256232976913452,
|
|
"num_tokens": 5685295.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.960014724731446,
|
|
"epoch": 0.17084580294621843,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00017516818661090738,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.153633750975132,
|
|
"num_tokens": 5694626.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.0667417526245115,
|
|
"epoch": 0.17113635703286167,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 5.7658,
|
|
"mean_token_accuracy": 0.1459375351667404,
|
|
"num_tokens": 5704247.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.021193981170654,
|
|
"epoch": 0.1714269111195049,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00017306213199536115,
|
|
"loss": 5.6436,
|
|
"mean_token_accuracy": 0.16296780705451966,
|
|
"num_tokens": 5712536.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.047375822067261,
|
|
"epoch": 0.1717174652061481,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001720132822891243,
|
|
"loss": 5.7144,
|
|
"mean_token_accuracy": 0.1537844330072403,
|
|
"num_tokens": 5721641.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.961157178878784,
|
|
"epoch": 0.17200801929279136,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001709672560157769,
|
|
"loss": 5.6832,
|
|
"mean_token_accuracy": 0.15097189098596572,
|
|
"num_tokens": 5731404.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.061686420440674,
|
|
"epoch": 0.17229857337943458,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00016992408185265758,
|
|
"loss": 5.7934,
|
|
"mean_token_accuracy": 0.14903522282838821,
|
|
"num_tokens": 5741006.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.143270587921142,
|
|
"epoch": 0.17258912746607782,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00016888378839891298,
|
|
"loss": 5.7955,
|
|
"mean_token_accuracy": 0.14605200439691543,
|
|
"num_tokens": 5751124.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.10756402015686,
|
|
"epoch": 0.17287968155272104,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001678464041747137,
|
|
"loss": 5.7795,
|
|
"mean_token_accuracy": 0.15010488629341126,
|
|
"num_tokens": 5761261.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.070679616928101,
|
|
"epoch": 0.17317023563936426,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00016681195762047223,
|
|
"loss": 5.7228,
|
|
"mean_token_accuracy": 0.1600602760910988,
|
|
"num_tokens": 5769855.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.0531364440917965,
|
|
"epoch": 0.1734607897260075,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016578047709606337,
|
|
"loss": 5.7577,
|
|
"mean_token_accuracy": 0.14312802702188493,
|
|
"num_tokens": 5779394.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.0887946605682375,
|
|
"epoch": 0.17375134381265073,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00016475199088004678,
|
|
"loss": 5.8185,
|
|
"mean_token_accuracy": 0.15020564645528794,
|
|
"num_tokens": 5789442.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.0648823261260985,
|
|
"epoch": 0.17404189789929395,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00016372652716889163,
|
|
"loss": 5.7086,
|
|
"mean_token_accuracy": 0.15479273200035096,
|
|
"num_tokens": 5798269.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.102453422546387,
|
|
"epoch": 0.1743324519859372,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001627041140762035,
|
|
"loss": 5.7227,
|
|
"mean_token_accuracy": 0.14934398829936982,
|
|
"num_tokens": 5808608.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.100079727172852,
|
|
"epoch": 0.1746230060725804,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016168477963195382,
|
|
"loss": 5.777,
|
|
"mean_token_accuracy": 0.14889512956142426,
|
|
"num_tokens": 5818134.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.064034843444825,
|
|
"epoch": 0.17491356015922363,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001606685517817114,
|
|
"loss": 5.7285,
|
|
"mean_token_accuracy": 0.15235000252723693,
|
|
"num_tokens": 5828237.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.020033311843872,
|
|
"epoch": 0.17520411424586688,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00015965545838587592,
|
|
"loss": 5.7206,
|
|
"mean_token_accuracy": 0.1511564001441002,
|
|
"num_tokens": 5837597.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.070970773696899,
