2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0811249323958896,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 1000,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"entropy": 10.742581367492676,
|
||
|
|
"epoch": 0.005408328826392645,
|
||
|
|
"grad_norm": 5.4375,
|
||
|
|
"learning_rate": 2e-06,
|
||
|
|
"loss": 10.7836,
|
||
|
|
"mean_token_accuracy": 0.0,
|
||
|
|
"num_tokens": 10942.0,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.742576122283936,
|
||
|
|
"epoch": 0.01081665765278529,
|
||
|
|
"grad_norm": 5.0,
|
||
|
|
"learning_rate": 4.5e-06,
|
||
|
|
"loss": 10.7725,
|
||
|
|
"mean_token_accuracy": 0.0,
|
||
|
|
"num_tokens": 21887.0,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.742580318450928,
|
||
|
|
"epoch": 0.016224986479177934,
|
||
|
|
"grad_norm": 5.46875,
|
||
|
|
"learning_rate": 7e-06,
|
||
|
|
"loss": 10.7688,
|
||
|
|
"mean_token_accuracy": 0.0,
|
||
|
|
"num_tokens": 34345.0,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.74259843826294,
|
||
|
|
"epoch": 0.02163331530557058,
|
||
|
|
"grad_norm": 5.3125,
|
||
|
|
"learning_rate": 9.5e-06,
|
||
|
|
"loss": 10.6831,
|
||
|
|
"mean_token_accuracy": 0.0005558153847232461,
|
||
|
|
"num_tokens": 45445.0,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.742469692230225,
|
||
|
|
"epoch": 0.02704164413196322,
|
||
|
|
"grad_norm": 4.34375,
|
||
|
|
"learning_rate": 1.2e-05,
|
||
|
|
"loss": 10.5988,
|
||
|
|
"mean_token_accuracy": 0.007499191560782492,
|
||
|
|
"num_tokens": 58237.0,
|
||
|
|
"step": 25
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.742159843444824,
|
||
|
|
"epoch": 0.03244997295835587,
|
||
|
|
"grad_norm": 4.25,
|
||
|
|
"learning_rate": 1.4500000000000002e-05,
|
||
|
|
"loss": 10.4088,
|
||
|
|
"mean_token_accuracy": 0.031410138495266436,
|
||
|
|
"num_tokens": 68098.0,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.740957832336425,
|
||
|
|
"epoch": 0.03785830178474851,
|
||
|
|
"grad_norm": 3.203125,
|
||
|
|
"learning_rate": 1.7000000000000003e-05,
|
||
|
|
"loss": 10.2937,
|
||
|
|
"mean_token_accuracy": 0.04090985879302025,
|
||
|
|
"num_tokens": 79867.0,
|
||
|
|
"step": 35
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.738431835174561,
|
||
|
|
"epoch": 0.04326663061114116,
|
||
|
|
"grad_norm": 2.625,
|
||
|
|
"learning_rate": 1.95e-05,
|
||
|
|
"loss": 10.1592,
|
||
|
|
"mean_token_accuracy": 0.037697183154523375,
|
||
|
|
"num_tokens": 91257.0,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.735397911071777,
|
||
|
|
"epoch": 0.048674959437533805,
|
||
|
|
"grad_norm": 2.1875,
|
||
|
|
"learning_rate": 2.2e-05,
|
||
|
|
"loss": 10.0705,
|
||
|
|
"mean_token_accuracy": 0.04106438439339399,
|
||
|
|
"num_tokens": 103474.0,
|
||
|
|
"step": 45
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.733191680908202,
|
||
|
|
"epoch": 0.05408328826392644,
|
||
|
|
"grad_norm": 2.28125,
|
||
|
|
"learning_rate": 2.4500000000000003e-05,
|
||
|
|
"loss": 9.9433,
|
||
|
|
"mean_token_accuracy": 0.037430111691355704,
|
||
|
|
"num_tokens": 114830.0,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.732486057281495,
|
||
|
|
"epoch": 0.05949161709031909,
|
||
|
|
"grad_norm": 2.046875,
|
||
|
|
"learning_rate": 2.7e-05,
|
||
|
|
"loss": 9.8905,
|
||
|
|
"mean_token_accuracy": 0.036898509971797465,
|
||
|
|
"num_tokens": 126017.0,
|
||
|
|
"step": 55
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.73141107559204,
|
||
|
|
"epoch": 0.06489994591671173,
|
||
|
|
"grad_norm": 2.03125,
|
||
|
|
"learning_rate": 2.95e-05,
|
||
|
|
"loss": 9.8185,
|
||
|
|
"mean_token_accuracy": 0.04504128023982048,
|
||
|
|
"num_tokens": 138509.0,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.729285430908202,
|
||
|
|
"epoch": 0.07030827474310439,
|
||
|
|
"grad_norm": 1.96875,
|
||
|
|
"learning_rate": 3.2e-05,
|
||
|
|
"loss": 9.764,
|
||
|
|
"mean_token_accuracy": 0.0458514004945755,
|
||
|
|
"num_tokens": 148353.0,
|
||
|
|
"step": 65
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.726380920410156,
|
||
|
|
"epoch": 0.07571660356949703,
|
||
|
|
"grad_norm": 1.9140625,
|
||
|
|
"learning_rate": 3.4500000000000005e-05,
|
||
|
|
"loss": 9.7481,
|
||
|
|
"mean_token_accuracy": 0.04072931818664074,
|
||
|
|
"num_tokens": 160269.0,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.723598289489747,
|
||
|
|
"epoch": 0.08112493239588967,
|
||
|
|
"grad_norm": 2.0,
|
||
|
|
"learning_rate": 3.7e-05,
|
||
|
|
"loss": 9.6457,
|
||
|
|
"mean_token_accuracy": 0.040256797894835473,
|
||
|
|
"num_tokens": 171568.0,
|
||
|
|
"step": 75
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.720184516906738,
|
||
|
|
"epoch": 0.08653326122228232,
|
||
|
|
"grad_norm": 2.046875,
|
||
|
|
"learning_rate": 3.95e-05,
|
||
|
|
"loss": 9.5709,
|
||
|
|
"mean_token_accuracy": 0.04319700971245766,
|
||
|
|
"num_tokens": 184243.0,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.715407657623292,
|
||
|
|
"epoch": 0.09194159004867496,
|
||
|
|
"grad_norm": 2.0,
|
||
|
|
"learning_rate": 4.2000000000000004e-05,
|
||
|
|
"loss": 9.5081,
|
||
|
|
"mean_token_accuracy": 0.044286393001675604,
|
||
|
|
"num_tokens": 194887.0,
|
||
|
|
"step": 85
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.70880765914917,
|
||
|
|
"epoch": 0.09734991887506761,
|
||
|
|
"grad_norm": 1.9453125,
|
||
|
|
"learning_rate": 4.45e-05,
|
||
|
|
"loss": 9.4386,
|
||
|
|
"mean_token_accuracy": 0.04779103547334671,
|
||
|
|
"num_tokens": 206476.0,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.699271774291992,
|
||
|
|
"epoch": 0.10275824770146025,
|
||
|
|
"grad_norm": 1.9453125,
|
||
|
|
"learning_rate": 4.7000000000000004e-05,
|
||
|
|
"loss": 9.3288,
|
||
|
|
"mean_token_accuracy": 0.052064205333590505,
|
||
|
|
"num_tokens": 216578.0,
|
||
|
|
"step": 95
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.6855544090271,
|
||
|
|
"epoch": 0.10816657652785289,
|
||
|
|
"grad_norm": 1.90625,
|
||
|
|
"learning_rate": 4.9500000000000004e-05,
|
||
|
|
"loss": 9.2367,
|
||
|
|
"mean_token_accuracy": 0.05270014926791191,
|
||
|
|
"num_tokens": 226941.0,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.670534229278564,
|
||
|
|
"epoch": 0.11357490535424554,
|
||
|
|
"grad_norm": 1.8828125,
|
||
|
|
"learning_rate": 5.2e-05,
|
||
|
|
"loss": 9.1416,
|
||
|
|
"mean_token_accuracy": 0.05506536141037941,
|
||
|
|
"num_tokens": 238060.0,
|
||
|
|
"step": 105
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.650165462493897,
|
||
|
|
"epoch": 0.11898323418063818,
|
||
|
|
"grad_norm": 2.015625,
|
||
|
|
"learning_rate": 5.45e-05,
|
||
|
|
"loss": 8.998,
|
||
|
|
"mean_token_accuracy": 0.06617085449397564,
|
||
|
|
"num_tokens": 249137.0,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.614893913269043,
|
||
|
|
"epoch": 0.12439156300703083,
|
||
|
|
"grad_norm": 1.8984375,
|
||
|
|
"learning_rate": 5.7e-05,
|
||
|
|
"loss": 8.9011,
|
||
|
|
"mean_token_accuracy": 0.0645027995109558,
|
||
|
|
"num_tokens": 260712.0,
|
||
|
|
"step": 115
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.56815938949585,
|
||
|
|
"epoch": 0.12979989183342347,
|
||
|
|
"grad_norm": 1.921875,
|
||
|
|
"learning_rate": 5.9499999999999996e-05,
|
||
|
|
"loss": 8.7629,
|
||
|
|
"mean_token_accuracy": 0.06179735250771046,
|
||
|
|
"num_tokens": 271039.0,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.511663913726807,
|
||
|
|
"epoch": 0.1352082206598161,
|
||
|
|
"grad_norm": 1.984375,
|
||
|
|
"learning_rate": 6.2e-05,
|
||
|
|
"loss": 8.6203,
|
||
|
|
"mean_token_accuracy": 0.0636001892387867,
|
||
|
|
"num_tokens": 281353.0,
|
||
|
|
"step": 125
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.432361125946045,
|
||
|
|
"epoch": 0.14061654948620878,