|
|
"epoch": 0.1754946683325101,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00015864552721891467,
|
|
"loss": 5.6728,
|
|
"mean_token_accuracy": 0.1517861396074295,
|
|
"num_tokens": 5846728.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.065035820007324,
|
|
"epoch": 0.17578522241915331,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00015763878596860076,
|
|
"loss": 5.831,
|
|
"mean_token_accuracy": 0.14596770107746124,
|
|
"num_tokens": 5857655.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.956533288955688,
|
|
"epoch": 0.17607577650579656,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00015663526223525412,
|
|
"loss": 5.6927,
|
|
"mean_token_accuracy": 0.1539006546139717,
|
|
"num_tokens": 5868854.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.1010034561157225,
|
|
"epoch": 0.17636633059243978,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0001556349835309848,
|
|
"loss": 5.8291,
|
|
"mean_token_accuracy": 0.1526801697909832,
|
|
"num_tokens": 5879362.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.131938362121582,
|
|
"epoch": 0.176656884679083,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001546379772789389,
|
|
"loss": 5.8241,
|
|
"mean_token_accuracy": 0.15398107543587686,
|
|
"num_tokens": 5888753.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.086278247833252,
|
|
"epoch": 0.17694743876572624,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00015364427081254622,
|
|
"loss": 5.7149,
|
|
"mean_token_accuracy": 0.15152743011713027,
|
|
"num_tokens": 5898860.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.081228303909302,
|
|
"epoch": 0.17723799285236946,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00015265389137477165,
|
|
"loss": 5.7008,
|
|
"mean_token_accuracy": 0.14895583540201188,
|
|
"num_tokens": 5908191.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.977576351165771,
|
|
"epoch": 0.1775285469390127,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00015166686611736786,
|
|
"loss": 5.6745,
|
|
"mean_token_accuracy": 0.15492946803569793,
|
|
"num_tokens": 5918266.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.968826675415039,
|
|
"epoch": 0.17781910102565593,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00015068322210013064,
|
|
"loss": 5.6773,
|
|
"mean_token_accuracy": 0.15108609497547149,
|
|
"num_tokens": 5927628.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.01093373298645,
|
|
"epoch": 0.17810965511229915,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001497029862901578,
|
|
"loss": 5.6803,
|
|
"mean_token_accuracy": 0.15184399709105492,
|
|
"num_tokens": 5937323.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.0069934844970705,
|
|
"epoch": 0.1784002091989424,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00014872618556110905,
|
|
"loss": 5.7033,
|
|
"mean_token_accuracy": 0.15416699647903442,
|
|
"num_tokens": 5946767.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.092588567733765,
|
|
"epoch": 0.1786907632855856,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00014775284669246992,
|
|
"loss": 5.7472,
|
|
"mean_token_accuracy": 0.1522138647735119,
|
|
"num_tokens": 5956469.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.093777990341186,
|
|
"epoch": 0.17898131737222883,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00014678299636881716,
|
|
"loss": 5.7564,
|
|
"mean_token_accuracy": 0.15174834728240966,
|
|
"num_tokens": 5965882.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.043925333023071,
|
|
"epoch": 0.17927187145887208,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001458166611790873,
|
|
"loss": 5.6561,
|
|
"mean_token_accuracy": 0.15863914489746095,
|
|
"num_tokens": 5974696.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.098592472076416,
|
|
"epoch": 0.1795624255455153,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00014485386761584773,
|
|
"loss": 5.771,
|
|
"mean_token_accuracy": 0.15556092113256453,
|
|
"num_tokens": 5984525.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.02237606048584,
|
|
"epoch": 0.17985297963215852,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00014389464207457042,
|
|
"loss": 5.7049,
|
|
"mean_token_accuracy": 0.1512262910604477,
|
|
"num_tokens": 5993382.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.985553407669068,
|
|
"epoch": 0.18014353371880176,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00014293901085290795,
|
|
"loss": 5.6662,
|
|
"mean_token_accuracy": 0.15523958802223206,
|
|
"num_tokens": 6003127.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.131969976425171,
|
|
"epoch": 0.18043408780544498,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00014198700014997307,
|
|
"loss": 5.8075,
|
|
"mean_token_accuracy": 0.1487472876906395,
|
|
"num_tokens": 6013336.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.103825998306275,
|
|
"epoch": 0.1807246418920882,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00014103863606562016,
|
|
"loss": 5.8092,
|
|
"mean_token_accuracy": 0.14103479385375978,
|
|
"num_tokens": 6023446.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.070917987823487,
|
|
"epoch": 0.18101519597873145,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00014009394459972964,
|
|
"loss": 5.7103,
|
|
"mean_token_accuracy": 0.15287835896015167,
|
|
"num_tokens": 6034264.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.039470911026001,
|
|
"epoch": 0.18130575006537467,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00013915295165149513,
|
|
"loss": 5.7217,
|
|
"mean_token_accuracy": 0.15219166725873948,
|
|
"num_tokens": 6043048.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.063743162155151,
|
|
"epoch": 0.18159630415201788,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00013821568301871384,
|
|
"loss": 5.7789,
|
|
"mean_token_accuracy": 0.14522581547498703,
|
|
"num_tokens": 6053255.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.9902284145355225,
|
|
"epoch": 0.18188685823866113,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00013728216439707862,
|
|
"loss": 5.6082,
|
|
"mean_token_accuracy": 0.1634930595755577,
|
|
"num_tokens": 6062968.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.016612720489502,
|
|
"epoch": 0.18217741232530435,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00013635242137947419,
|
|
"loss": 5.5827,
|
|
"mean_token_accuracy": 0.16410193741321563,
|
|
"num_tokens": 6072071.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.09333872795105,
|
|
"epoch": 0.1824679664119476,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00013542647945527498,
|
|
"loss": 5.8434,
|
|
"mean_token_accuracy": 0.1482686847448349,
|
|
"num_tokens": 6081887.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.0756289005279545,
|
|
"epoch": 0.18275852049859082,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001345043640096465,
|
|
"loss": 5.7092,
|
|
"mean_token_accuracy": 0.1483364373445511,
|
|
"num_tokens": 6092254.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.047313165664673,
|
|
"epoch": 0.18304907458523403,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00013358610032284957,
|
|
"loss": 5.72,
|
|
"mean_token_accuracy": 0.15549504160881042,
|
|
"num_tokens": 6102497.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.062089967727661,
|
|
"epoch": 0.18333962867187728,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000132671713569547,
|
|
"loss": 5.734,
|
|
"mean_token_accuracy": 0.1556983083486557,
|
|
"num_tokens": 6112277.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.076785612106323,
|
|
"epoch": 0.1836301827585205,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001317612288181136,
|
|
"loss": 5.7002,
|
|
"mean_token_accuracy": 0.1554645776748657,
|
|
"num_tokens": 6121471.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.072947454452515,
|
|
"epoch": 0.18392073684516372,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00013085467102994864,
|
|
"loss": 5.7982,
|
|
"mean_token_accuracy": 0.14808910414576532,
|
|
"num_tokens": 6131717.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.053280687332153,
|
|
"epoch": 0.18421129093180696,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00012995206505879198,
|
|
"loss": 5.708,
|
|
"mean_token_accuracy": 0.15172735154628753,
|
|
"num_tokens": 6142002.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.130218172073365,
|
|
"epoch": 0.18450184501845018,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001290534356500421,
|
|
"loss": 5.7836,
|
|
"mean_token_accuracy": 0.14181277081370353,
|
|
"num_tokens": 6152034.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.088483095169067,
|
|
"epoch": 0.1847923991050934,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00012815880744007827,
|
|