|
||
|
|
"grad_norm": 1.671875,
|
||
|
|
"learning_rate": 6.450000000000001e-05,
|
||
|
|
"loss": 8.5429,
|
||
|
|
"mean_token_accuracy": 0.07175168134272099,
|
||
|
|
"num_tokens": 292221.0,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.352851963043213,
|
||
|
|
"epoch": 0.1460248783126014,
|
||
|
|
"grad_norm": 1.671875,
|
||
|
|
"learning_rate": 6.7e-05,
|
||
|
|
"loss": 8.4081,
|
||
|
|
"mean_token_accuracy": 0.06957007274031639,
|
||
|
|
"num_tokens": 303353.0,
|
||
|
|
"step": 135
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.268811798095703,
|
||
|
|
"epoch": 0.15143320713899405,
|
||
|
|
"grad_norm": 1.5703125,
|
||
|
|
"learning_rate": 6.950000000000001e-05,
|
||
|
|
"loss": 8.292,
|
||
|
|
"mean_token_accuracy": 0.07166561894118786,
|
||
|
|
"num_tokens": 314165.0,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.154801845550537,
|
||
|
|
"epoch": 0.1568415359653867,
|
||
|
|
"grad_norm": 1.5234375,
|
||
|
|
"learning_rate": 7.2e-05,
|
||
|
|
"loss": 8.1152,
|
||
|
|
"mean_token_accuracy": 0.06947772391140461,
|
||
|
|
"num_tokens": 326387.0,
|
||
|
|
"step": 145
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.024849987030029,
|
||
|
|
"epoch": 0.16224986479177933,
|
||
|
|
"grad_norm": 1.46875,
|
||
|
|
"learning_rate": 7.45e-05,
|
||
|
|
"loss": 8.0602,
|
||
|
|
"mean_token_accuracy": 0.06997468769550323,
|
||
|
|
"num_tokens": 339155.0,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.84305181503296,
|
||
|
|
"epoch": 0.167658193618172,
|
||
|
|
"grad_norm": 1.609375,
|
||
|
|
"learning_rate": 7.7e-05,
|
||
|
|
"loss": 7.9106,
|
||
|
|
"mean_token_accuracy": 0.07246845588088036,
|
||
|
|
"num_tokens": 350643.0,
|
||
|
|
"step": 155
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.628917694091797,
|
||
|
|
"epoch": 0.17306652244456464,
|
||
|
|
"grad_norm": 1.296875,
|
||
|
|
"learning_rate": 7.950000000000001e-05,
|
||
|
|
"loss": 7.7825,
|
||
|
|
"mean_token_accuracy": 0.07223140001296997,
|
||
|
|
"num_tokens": 361977.0,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.454836750030518,
|
||
|
|
"epoch": 0.17847485127095727,
|
||
|
|
"grad_norm": 1.359375,
|
||
|
|
"learning_rate": 8.2e-05,
|
||
|
|
"loss": 7.7462,
|
||
|
|
"mean_token_accuracy": 0.07521784901618958,
|
||
|
|
"num_tokens": 373604.0,
|
||
|
|
"step": 165
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.20318956375122,
|
||
|
|
"epoch": 0.1838831800973499,
|
||
|
|
"grad_norm": 0.98046875,
|
||
|
|
"learning_rate": 8.450000000000001e-05,
|
||
|
|
"loss": 7.6587,
|
||
|
|
"mean_token_accuracy": 0.07736964598298073,
|
||
|
|
"num_tokens": 384306.0,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.976179218292236,
|
||
|
|
"epoch": 0.18929150892374255,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 8.7e-05,
|
||
|
|
"loss": 7.5228,
|
||
|
|
"mean_token_accuracy": 0.08294685482978821,
|
||
|
|
"num_tokens": 396479.0,
|
||
|
|
"step": 175
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.758621406555175,
|
||
|
|
"epoch": 0.19469983775013522,
|
||
|
|
"grad_norm": 0.95703125,
|
||
|
|
"learning_rate": 8.95e-05,
|
||
|
|
"loss": 7.4294,
|
||
|
|
"mean_token_accuracy": 0.08506134003400803,
|
||
|
|
"num_tokens": 408489.0,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.530921840667725,
|
||
|
|
"epoch": 0.20010816657652786,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 9.2e-05,
|
||
|
|
"loss": 7.3995,
|
||
|
|
"mean_token_accuracy": 0.08729378059506417,
|
||
|
|
"num_tokens": 419339.0,
|
||
|
|
"step": 185
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.382069301605224,
|
||
|
|
"epoch": 0.2055164954029205,
|
||
|
|
"grad_norm": 0.9296875,
|
||
|
|
"learning_rate": 9.45e-05,
|
||
|
|
"loss": 7.3478,
|
||
|
|
"mean_token_accuracy": 0.08495231121778488,
|
||
|
|
"num_tokens": 430486.0,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.212819290161132,
|
||
|
|
"epoch": 0.21092482422931313,
|
||
|
|
"grad_norm": 0.92578125,
|
||
|
|
"learning_rate": 9.7e-05,
|
||
|
|
"loss": 7.3377,
|
||
|
|
"mean_token_accuracy": 0.08699661865830421,
|
||
|
|
"num_tokens": 440414.0,
|
||
|
|
"step": 195
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.141395664215088,
|
||
|
|
"epoch": 0.21633315305570577,
|
||
|
|
"grad_norm": 0.84375,
|
||
|
|
"learning_rate": 9.95e-05,
|
||
|
|
"loss": 7.2746,
|
||
|
|
"mean_token_accuracy": 0.0853013277053833,
|
||
|
|
"num_tokens": 451475.0,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.020270109176636,
|
||
|
|
"epoch": 0.22174148188209844,
|
||
|
|
"grad_norm": 0.86328125,
|
||
|
|
"learning_rate": 0.000102,
|
||
|
|
"loss": 7.3305,
|
||
|
|
"mean_token_accuracy": 0.08908139243721962,
|
||
|
|
"num_tokens": 462053.0,
|
||
|
|
"step": 205
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.034120750427245,
|
||
|
|
"epoch": 0.22714981070849108,
|
||
|
|
"grad_norm": 0.9609375,
|
||
|
|
"learning_rate": 0.00010449999999999999,
|
||
|
|
"loss": 7.2768,
|
||
|
|
"mean_token_accuracy": 0.08668714761734009,
|
||
|
|
"num_tokens": 473207.0,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.990876579284668,
|
||
|
|
"epoch": 0.23255813953488372,
|
||
|
|
"grad_norm": 0.9140625,
|
||
|
|
"learning_rate": 0.000107,
|
||
|
|
"loss": 7.2574,
|
||
|
|
"mean_token_accuracy": 0.08839641213417053,
|
||
|
|
"num_tokens": 484999.0,
|
||
|
|
"step": 215
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.926925563812256,
|
||
|
|
"epoch": 0.23796646836127636,
|
||
|
|
"grad_norm": 0.9453125,
|
||
|
|
"learning_rate": 0.0001095,
|
||
|
|
"loss": 7.2424,
|
||
|
|
"mean_token_accuracy": 0.08917610868811607,
|
||
|
|
"num_tokens": 496772.0,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.89932656288147,
|
||
|
|
"epoch": 0.24337479718766902,
|
||
|
|
"grad_norm": 1.0,
|
||
|
|
"learning_rate": 0.000112,
|
||
|
|
"loss": 7.222,
|
||
|
|
"mean_token_accuracy": 0.08830738216638565,
|
||
|
|
"num_tokens": 507782.0,
|
||
|
|
"step": 225
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.855933761596679,
|
||
|
|
"epoch": 0.24878312601406166,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.0001145,
|
||
|
|
"loss": 7.1208,
|
||
|
|
"mean_token_accuracy": 0.09142142534255981,
|
||
|
|
"num_tokens": 518950.0,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.874954509735107,
|
||
|
|
"epoch": 0.2541914548404543,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 0.00011700000000000001,
|
||
|
|
"loss": 7.1844,
|
||
|
|
"mean_token_accuracy": 0.09523893147706985,
|
||
|
|
"num_tokens": 530795.0,
|
||
|
|
"step": 235
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.853066539764404,
|
||
|
|
"epoch": 0.25959978366684694,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.00011949999999999999,
|
||
|
|
"loss": 7.202,
|
||
|
|
"mean_token_accuracy": 0.09066410064697265,
|
||
|
|
"num_tokens": 542878.0,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.815495109558105,
|
||
|
|
"epoch": 0.2650081124932396,
|
||
|
|
"grad_norm": 0.93359375,
|
||
|
|
"learning_rate": 0.000122,
|
||
|
|
"loss": 7.0976,
|
||
|
|
"mean_token_accuracy": 0.09599302783608436,
|
||
|
|
"num_tokens": 553682.0,
|
||
|
|
"step": 245
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.718054342269897,
|
||
|
|
"epoch": 0.2704164413196322,
|
||
|
|
"grad_norm": 0.9453125,
|
||
|
|
"learning_rate": 0.0001245,
|
||
|
|
"loss": 7.0854,
|
||
|
|
"mean_token_accuracy": 0.10266833081841469,
|
||
|
|
"num_tokens": 564433.0,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.75045108795166,
|
||
|
|
"epoch": 0.27582477014602486,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.000127,
|
||
|
|
"loss": 7.1247,
|
||
|
|
"mean_token_accuracy": 0.09944702386856079,
|
||
|
|
"num_tokens": 576001.0,
|
||