"loss": 5.7851,
|
|
"mean_token_accuracy": 0.1522969976067543,
|
|
"num_tokens": 6161737.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.9690474510192875,
|
|
"epoch": 0.18508295319173665,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00012726820495558483,
|
|
"loss": 5.6982,
|
|
"mean_token_accuracy": 0.153187994658947,
|
|
"num_tokens": 6170088.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.107737588882446,
|
|
"epoch": 0.18537350727837987,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00012638165261287868,
|
|
"loss": 5.8119,
|
|
"mean_token_accuracy": 0.1520651862025261,
|
|
"num_tokens": 6179914.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.0809577941894535,
|
|
"epoch": 0.1856640613650231,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001254991747172402,
|
|
"loss": 5.7642,
|
|
"mean_token_accuracy": 0.14715377390384674,
|
|
"num_tokens": 6191145.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.9991779804229735,
|
|
"epoch": 0.18595461545166633,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00012462079546224662,
|
|
"loss": 5.6445,
|
|
"mean_token_accuracy": 0.15297795236110687,
|
|
"num_tokens": 6201587.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.071609449386597,
|
|
"epoch": 0.18624516953830955,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00012374653892910896,
|
|
"loss": 5.7869,
|
|
"mean_token_accuracy": 0.14952635765075684,
|
|
"num_tokens": 6210659.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.06493353843689,
|
|
"epoch": 0.1865357236249528,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00012287642908601166,
|
|
"loss": 5.6995,
|
|
"mean_token_accuracy": 0.16741256564855575,
|
|
"num_tokens": 6220179.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.087040662765503,
|
|
"epoch": 0.18682627771159602,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00012201048978745569,
|
|
"loss": 5.8608,
|
|
"mean_token_accuracy": 0.15103042125701904,
|
|
"num_tokens": 6230568.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.1079224109649655,
|
|
"epoch": 0.18711683179823924,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00012114874477360427,
|
|
"loss": 5.7732,
|
|
"mean_token_accuracy": 0.15811807066202163,
|
|
"num_tokens": 6240571.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.066962003707886,
|
|
"epoch": 0.18740738588488248,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00012029121766963236,
|
|
"loss": 5.6959,
|
|
"mean_token_accuracy": 0.151812843978405,
|
|
"num_tokens": 6250232.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.136503267288208,
|
|
"epoch": 0.1876979399715257,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00011943793198507858,
|
|
"loss": 5.8473,
|
|
"mean_token_accuracy": 0.14256954118609427,
|
|
"num_tokens": 6260191.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.1150651454925535,
|
|
"epoch": 0.18798849405816892,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00011858891111320104,
|
|
"loss": 5.7593,
|
|
"mean_token_accuracy": 0.15107759982347488,
|
|
"num_tokens": 6270348.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.040066814422607,
|
|
"epoch": 0.18827904814481217,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001177441783303359,
|
|
"loss": 5.7222,
|
|
"mean_token_accuracy": 0.14626873433589935,
|
|
"num_tokens": 6281212.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.0591840744018555,
|
|
"epoch": 0.18856960223145539,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00011690375679525896,
|
|
"loss": 5.7569,
|
|
"mean_token_accuracy": 0.15225170105695723,
|
|
"num_tokens": 6290194.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.092053413391113,
|
|
"epoch": 0.1888601563180986,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00011606766954855124,
|
|
"loss": 5.7817,
|
|
"mean_token_accuracy": 0.15150906667113304,
|
|
"num_tokens": 6299875.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.102940845489502,
|
|
"epoch": 0.18915071040474185,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00011523593951196702,
|
|
"loss": 5.7722,
|
|
"mean_token_accuracy": 0.15258708745241165,
|
|
"num_tokens": 6309359.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.080491304397583,
|
|
"epoch": 0.18944126449138507,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00011440858948780523,
|
|
"loss": 5.7347,
|
|
"mean_token_accuracy": 0.1517005071043968,
|
|
"num_tokens": 6320783.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.987936544418335,
|
|
"epoch": 0.1897318185780283,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00011358564215828484,
|
|
"loss": 5.5715,
|
|
"mean_token_accuracy": 0.1621371328830719,
|
|
"num_tokens": 6330150.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.964755249023438,
|
|
"epoch": 0.19002237266467153,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00011276712008492254,
|
|
"loss": 5.7216,
|
|
"mean_token_accuracy": 0.15560902208089827,
|
|
"num_tokens": 6340421.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.003197145462036,
|
|
"epoch": 0.19031292675131475,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00011195304570791451,
|
|
"loss": 5.7291,
|
|
"mean_token_accuracy": 0.14981625527143477,
|
|
"num_tokens": 6351093.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.069776487350464,
|
|
"epoch": 0.19060348083795797,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00011114344134552094,
|
|
"loss": 5.7178,
|
|
"mean_token_accuracy": 0.14927603229880332,
|
|
"num_tokens": 6360373.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.128297328948975,
|
|
"epoch": 0.19089403492460122,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001103383291934545,
|
|
"loss": 5.8211,
|
|
"mean_token_accuracy": 0.15346538573503493,
|
|
"num_tokens": 6371291.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.147673416137695,
|
|
"epoch": 0.19118458901124444,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00010953773132427141,
|
|
"loss": 5.706,
|
|
"mean_token_accuracy": 0.15231086164712906,
|
|
"num_tokens": 6380869.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.075566673278809,
|
|
"epoch": 0.19147514309788768,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00010874166968676677,
|
|
"loss": 5.7058,
|
|
"mean_token_accuracy": 0.15668356716632842,
|
|
"num_tokens": 6390174.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.030622339248657,
|
|
"epoch": 0.1917656971845309,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00010795016610537251,
|
|
"loss": 5.6748,
|
|
"mean_token_accuracy": 0.15518099069595337,
|
|
"num_tokens": 6399768.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.077046346664429,
|
|
"epoch": 0.19205625127117412,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00010716324227955904,
|
|
"loss": 5.802,
|
|
"mean_token_accuracy": 0.1477528505027294,
|
|
"num_tokens": 6409460.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.034036684036255,
|
|
"epoch": 0.19234680535781737,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001063809197832406,
|
|
"loss": 5.6572,
|
|
"mean_token_accuracy": 0.16222479790449143,
|
|
"num_tokens": 6417968.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.085168313980103,
|
|
"epoch": 0.1926373594444606,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00010560322006418368,
|
|
"loss": 5.7371,
|
|
"mean_token_accuracy": 0.1469581514596939,
|
|
"num_tokens": 6427402.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.114069175720215,
|
|
"epoch": 0.1929279135311038,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00010483016444341887,
|
|
"loss": 5.7702,
|
|
"mean_token_accuracy": 0.15116747766733168,
|
|
"num_tokens": 6437203.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.091079235076904,
|
|
"epoch": 0.19321846761774705,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00010406177411465654,
|
|
"loss": 5.6856,
|
|
"mean_token_accuracy": 0.15697493702173232,
|
|
"num_tokens": 6446440.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.068310308456421,
|
|
"epoch": 0.19350902170439027,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00010329807014370562,
|
|
"loss": 5.6496,
|
|
"mean_token_accuracy": 0.157624289393425,
|
|
"num_tokens": 6455842.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.069522714614868,
|
|
"epoch": 0.1937995757910335,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00010253907346789632,
|
|
"loss": 5.6689,
|
|
"mean_token_accuracy": 0.15749077796936034,
|
|
"num_tokens": 6464538.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.971843290328979,
|
|
"epoch": 0.19409012987767674,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00010178480489550596,
|
|
"loss": 5.6299,
|
|
"mean_token_accuracy": 0.15191589742898942,
|
|
"num_tokens": 6474646.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.010790205001831,