|
|
"step": 255
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.712650632858276,
|
||
|
|
"epoch": 0.28123309897241755,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.0001295,
|
||
|
|
"loss": 7.0446,
|
||
|
|
"mean_token_accuracy": 0.10567303821444511,
|
||
|
|
"num_tokens": 587138.0,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.598682832717896,
|
||
|
|
"epoch": 0.2866414277988102,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.000132,
|
||
|
|
"loss": 7.0066,
|
||
|
|
"mean_token_accuracy": 0.10336644649505615,
|
||
|
|
"num_tokens": 597572.0,
|
||
|
|
"step": 265
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.61051869392395,
|
||
|
|
"epoch": 0.2920497566252028,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.00013450000000000002,
|
||
|
|
"loss": 6.9642,
|
||
|
|
"mean_token_accuracy": 0.10717165470123291,
|
||
|
|
"num_tokens": 608879.0,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.656619739532471,
|
||
|
|
"epoch": 0.29745808545159547,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.00013700000000000002,
|
||
|
|
"loss": 7.0022,
|
||
|
|
"mean_token_accuracy": 0.09820161312818527,
|
||
|
|
"num_tokens": 620439.0,
|
||
|
|
"step": 275
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.668045330047607,
|
||
|
|
"epoch": 0.3028664142779881,
|
||
|
|
"grad_norm": 0.9453125,
|
||
|
|
"learning_rate": 0.0001395,
|
||
|
|
"loss": 7.0632,
|
||
|
|
"mean_token_accuracy": 0.10111433267593384,
|
||
|
|
"num_tokens": 632543.0,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.635419225692749,
|
||
|
|
"epoch": 0.30827474310438074,
|
||
|
|
"grad_norm": 0.94140625,
|
||
|
|
"learning_rate": 0.00014199999999999998,
|
||
|
|
"loss": 7.1026,
|
||
|
|
"mean_token_accuracy": 0.10522567853331566,
|
||
|
|
"num_tokens": 643369.0,
|
||
|
|
"step": 285
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.584413766860962,
|
||
|
|
"epoch": 0.3136830719307734,
|
||
|
|
"grad_norm": 0.86328125,
|
||
|
|
"learning_rate": 0.0001445,
|
||
|
|
"loss": 6.9646,
|
||
|
|
"mean_token_accuracy": 0.1047127217054367,
|
||
|
|
"num_tokens": 655328.0,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.67089467048645,
|
||
|
|
"epoch": 0.319091400757166,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.000147,
|
||
|
|
"loss": 7.0248,
|
||
|
|
"mean_token_accuracy": 0.10310934484004974,
|
||
|
|
"num_tokens": 665780.0,
|
||
|
|
"step": 295
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.552512168884277,
|
||
|
|
"epoch": 0.32449972958355866,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.0001495,
|
||
|
|
"loss": 6.9293,
|
||
|
|
"mean_token_accuracy": 0.10721782594919205,
|
||
|
|
"num_tokens": 676566.0,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.559945297241211,
|
||
|
|
"epoch": 0.3299080584099513,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.000152,
|
||
|
|
"loss": 6.9251,
|
||
|
|
"mean_token_accuracy": 0.1060689315199852,
|
||
|
|
"num_tokens": 687407.0,
|
||
|
|
"step": 305
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.49097957611084,
|
||
|
|
"epoch": 0.335316387236344,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 0.00015450000000000001,
|
||
|
|
"loss": 6.8497,
|
||
|
|
"mean_token_accuracy": 0.11217133775353431,
|
||
|
|
"num_tokens": 698162.0,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.46426739692688,
|
||
|
|
"epoch": 0.34072471606273663,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.000157,
|
||
|
|
"loss": 6.8316,
|
||
|
|
"mean_token_accuracy": 0.10524874627590179,
|
||
|
|
"num_tokens": 708743.0,
|
||
|
|
"step": 315
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.558863353729248,
|
||
|
|
"epoch": 0.34613304488912927,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.0001595,
|
||
|
|
"loss": 7.0442,
|
||
|
|
"mean_token_accuracy": 0.11182164400815964,
|
||
|
|
"num_tokens": 719828.0,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.583437824249268,
|
||
|
|
"epoch": 0.3515413737155219,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.000162,
|
||
|
|
"loss": 6.9937,
|
||
|
|
"mean_token_accuracy": 0.1039721980690956,
|
||
|
|
"num_tokens": 732514.0,
|
||
|
|
"step": 325
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.449002027511597,
|
||
|
|
"epoch": 0.35694970254191455,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.00016450000000000001,
|
||
|
|
"loss": 6.8317,
|
||
|
|
"mean_token_accuracy": 0.11070109233260154,
|
||
|
|
"num_tokens": 742391.0,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.36940336227417,
|
||
|
|
"epoch": 0.3623580313683072,
|
||
|
|
"grad_norm": 1.1875,
|
||
|
|
"learning_rate": 0.00016700000000000002,
|
||
|
|
"loss": 6.8335,
|
||
|
|
"mean_token_accuracy": 0.10628218725323677,
|
||
|
|
"num_tokens": 753638.0,
|
||
|
|
"step": 335
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.5532920360565186,
|
||
|
|
"epoch": 0.3677663601946998,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.00016950000000000003,
|
||
|
|
"loss": 7.0485,
|
||
|
|
"mean_token_accuracy": 0.10574989169836044,
|
||
|
|
"num_tokens": 764850.0,
|
||
|
|
"step": 340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.4345468044281,
|
||
|
|
"epoch": 0.37317468902109246,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.00017199999999999998,
|
||
|
|
"loss": 6.8115,
|
||
|
|
"mean_token_accuracy": 0.11775769889354706,
|
||
|
|
"num_tokens": 776192.0,
|
||
|
|
"step": 345
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.336619281768799,
|
||
|
|
"epoch": 0.3785830178474851,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.00017449999999999999,
|
||
|
|
"loss": 6.7691,
|
||
|
|
"mean_token_accuracy": 0.11614358723163605,
|
||
|
|
"num_tokens": 787051.0,
|
||
|
|
"step": 350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.326699829101562,
|
||
|
|
"epoch": 0.3839913466738778,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.000177,
|
||
|
|
"loss": 6.7934,
|
||
|
|
"mean_token_accuracy": 0.11272966936230659,
|
||
|
|
"num_tokens": 799281.0,
|
||
|
|
"step": 355
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.331569719314575,
|
||
|
|
"epoch": 0.38939967550027044,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.0001795,
|
||
|
|
"loss": 6.813,
|
||
|
|
"mean_token_accuracy": 0.11107392236590385,
|
||
|
|
"num_tokens": 810003.0,
|
||
|
|
"step": 360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.336561441421509,
|
||
|
|
"epoch": 0.3948080043266631,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.000182,
|
||
|
|
"loss": 6.7821,
|
||
|
|
"mean_token_accuracy": 0.11116143241524697,
|
||
|
|
"num_tokens": 819749.0,
|
||
|
|
"step": 365
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.329193019866944,
|
||
|
|
"epoch": 0.4002163331530557,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.0001845,
|
||
|
|
"loss": 6.7888,
|
||
|
|
"mean_token_accuracy": 0.11996297091245652,
|
||
|
|
"num_tokens": 830552.0,
|
||
|
|
"step": 370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.390970516204834,
|
||
|
|
"epoch": 0.40562466197944835,
|
||
|
|
"grad_norm": 1.3125,
|
||
|
|
"learning_rate": 0.000187,
|
||
|
|
"loss": 6.8901,
|
||
|
|
"mean_token_accuracy": 0.1121357448399067,
|
||
|
|
"num_tokens": 840495.0,
|
||
|
|
"step": 375
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.275109004974365,
|
||
|
|
"epoch": 0.411032990805841,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.0001895,
|
||
|
|
"loss": 6.8114,
|
||
|
|
"mean_token_accuracy": 0.1169951744377613,
|
||
|
|
"num_tokens": 852915.0,
|
||
|
|
"step": 380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.377492761611938,
|
||
|
|
"epoch": 0.41644131963223363,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 0.000192,
|
||
|
|
"loss": 6.7822,
|
||
|
|
"mean_token_accuracy": 0.11362324208021164,
|
||
|
|
"num_tokens": 862762.0,
|
||
|
|
"step": 385
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.277709007263184,
|
||
|
|
"epoch": 0.42184964845862627,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.0001945,
|
||
|
|
"loss": 6.7274,
|
||
|
|
"mean_token_accuracy": 0.11747928410768509,
|
||
|
|
"num_tokens": 873023.0,
|
||
|
|
"step": 390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.28900933265686,
|
||
|
|
"epoch": 0.4272579772850189,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.00019700000000000002,
|
||
|
|
"loss": 6.8274,
|
||
|
|
"mean_token_accuracy": 0.10971275642514229,
|
||
|
|
"num_tokens": 885281.0,
|
||
|
|
"step": 395
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.2772074222564695,
|
||
|
|
"epoch": 0.43266630611141155,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.00019950000000000002,
|
||
|
|
"loss": 6.8206,
|
||
|
|
"mean_token_accuracy": 0.11502003371715545,
|
||
|
|
"num_tokens": 896210.0,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.189112377166748,
|
||
|
|
"epoch": 0.43807463493780424,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.000202,
|
||
|
|
"loss": 6.7468,
|
||
|
|
"mean_token_accuracy": 0.11514622867107391,
|
||
|
|
"num_tokens": 909028.0,
|
||
|
|
"step": 405
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.229736566543579,
|
||
|
|
"epoch": 0.4434829637641969,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.00020449999999999998,
|
||
|
|
"loss": 6.6815,
|
||
|
|
"mean_token_accuracy": 0.11853564456105233,
|
||
|
|
"num_tokens": 919710.0,
|
||
|
|
"step": 410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.304251861572266,
|
||
|
|
"epoch": 0.4488912925905895,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.000207,
|
||
|
|
"loss": 6.736,
|
||
|
|
"mean_token_accuracy": 0.11397984251379967,
|
||
|
|
"num_tokens": 930019.0,
|
||
|
|
"step": 415
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.166671371459961,
|
||
|
|
"epoch": 0.45429962141698216,
|
||
|
|
"grad_norm": 0.99609375,
|
||
|
|
"learning_rate": 0.0002095,
|
||
|
|
"loss": 6.6524,
|
||
|
|
"mean_token_accuracy": 0.11951824501156808,
|
||
|
|
"num_tokens": 940535.0,
|
||
|
|
"step": 420
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.0795973777771,
|
||
|
|
"epoch": 0.4597079502433748,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.000212,
|
||
|
|
"loss": 6.6119,
|
||
|
|
"mean_token_accuracy": 0.12140549421310425,
|
||
|
|
"num_tokens": 951390.0,
|
||
|
|
"step": 425
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.141013669967651,
|
||
|
|
"epoch": 0.46511627906976744,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.0002145,
|
||
|
|
"loss": 6.6636,
|
||
|
|
"mean_token_accuracy": 0.1212614968419075,
|
||
|
|
"num_tokens": 963010.0,
|
||
|
|
"step": 430
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.112875986099243,
|
||
|
|
"epoch": 0.4705246078961601,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.00021700000000000002,
|
||
|
|
"loss": 6.6152,
|
||
|
|
"mean_token_accuracy": 0.12599240839481354,
|
||
|
|
"num_tokens": 974032.0,
|
||
|
|
"step": 435
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.222216367721558,
|
||
|
|
"epoch": 0.4759329367225527,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 0.0002195,
|
||
|
|
"loss": 6.6586,
|
||
|
|
"mean_token_accuracy": 0.11954770833253861,
|
||
|
|
"num_tokens": 983954.0,
|
||
|
|
"step": 440
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.11478705406189,
|
||
|
|
"epoch": 0.48134126554894535,
|
||
|
|
"grad_norm": 1.2421875,
|
||
|
|
"learning_rate": 0.000222,
|
||
|
|
"loss": 6.689,
|
||
|
|
"mean_token_accuracy": 0.12196131646633149,
|
||
|
|
"num_tokens": 994135.0,
|
||
|
|
"step": 445
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.173604154586792,
|
||
|
|
"epoch": 0.48674959437533805,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.0002245,
|
||
|
|
"loss": 6.6493,
|
||
|
|
"mean_token_accuracy": 0.12306504994630814,
|
||
|
|
"num_tokens": 1006275.0,
|
||
|
|
"step": 450
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.988444137573242,
|
||
|
|
"epoch": 0.4921579232017307,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.00022700000000000002,
|
||
|
|
"loss": 6.4801,
|
||
|
|
"mean_token_accuracy": 0.129085473716259,
|
||
|
|
"num_tokens": 1017767.0,
|
||
|
|
"step": 455
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.0636707782745365,
|
||
|
|
"epoch": 0.4975662520281233,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.00022950000000000002,
|
||
|
|
"loss": 6.6038,
|
||
|
|
"mean_token_accuracy": 0.1241903193295002,
|
||
|
|
"num_tokens": 1028286.0,
|
||
|
|
"step": 460
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.059632682800293,
|
||
|
|
"epoch": 0.502974580854516,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.00023200000000000003,
|
||
|
|
"loss": 6.6151,
|
||
|
|
"mean_token_accuracy": 0.12029262557625771,
|
||
|
|
"num_tokens": 1040555.0,
|
||
|
|
"step": 465
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.171442461013794,
|
||
|
|
"epoch": 0.5083829096809086,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 0.00023449999999999998,
|
||
|
|
"loss": 6.7175,
|
||
|
|
"mean_token_accuracy": 0.1209532730281353,
|
||
|
|
"num_tokens": 1053668.0,
|
||
|
|
"step": 470
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.12706880569458,
|
||
|
|
"epoch": 0.5137912385073012,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.000237,
|
||
|
|
"loss": 6.6597,
|
||
|
|
"mean_token_accuracy": 0.11960283890366555,
|
||
|
|
"num_tokens": 1066042.0,
|
||
|
|
"step": 475
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.984622383117676,
|
||
|
|
"epoch": 0.5191995673336939,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.0002395,
|
||
|
|
"loss": 6.6701,
|
||
|
|
"mean_token_accuracy": 0.12059814557433128,
|
||
|
|
"num_tokens": 1077843.0,
|
||
|
|
"step": 480
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.047416830062867,
|
||
|
|
"epoch": 0.5246078961600865,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.000242,
|
||
|
|
"loss": 6.5605,
|
||
|
|
"mean_token_accuracy": 0.1252925641834736,
|
||
|
|
"num_tokens": 1089448.0,
|
||
|
|
"step": 485
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.041959238052368,
|
||
|
|
"epoch": 0.5300162249864792,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.0002445,
|
||
|
|
"loss": 6.6615,
|
||
|
|
"mean_token_accuracy": 0.12150585278868675,
|
||
|
|
"num_tokens": 1100989.0,
|
||
|
|
"step": 490
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.162778997421265,
|
||
|
|
"epoch": 0.5354245538128718,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.000247,
|
||
|
|
"loss": 6.7004,
|
||
|
|
"mean_token_accuracy": 0.1256784491240978,
|
||
|
|
"num_tokens": 1113012.0,
|
||
|
|
"step": 495
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.056042337417603,
|
||
|
|
"epoch": 0.5408328826392644,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.0002495,
|
||
|
|
"loss": 6.6356,
|
||
|
|
"mean_token_accuracy": 0.12231024205684662,
|
||
|
|
"num_tokens": 1123551.0,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5408328826392644,
|
||
|
|
"eval_entropy": 7.0043568221005525,
|
||
|
|
"eval_loss": 6.780564308166504,
|
||
|
|
"eval_mean_token_accuracy": 0.12075253836810589,
|
||
|
|
"eval_num_tokens": 1123551.0,
|
||
|
|
"eval_runtime": 1.9941,
|
||
|
|
"eval_samples_per_second": 1762.721,
|
||
|
|
"eval_steps_per_second": 220.654,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.074879932403564,
|
||
|
|
"epoch": 0.5462412114656571,
|
||
|
|