|
|
"epoch": 0.19438068396431996,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00010103528510518836,
|
|
"loss": 5.7641,
|
|
"mean_token_accuracy": 0.14547111392021178,
|
|
"num_tokens": 6484397.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.044341564178467,
|
|
"epoch": 0.19467123805096317,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001002905346454073,
|
|
"loss": 5.7108,
|
|
"mean_token_accuracy": 0.14943148642778398,
|
|
"num_tokens": 6494254.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.041453218460083,
|
|
"epoch": 0.19496179213760642,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 9.955057393387285e-05,
|
|
"loss": 5.6536,
|
|
"mean_token_accuracy": 0.16039068698883058,
|
|
"num_tokens": 6503862.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.147222709655762,
|
|
"epoch": 0.19525234622424964,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 9.88154232569816e-05,
|
|
"loss": 5.7122,
|
|
"mean_token_accuracy": 0.14973994195461274,
|
|
"num_tokens": 6513644.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.084700441360473,
|
|
"epoch": 0.1955429003108929,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.808510276926075e-05,
|
|
"loss": 5.7,
|
|
"mean_token_accuracy": 0.14390370547771453,
|
|
"num_tokens": 6523991.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.049271821975708,
|
|
"epoch": 0.1958334543975361,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.735963249281549e-05,
|
|
"loss": 5.7439,
|
|
"mean_token_accuracy": 0.14755677580833435,
|
|
"num_tokens": 6533761.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.067014598846436,
|
|
"epoch": 0.19612400848417932,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.663903231677974e-05,
|
|
"loss": 5.7136,
|
|
"mean_token_accuracy": 0.1520987056195736,
|
|
"num_tokens": 6544115.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.102639961242676,
|
|
"epoch": 0.19641456257082257,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 9.592332199677145e-05,
|
|
"loss": 5.7757,
|
|
"mean_token_accuracy": 0.1473358005285263,
|
|
"num_tokens": 6555207.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.06288423538208,
|
|
"epoch": 0.1967051166574658,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 9.521252115435061e-05,
|
|
"loss": 5.7049,
|
|
"mean_token_accuracy": 0.15305796936154364,
|
|
"num_tokens": 6564725.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.05938458442688,
|
|
"epoch": 0.196995670744109,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 9.450664927648126e-05,
|
|
"loss": 5.7016,
|
|
"mean_token_accuracy": 0.14934950321912766,
|
|
"num_tokens": 6575036.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.02567343711853,
|
|
"epoch": 0.19728622483075225,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 9.380572571499758e-05,
|
|
"loss": 5.7274,
|
|
"mean_token_accuracy": 0.1436111845076084,
|
|
"num_tokens": 6585489.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.0807962894439695,
|
|
"epoch": 0.19757677891739547,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 9.310976968607307e-05,
|
|
"loss": 5.7484,
|
|
"mean_token_accuracy": 0.1487852841615677,
|
|
"num_tokens": 6594848.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.089216613769532,
|
|
"epoch": 0.1978673330040387,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.241880026969381e-05,
|
|
"loss": 5.7464,
|
|
"mean_token_accuracy": 0.14339143857359887,
|
|
"num_tokens": 6605983.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.055511140823365,
|
|
"epoch": 0.19815788709068194,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 9.173283640913537e-05,
|
|
"loss": 5.6737,
|
|
"mean_token_accuracy": 0.1530997022986412,
|
|
"num_tokens": 6615093.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.013196659088135,
|
|
"epoch": 0.19844844117732516,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 9.10518969104436e-05,
|
|
"loss": 5.6165,
|
|
"mean_token_accuracy": 0.15827906131744385,
|
|
"num_tokens": 6623551.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.9676004409790036,
|
|
"epoch": 0.19873899526396838,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.037600044191868e-05,
|
|
"loss": 5.7311,
|
|
"mean_token_accuracy": 0.1509920448064804,
|
|
"num_tokens": 6633611.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.004412841796875,
|
|