"grad_norm": 1.359375,
|
||
|
|
"learning_rate": 0.000252,
|
||
|
|
"loss": 6.5659,
|
||
|
|
"mean_token_accuracy": 0.12208069637417793,
|
||
|
|
"num_tokens": 1133838.0,
|
||
|
|
"step": 505
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.973394107818604,
|
||
|
|
"epoch": 0.5516495402920497,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.0002545,
|
||
|
|
"loss": 6.6027,
|
||
|
|
"mean_token_accuracy": 0.12259892076253891,
|
||
|
|
"num_tokens": 1146005.0,
|
||
|
|
"step": 510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.00295934677124,
|
||
|
|
"epoch": 0.5570578691184424,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000257,
|
||
|
|
"loss": 6.6885,
|
||
|
|
"mean_token_accuracy": 0.1244613103568554,
|
||
|
|
"num_tokens": 1157328.0,
|
||
|
|
"step": 515
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.057001447677612,
|
||
|
|
"epoch": 0.5624661979448351,
|
||
|
|
"grad_norm": 0.98046875,
|
||
|
|
"learning_rate": 0.0002595,
|
||
|
|
"loss": 6.4842,
|
||
|
|
"mean_token_accuracy": 0.1322474293410778,
|
||
|
|
"num_tokens": 1168592.0,
|
||
|
|
"step": 520
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.032931184768676,
|
||
|
|
"epoch": 0.5678745267712277,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000262,
|
||
|
|
"loss": 6.6161,
|
||
|
|
"mean_token_accuracy": 0.12481116801500321,
|
||
|
|
"num_tokens": 1178680.0,
|
||
|
|
"step": 525
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.9781468391418455,
|
||
|
|
"epoch": 0.5732828555976204,
|
||
|
|
"grad_norm": 1.1640625,
|
||
|
|
"learning_rate": 0.00026450000000000003,
|
||
|
|
"loss": 6.5535,
|
||
|
|
"mean_token_accuracy": 0.127150696516037,
|
||
|
|
"num_tokens": 1188825.0,
|
||
|
|
"step": 530
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.008031415939331,
|
||
|
|
"epoch": 0.578691184424013,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.00026700000000000004,
|
||
|
|
"loss": 6.5956,
|
||
|
|
"mean_token_accuracy": 0.12398400753736497,
|
||
|
|
"num_tokens": 1201375.0,
|
||
|
|
"step": 535
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.072990751266479,
|
||
|
|
"epoch": 0.5840995132504057,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.00026950000000000005,
|
||
|
|
"loss": 6.7227,
|
||
|
|
"mean_token_accuracy": 0.11844412088394166,
|
||
|
|
"num_tokens": 1211915.0,
|
||
|
|
"step": 540
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.957782316207886,
|
||
|
|
"epoch": 0.5895078420767983,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.00027200000000000005,
|
||
|
|
"loss": 6.6289,
|
||
|
|
"mean_token_accuracy": 0.12811450883746148,
|
||
|
|
"num_tokens": 1223151.0,
|
||
|
|
"step": 545
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.041991472244263,
|
||
|
|
"epoch": 0.5949161709031909,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.0002745,
|
||
|
|
"loss": 6.5414,
|
||
|
|
"mean_token_accuracy": 0.13184184953570366,
|
||
|
|
"num_tokens": 1234229.0,
|
||
|
|
"step": 550
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.023768568038941,
|
||
|
|
"epoch": 0.6003244997295836,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.000277,
|
||
|
|
"loss": 6.5529,
|
||
|
|
"mean_token_accuracy": 0.13157471716403962,
|
||
|
|
"num_tokens": 1245298.0,
|
||
|
|
"step": 555
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.8409104347229,
|
||
|
|
"epoch": 0.6057328285559762,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.0002795,
|
||
|
|
"loss": 6.5595,
|
||
|
|
"mean_token_accuracy": 0.12466495037078858,
|
||
|
|
"num_tokens": 1256154.0,
|
||
|
|
"step": 560
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.922661924362183,
|
||
|
|
"epoch": 0.6111411573823688,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.00028199999999999997,
|
||
|
|
"loss": 6.4705,
|
||
|
|
"mean_token_accuracy": 0.129321701079607,
|
||
|
|
"num_tokens": 1267404.0,
|
||
|
|
"step": 565
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.944332456588745,
|
||
|
|
"epoch": 0.6165494862087615,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.0002845,
|
||
|
|
"loss": 6.5185,
|
||
|
|
"mean_token_accuracy": 0.12794106900691987,
|
||
|
|
"num_tokens": 1278894.0,
|
||
|
|
"step": 570
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.9090385913848875,
|
||
|
|
"epoch": 0.6219578150351541,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000287,
|
||
|
|
"loss": 6.502,
|
||
|
|
"mean_token_accuracy": 0.13496915996074677,
|
||
|
|
"num_tokens": 1290408.0,
|
||
|
|
"step": 575
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.971935558319092,
|
||
|
|
"epoch": 0.6273661438615468,
|
||
|
|
"grad_norm": 1.1875,
|
||
|
|
"learning_rate": 0.0002895,
|
||
|
|
"loss": 6.5806,
|
||
|
|
"mean_token_accuracy": 0.12681553587317468,
|
||
|
|
"num_tokens": 1301508.0,
|
||
|
|
"step": 580
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.989264011383057,
|
||
|
|
"epoch": 0.6327744726879394,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.000292,
|
||
|
|
"loss": 6.5447,
|
||
|
|
"mean_token_accuracy": 0.13189456537365912,
|
||
|
|
"num_tokens": 1312551.0,
|
||
|
|
"step": 585
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.965842962265015,
|
||
|
|
"epoch": 0.638182801514332,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.0002945,
|
||
|
|
"loss": 6.6459,
|
||
|
|
"mean_token_accuracy": 0.12647571191191673,
|
||
|
|
"num_tokens": 1324876.0,
|
||
|
|
"step": 590
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.939603042602539,
|
||
|
|
"epoch": 0.6435911303407247,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.000297,
|
||
|
|
"loss": 6.531,
|
||
|
|
"mean_token_accuracy": 0.12691377103328705,
|
||
|
|
"num_tokens": 1335401.0,
|
||
|
|
"step": 595
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.8328300476074215,
|
||
|
|
"epoch": 0.6489994591671173,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.0002995,
|
||
|
|
"loss": 6.4707,
|
||
|
|
"mean_token_accuracy": 0.127620816975832,
|
||
|
|
"num_tokens": 1348265.0,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.7745624542236325,
|
||
|
|
"epoch": 0.65440778799351,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.000302,
|
||
|
|
"loss": 6.4306,
|
||
|
|
"mean_token_accuracy": 0.13703566640615464,
|
||
|
|
"num_tokens": 1359078.0,
|
||
|
|
"step": 605
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.712151575088501,
|
||
|
|
"epoch": 0.6598161168199026,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.0003045,
|
||
|
|
"loss": 6.3318,
|
||
|
|
"mean_token_accuracy": 0.13108599931001663,
|
||
|
|
"num_tokens": 1369499.0,
|
||
|
|
"step": 610
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.8127624034881595,
|
||
|
|
"epoch": 0.6652244456462953,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.000307,
|
||
|
|
"loss": 6.4323,
|
||
|
|
"mean_token_accuracy": 0.1286877028644085,
|
||
|
|
"num_tokens": 1380329.0,
|
||
|
|
"step": 615
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.824408864974975,
|
||
|
|
"epoch": 0.670632774472688,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.0003095,
|
||
|
|
"loss": 6.5116,
|
||
|
|
"mean_token_accuracy": 0.12648341283202172,
|
||
|
|
"num_tokens": 1393047.0,
|
||
|
|
"step": 620
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.872823572158813,
|
||
|
|
"epoch": 0.6760411032990806,
|
||
|
|
"grad_norm": 0.9140625,
|
||
|
|
"learning_rate": 0.000312,
|
||
|
|
"loss": 6.5067,
|
||
|
|
"mean_token_accuracy": 0.12922688201069832,
|
||
|
|
"num_tokens": 1405600.0,
|
||
|
|
"step": 625
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.811132955551147,
|
||
|
|
"epoch": 0.6814494321254733,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.0003145,
|
||
|
|
"loss": 6.4417,
|
||
|
|
"mean_token_accuracy": 0.13407272696495057,
|
||
|
|
"num_tokens": 1416670.0,
|
||
|
|
"step": 630
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.825539922714233,
|
||
|
|
"epoch": 0.6868577609518659,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.000317,
|
||
|
|
"loss": 6.4599,
|
||
|
|
"mean_token_accuracy": 0.12846649512648584,
|
||
|
|
"num_tokens": 1428929.0,
|
||
|
|
"step": 635
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.827125072479248,
|
||
|
|
"epoch": 0.6922660897782585,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.0003195,
|
||
|
|
"loss": 6.3887,
|
||
|
|
"mean_token_accuracy": 0.13251324519515037,
|
||
|
|
"num_tokens": 1439298.0,
|
||
|
|
"step": 640
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.817316055297852,
|
||
|
|
"epoch": 0.6976744186046512,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.000322,
|
||
|
|
"loss": 6.4844,
|
||
|
|
"mean_token_accuracy": 0.13544763550162314,
|
||
|
|
"num_tokens": 1449853.0,
|
||
|
|
"step": 645
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.891280555725098,
|
||
|
|
"epoch": 0.7030827474310438,
|
||
|
|
"grad_norm": 1.171875,
|
||
|
|
"learning_rate": 0.00032450000000000003,
|
||
|
|
"loss": 6.5515,
|
||
|
|
"mean_token_accuracy": 0.12688857838511466,
|
||
|
|
"num_tokens": 1460816.0,
|
||
|
|
"step": 650
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.740885925292969,
|
||
|
|
"epoch": 0.7084910762574365,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.00032700000000000003,
|
||
|
|
"loss": 6.4014,
|
||
|
|
"mean_token_accuracy": 0.13262718617916108,
|
||
|
|
"num_tokens": 1473003.0,
|
||
|
|
"step": 655
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.843165111541748,
|
||
|
|
"epoch": 0.7138994050838291,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.00032950000000000004,
|
||
|
|
"loss": 6.5206,
|
||
|
|
"mean_token_accuracy": 0.1229576200246811,
|
||
|
|
"num_tokens": 1484745.0,
|
||
|
|
"step": 660
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.823391914367676,
|
||
|
|
"epoch": 0.7193077339102217,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.00033200000000000005,
|
||
|
|
"loss": 6.4011,
|
||
|
|
"mean_token_accuracy": 0.13305559009313583,
|
||
|
|
"num_tokens": 1495803.0,
|
||
|
|
"step": 665
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.803869295120239,
|
||
|
|
"epoch": 0.7247160627366144,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.00033450000000000005,
|
||
|
|
"loss": 6.4092,
|
||
|
|
"mean_token_accuracy": 0.141731408983469,
|
||
|
|
"num_tokens": 1507504.0,
|
||
|
|
"step": 670
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.852431774139404,
|
||
|
|
"epoch": 0.730124391563007,
|
||
|
|
"grad_norm": 0.98046875,
|
||
|
|
"learning_rate": 0.000337,
|
||
|
|
"loss": 6.5033,
|
||
|
|
"mean_token_accuracy": 0.12678006961941718,
|
||
|
|
"num_tokens": 1518765.0,
|
||
|
|
"step": 675
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.650970602035523,
|
||
|
|
"epoch": 0.7355327203893997,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.0003395,
|
||
|
|
"loss": 6.3839,
|
||
|
|
"mean_token_accuracy": 0.1324382781982422,
|
||
|
|
"num_tokens": 1529481.0,
|
||
|
|
"step": 680
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.810516691207885,
|
||
|
|
"epoch": 0.7409410492157923,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.000342,
|
||
|
|
"loss": 6.3986,
|
||
|
|
"mean_token_accuracy": 0.13725290223956108,
|
||
|
|
"num_tokens": 1539612.0,
|
||
|
|
"step": 685
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.801828384399414,
|
||
|
|
"epoch": 0.7463493780421849,
|
||
|
|
"grad_norm": 0.984375,
|
||
|
|
"learning_rate": 0.00034449999999999997,
|
||
|
|
"loss": 6.558,
|
||
|
|
"mean_token_accuracy": 0.12963267266750336,
|
||
|
|
"num_tokens": 1551370.0,
|
||
|
|
"step": 690
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.8387431621551515,
|
||
|
|
"epoch": 0.7517577068685776,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.000347,
|
||
|
|
"loss": 6.44,
|
||
|
|
"mean_token_accuracy": 0.1378513492643833,
|
||
|
|
"num_tokens": 1561662.0,
|
||
|
|
"step": 695
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.7489661693573,
|
||
|
|
"epoch": 0.7571660356949702,
|
||
|
|
"grad_norm": 0.91796875,
|
||
|
|
"learning_rate": 0.0003495,
|
||
|
|
"loss": 6.4232,
|
||
|
|
"mean_token_accuracy": 0.13230135440826415,
|
||
|
|
"num_tokens": 1573479.0,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.756485414505005,
|
||
|
|
"epoch": 0.7625743645213628,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 0.000352,
|
||
|
|
"loss": 6.366,
|
||
|
|
"mean_token_accuracy": 0.1389087036252022,
|
||
|
|
"num_tokens": 1583684.0,
|
||
|
|
"step": 705
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.7295139789581295,
|
||
|
|
"epoch": 0.7679826933477556,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0003545,
|
||
|
|
"loss": 6.4101,
|
||
|
|
"mean_token_accuracy": 0.13583654090762137,
|
||
|
|
"num_tokens": 1595694.0,
|
||
|
|
"step": 710
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.696416282653809,
|
||
|
|
"epoch": 0.7733910221741482,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.000357,
|
||
|
|
"loss": 6.4184,
|
||
|
|
"mean_token_accuracy": 0.13240900188684462,
|
||
|
|
"num_tokens": 1606104.0,
|
||
|
|
"step": 715
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.7834718227386475,
|
||
|
|
"epoch": 0.7787993510005409,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.0003595,
|
||
|
|
"loss": 6.4803,
|
||
|
|
"mean_token_accuracy": 0.13445528894662856,
|
||
|
|
"num_tokens": 1617368.0,
|
||
|
|
"step": 720
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.790578842163086,
|
||
|
|
"epoch": 0.7842076798269335,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.000362,
|
||
|
|
"loss": 6.4893,
|
||
|
|
"mean_token_accuracy": 0.1305526874959469,
|
||
|
|
"num_tokens": 1630485.0,
|
||
|
|
"step": 725
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.788717746734619,
|
||
|
|
"epoch": 0.7896160086533262,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.0003645,
|
||
|
|
"loss": 6.3357,
|
||
|
|
"mean_token_accuracy": 0.1369060456752777,
|
||
|
|
"num_tokens": 1641411.0,
|
||
|
|
"step": 730
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.644323205947876,
|
||
|
|
"epoch": 0.7950243374797188,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.000367,
|
||
|
|
"loss": 6.3465,
|
||
|
|
"mean_token_accuracy": 0.13824027627706528,
|
||
|
|
"num_tokens": 1653737.0,
|
||
|
|
"step": 735
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.823546600341797,
|
||
|
|
"epoch": 0.8004326663061114,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0003695,
|
||
|
|
"loss": 6.5281,
|
||
|
|
"mean_token_accuracy": 0.12467064261436463,
|
||
|
|
"num_tokens": 1665910.0,
|
||
|
|
"step": 740
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.726630735397339,
|
||
|
|
"epoch": 0.8058409951325041,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.000372,
|
||
|
|
"loss": 6.3094,
|
||
|
|
"mean_token_accuracy": 0.134908963739872,
|
||
|
|
"num_tokens": 1676927.0,
|
||
|
|
"step": 745
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.629592800140381,
|
||
|
|
"epoch": 0.8112493239588967,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.0003745,
|
||
|
|
"loss": 6.357,
|
||
|
|
"mean_token_accuracy": 0.13599003627896308,
|
||
|
|
"num_tokens": 1688167.0,
|
||
|
|
"step": 750
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.650397443771363,
|
||
|
|
"epoch": 0.8166576527852893,
|
||
|
|
"grad_norm": 1.0,
|
||
|
|
"learning_rate": 0.000377,
|
||
|
|
"loss": 6.4308,
|
||
|
|