"epoch": 0.19902954935061162,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 8.970516553360383e-05,
|
|
"loss": 5.6986,
|
|
"mean_token_accuracy": 0.1615770772099495,
|
|
"num_tokens": 6642605.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.040467405319214,
|
|
"epoch": 0.19932010343725484,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.903941057677692e-05,
|
|
"loss": 5.7086,
|
|
"mean_token_accuracy": 0.15024487525224686,
|
|
"num_tokens": 6652398.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.084765291213989,
|
|
"epoch": 0.19961065752389806,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.837875382344635e-05,
|
|
"loss": 5.6634,
|
|
"mean_token_accuracy": 0.15238699167966843,
|
|
"num_tokens": 6661667.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.060661029815674,
|
|
"epoch": 0.1999012116105413,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.772321338585076e-05,
|
|
"loss": 5.6415,
|
|
"mean_token_accuracy": 0.16024876087903978,
|
|
"num_tokens": 6670613.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.083744382858276,
|
|
"epoch": 0.20019176569718453,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 8.707280723596242e-05,
|
|
"loss": 5.7538,
|
|
"mean_token_accuracy": 0.1504388488829136,
|
|
"num_tokens": 6679543.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.077122068405151,
|
|
"epoch": 0.20048231978382777,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 8.64275532049944e-05,
|
|
"loss": 5.8425,
|
|
"mean_token_accuracy": 0.140785413980484,
|
|
"num_tokens": 6689638.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.026986122131348,
|
|
"epoch": 0.200772873870471,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 8.578746898291198e-05,
|
|
"loss": 5.7096,
|
|
"mean_token_accuracy": 0.15388644784688948,
|
|
"num_tokens": 6699561.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.036118078231811,
|
|
"epoch": 0.2010634279571142,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.515257211794742e-05,
|
|
"loss": 5.7424,
|
|
"mean_token_accuracy": 0.1555853232741356,
|
|
"num_tokens": 6709542.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.055869913101196,
|
|
"epoch": 0.20135398204375746,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 8.452288001611896e-05,
|
|
"loss": 5.6998,
|
|
"mean_token_accuracy": 0.15517012774944305,
|
|
"num_tokens": 6718545.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.0092888355255125,
|
|
"epoch": 0.20164453613040068,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 8.389840994075379e-05,
|
|
"loss": 5.5914,
|
|
"mean_token_accuracy": 0.15963410586118698,
|
|
"num_tokens": 6727491.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.133456993103027,
|
|
"epoch": 0.2019350902170439,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 8.327917901201435e-05,
|
|
"loss": 5.862,
|
|
"mean_token_accuracy": 0.1492785707116127,
|
|
"num_tokens": 6737690.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.023900270462036,
|
|
"epoch": 0.20222564430368714,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 8.266520420642931e-05,
|
|
"loss": 5.638,
|
|
"mean_token_accuracy": 0.16246868669986725,
|
|
"num_tokens": 6747049.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.069623184204102,
|
|
"epoch": 0.20251619839033036,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 8.205650235642828e-05,
|
|
"loss": 5.7306,
|
|
"mean_token_accuracy": 0.15011803209781646,
|
|
"num_tokens": 6756199.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.060231304168701,
|
|
"epoch": 0.20280675247697358,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.145309014987978e-05,
|
|
"loss": 5.6926,
|
|
"mean_token_accuracy": 0.14650730416178703,
|
|
"num_tokens": 6765595.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.029575967788697,
|
|
"epoch": 0.20309730656361683,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.085498412963437e-05,
|
|
"loss": 5.628,
|
|
"mean_token_accuracy": 0.16038369089365007,
|
|
"num_tokens": 6775078.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.112934684753418,
|
|
"epoch": 0.20338786065026004,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 8.026220069307078e-05,
|
|
"loss": 5.7506,
|
|
"mean_token_accuracy": 0.15270390585064889,
|
|
"num_tokens": 6785419.0,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1478035918356480.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|