"mean_token_accuracy": 0.13610320165753365,
|
||
|
|
"num_tokens": 1699899.0,
|
||
|
|
"step": 755
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.7162903308868405,
|
||
|
|
"epoch": 0.822065981611682,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.0003795,
|
||
|
|
"loss": 6.3533,
|
||
|
|
"mean_token_accuracy": 0.1412688173353672,
|
||
|
|
"num_tokens": 1709862.0,
|
||
|
|
"step": 760
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.640978765487671,
|
||
|
|
"epoch": 0.8274743104380746,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.000382,
|
||
|
|
"loss": 6.3481,
|
||
|
|
"mean_token_accuracy": 0.1376900367438793,
|
||
|
|
"num_tokens": 1719902.0,
|
||
|
|
"step": 765
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.691372346878052,
|
||
|
|
"epoch": 0.8328826392644673,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.0003845,
|
||
|
|
"loss": 6.3541,
|
||
|
|
"mean_token_accuracy": 0.13510108962655068,
|
||
|
|
"num_tokens": 1730351.0,
|
||
|
|
"step": 770
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.589955043792725,
|
||
|
|
"epoch": 0.8382909680908599,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.00038700000000000003,
|
||
|
|
"loss": 6.3233,
|
||
|
|
"mean_token_accuracy": 0.1341069035232067,
|
||
|
|
"num_tokens": 1741783.0,
|
||
|
|
"step": 775
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.609476327896118,
|
||
|
|
"epoch": 0.8436992969172525,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.00038950000000000003,
|
||
|
|
"loss": 6.3639,
|
||
|
|
"mean_token_accuracy": 0.13918454721570014,
|
||
|
|
"num_tokens": 1752952.0,
|
||
|
|
"step": 780
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.713692474365234,
|
||
|
|
"epoch": 0.8491076257436452,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.00039200000000000004,
|
||
|
|
"loss": 6.3796,
|
||
|
|
"mean_token_accuracy": 0.1377579912543297,
|
||
|
|
"num_tokens": 1764014.0,
|
||
|
|
"step": 785
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.692055416107178,
|
||
|
|
"epoch": 0.8545159545700378,
|
||
|
|
"grad_norm": 0.9921875,
|
||
|
|
"learning_rate": 0.00039450000000000005,
|
||
|
|
"loss": 6.3501,
|
||
|
|
"mean_token_accuracy": 0.13647191673517228,
|
||
|
|
"num_tokens": 1776154.0,
|
||
|
|
"step": 790
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.651118326187134,
|
||
|
|
"epoch": 0.8599242833964305,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.00039700000000000005,
|
||
|
|
"loss": 6.405,
|
||
|
|
"mean_token_accuracy": 0.1311624176800251,
|
||
|
|
"num_tokens": 1787292.0,
|
||
|
|
"step": 795
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.63629379272461,
|
||
|
|
"epoch": 0.8653326122228231,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.0003995,
|
||
|
|
"loss": 6.3667,
|
||
|
|
"mean_token_accuracy": 0.13610742837190629,
|
||
|
|
"num_tokens": 1799469.0,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.567807674407959,
|
||
|
|
"epoch": 0.8707409410492158,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.000402,
|
||
|
|
"loss": 6.2441,
|
||
|
|
"mean_token_accuracy": 0.14459661692380904,
|
||
|
|
"num_tokens": 1810091.0,
|
||
|
|
"step": 805
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.723603248596191,
|
||
|
|
"epoch": 0.8761492698756085,
|
||
|
|
"grad_norm": 0.9921875,
|
||
|
|
"learning_rate": 0.0004045,
|
||
|
|
"loss": 6.3925,
|
||
|
|
"mean_token_accuracy": 0.14025203064084052,
|
||
|
|
"num_tokens": 1820835.0,
|
||
|
|
"step": 810
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.633598756790161,
|
||
|
|
"epoch": 0.8815575987020011,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.00040699999999999997,
|
||
|
|
"loss": 6.3549,
|
||
|
|
"mean_token_accuracy": 0.13814282566308975,
|
||
|
|
"num_tokens": 1831868.0,
|
||
|
|
"step": 815
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.587992763519287,
|
||
|
|
"epoch": 0.8869659275283938,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.0004095,
|
||
|
|
"loss": 6.3383,
|
||
|
|
"mean_token_accuracy": 0.13848361223936081,
|
||
|
|
"num_tokens": 1842040.0,
|
||
|
|
"step": 820
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.723627090454102,
|
||
|
|
"epoch": 0.8923742563547864,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.000412,
|
||
|
|
"loss": 6.4083,
|
||
|
|
"mean_token_accuracy": 0.13382988572120666,
|
||
|
|
"num_tokens": 1853765.0,
|
||
|
|
"step": 825
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.673075199127197,
|
||
|
|
"epoch": 0.897782585181179,
|
||
|
|
"grad_norm": 1.0,
|
||
|
|
"learning_rate": 0.0004145,
|
||
|
|
"loss": 6.4141,
|
||
|
|
"mean_token_accuracy": 0.13732750713825226,
|
||
|
|
"num_tokens": 1866081.0,
|
||
|
|
"step": 830
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.593461990356445,
|
||
|
|
"epoch": 0.9031909140075717,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000417,
|
||
|
|
"loss": 6.3155,
|
||
|
|
"mean_token_accuracy": 0.13803164809942245,
|
||
|
|
"num_tokens": 1876754.0,
|
||
|
|
"step": 835
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.714284658432007,
|
||
|
|
"epoch": 0.9085992428339643,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.0004195,
|
||
|
|
"loss": 6.4263,
|
||
|
|
"mean_token_accuracy": 0.1334732621908188,
|
||
|
|
"num_tokens": 1888299.0,
|
||
|
|
"step": 840
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.475223159790039,
|
||
|
|
"epoch": 0.914007571660357,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.000422,
|
||
|
|
"loss": 6.2122,
|
||
|
|
"mean_token_accuracy": 0.1465678930282593,
|
||
|
|
"num_tokens": 1899027.0,
|
||
|
|
"step": 845
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.625812816619873,
|
||
|
|
"epoch": 0.9194159004867496,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.0004245,
|
||
|
|
"loss": 6.3374,
|
||
|
|
"mean_token_accuracy": 0.14080506712198257,
|
||
|
|
"num_tokens": 1911043.0,
|
||
|
|
"step": 850
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.590130615234375,
|
||
|
|
"epoch": 0.9248242293131422,
|
||
|
|
"grad_norm": 1.0390625,
|
||
|
|
"learning_rate": 0.000427,
|
||
|
|
"loss": 6.2567,
|
||
|
|
"mean_token_accuracy": 0.13707533478736877,
|
||
|
|
"num_tokens": 1921295.0,
|
||
|
|
"step": 855
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.61051025390625,
|
||
|
|
"epoch": 0.9302325581395349,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.0004295,
|
||
|
|
"loss": 6.3569,
|
||
|
|
"mean_token_accuracy": 0.13993956744670868,
|
||
|
|
"num_tokens": 1932944.0,
|
||
|
|
"step": 860
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.553638172149658,
|
||
|
|
"epoch": 0.9356408869659275,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.000432,
|
||
|
|
"loss": 6.3027,
|
||
|
|
"mean_token_accuracy": 0.13260096460580825,
|
||
|
|
"num_tokens": 1944072.0,
|
||
|
|
"step": 865
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.569744491577149,
|
||
|
|
"epoch": 0.9410492157923201,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.0004345,
|
||
|
|
"loss": 6.2956,
|
||
|
|
"mean_token_accuracy": 0.13898471891880035,
|
||
|
|
"num_tokens": 1954798.0,
|
||
|
|
"step": 870
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.513964128494263,
|
||
|
|
"epoch": 0.9464575446187128,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.000437,
|
||
|
|
"loss": 6.2698,
|
||
|
|
"mean_token_accuracy": 0.1473902389407158,
|
||
|
|
"num_tokens": 1965304.0,
|
||
|
|
"step": 875
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.6614869117736815,
|
||
|
|
"epoch": 0.9518658734451054,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.0004395,
|
||
|
|
"loss": 6.3094,
|
||
|
|
"mean_token_accuracy": 0.13569752722978592,
|
||
|
|
"num_tokens": 1976797.0,
|
||
|
|
"step": 880
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.550220966339111,
|
||
|
|
"epoch": 0.9572742022714981,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.000442,
|
||
|
|
"loss": 6.3248,
|
||
|
|
"mean_token_accuracy": 0.13910794109106064,
|
||
|
|
"num_tokens": 1987448.0,
|
||
|
|
"step": 885
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.369886207580566,
|
||
|
|
"epoch": 0.9626825310978907,
|
||
|
|
"grad_norm": 0.9453125,
|
||
|
|
"learning_rate": 0.0004445,
|
||
|
|
"loss": 6.1508,
|
||
|
|
"mean_token_accuracy": 0.1404670625925064,
|
||
|
|
"num_tokens": 2000948.0,
|
||
|
|
"step": 890
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.55863881111145,
|
||
|
|
"epoch": 0.9680908599242833,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.000447,
|
||
|
|
"loss": 6.2643,
|
||
|
|
"mean_token_accuracy": 0.14369212165474893,
|
||
|
|
"num_tokens": 2011807.0,
|
||
|
|
"step": 895
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.6163318157196045,
|
||
|
|
"epoch": 0.9734991887506761,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.00044950000000000003,
|
||
|
|
"loss": 6.3207,
|
||
|
|
"mean_token_accuracy": 0.13593853637576103,
|
||
|
|
"num_tokens": 2022835.0,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.3906701564788815,
|
||
|
|
"epoch": 0.9789075175770687,
|
||
|
|
"grad_norm": 1.0078125,
|
||
|
|
"learning_rate": 0.00045200000000000004,
|
||
|
|
"loss": 6.2339,
|
||
|
|
"mean_token_accuracy": 0.14182498455047607,
|
||
|
|
"num_tokens": 2034331.0,
|
||
|
|
"step": 905
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.662575387954712,
|
||
|
|
"epoch": 0.9843158464034614,
|
||
|
|
"grad_norm": 1.0,
|
||
|
|
"learning_rate": 0.00045450000000000004,
|
||
|
|
"loss": 6.2328,
|
||
|
|
"mean_token_accuracy": 0.13782825246453284,
|
||
|
|
"num_tokens": 2047002.0,
|
||
|
|
"step": 910
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.408291864395141,
|
||
|
|
"epoch": 0.989724175229854,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.00045700000000000005,
|
||
|
|
"loss": 6.1835,
|
||
|
|
"mean_token_accuracy": 0.14051275625824927,
|
||
|
|
"num_tokens": 2056299.0,
|
||
|
|
"step": 915
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.519381284713745,
|
||
|
|
"epoch": 0.9951325040562466,
|
||
|
|
"grad_norm": 1.0,
|
||
|
|
"learning_rate": 0.00045950000000000006,
|
||
|
|
"loss": 6.3108,
|
||
|
|
"mean_token_accuracy": 0.13649527579545975,
|
||
|
|
"num_tokens": 2068031.0,
|
||
|
|
"step": 920
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.533435503641765,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"grad_norm": 1.8125,
|
||
|
|
"learning_rate": 0.000462,
|
||
|
|
"loss": 6.2559,
|
||
|
|
"mean_token_accuracy": 0.1411174734433492,
|
||
|
|
"num_tokens": 2076889.0,
|
||
|
|
"step": 925
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.518666791915893,
|
||
|
|
"epoch": 1.0054083288263926,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.0004645,
|
||
|
|
"loss": 6.1187,
|
||
|
|
"mean_token_accuracy": 0.14975121915340422,
|
||
|
|
"num_tokens": 2087316.0,
|
||
|
|
"step": 930
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.499506664276123,
|
||
|
|
"epoch": 1.0108166576527853,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.000467,
|
||
|
|
"loss": 6.1223,
|
||
|
|
"mean_token_accuracy": 0.14297691583633423,
|
||
|
|
"num_tokens": 2098522.0,
|
||
|
|
"step": 935
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.383229446411133,
|
||
|
|
"epoch": 1.016224986479178,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.0004695,
|
||
|
|
"loss": 6.0878,
|
||
|
|
"mean_token_accuracy": 0.15435394942760466,
|
||
|
|
"num_tokens": 2110015.0,
|
||
|
|
"step": 940
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.525584554672241,
|
||
|
|
"epoch": 1.0216333153055706,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.000472,
|
||
|
|
"loss": 6.1924,
|
||
|
|
"mean_token_accuracy": 0.1424684338271618,
|
||
|
|
"num_tokens": 2121333.0,
|
||
|
|
"step": 945
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.4724767208099365,
|
||
|
|
"epoch": 1.0270416441319632,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.0004745,
|
||
|
|
"loss": 6.0838,
|
||
|
|
"mean_token_accuracy": 0.14994525909423828,
|
||
|
|
"num_tokens": 2131385.0,
|
||
|
|
"step": 950
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.488972854614258,
|
||
|
|
"epoch": 1.0324499729583558,
|
||
|
|
"grad_norm": 1.015625,
|
||
|
|
"learning_rate": 0.000477,
|
||
|
|
"loss": 6.1275,
|
||
|
|
"mean_token_accuracy": 0.1522180952131748,
|
||
|
|
"num_tokens": 2143274.0,
|
||
|
|
"step": 955
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.368765354156494,
|
||
|
|
"epoch": 1.0378583017847485,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.0004795,
|
||
|
|
"loss": 6.0124,
|
||
|
|
"mean_token_accuracy": 0.15076057612895966,
|
||
|
|
"num_tokens": 2154529.0,
|
||
|
|
"step": 960
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.4038872718811035,
|
||
|
|
"epoch": 1.043266630611141,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.000482,
|
||
|
|
"loss": 6.1603,
|
||
|
|
"mean_token_accuracy": 0.1439424842596054,
|
||
|
|
"num_tokens": 2165949.0,
|
||
|
|
"step": 965
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.407582426071167,
|
||
|
|
"epoch": 1.0486749594375337,
|
||
|
|
"grad_norm": 1.0546875,
|
||
|
|
"learning_rate": 0.0004845,
|
||
|
|
"loss": 6.1335,
|
||
|
|
"mean_token_accuracy": 0.1485764428973198,
|
||
|
|
"num_tokens": 2176710.0,
|
||
|
|
"step": 970
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.463790941238403,
|
||
|
|
"epoch": 1.0540832882639264,
|
||
|
|
"grad_norm": 0.99609375,
|
||
|
|
"learning_rate": 0.000487,
|
||
|
|
"loss": 6.1074,
|
||
|
|
"mean_token_accuracy": 0.15412099212408065,
|
||
|
|
"num_tokens": 2189470.0,
|
||
|
|
"step": 975
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.383792352676392,
|
||
|
|
"epoch": 1.059491617090319,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.0004895,
|
||
|
|
"loss": 6.0844,
|
||
|
|
"mean_token_accuracy": 0.1579994946718216,
|
||
|
|
"num_tokens": 2200266.0,
|
||
|
|
"step": 980
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.395930194854737,
|
||
|
|
"epoch": 1.0648999459167117,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.000492,
|
||
|
|
"loss": 6.0915,
|
||
|
|
"mean_token_accuracy": 0.14566799476742745,
|
||
|
|
"num_tokens": 2210736.0,
|
||
|
|
"step": 985
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.478207111358643,
|
||
|
|
"epoch": 1.0703082747431043,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0004945,
|
||
|
|
"loss": 6.1985,
|
||
|
|
"mean_token_accuracy": 0.14791767448186874,
|
||
|
|
"num_tokens": 2221057.0,
|
||
|
|
"step": 990
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.455506134033203,
|
||
|
|
"epoch": 1.075716603569497,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.000497,
|
||
|
|
"loss": 6.1738,
|
||
|
|
"mean_token_accuracy": 0.14685477688908577,
|
||
|
|
"num_tokens": 2232352.0,
|
||
|
|
"step": 995
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.510460042953492,
|
||
|
|
"epoch": 1.0811249323958896,
|
||
|
|
"grad_norm": 0.89453125,
|
||
|
|
"learning_rate": 0.0004995,
|
||
|
|
"loss": 6.0805,
|
||
|
|
"mean_token_accuracy": 0.15254651606082917,
|
||
|
|
"num_tokens": 2244655.0,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0811249323958896,
|
||
|
|
"eval_entropy": 6.124205052852631,
|
||
|
|
"eval_loss": 6.387869834899902,
|
||
|
|
"eval_mean_token_accuracy": 0.1444568099030717,
|
||
|
|
"eval_num_tokens": 2244655.0,
|
||
|
|
"eval_runtime": 1.9538,
|
||
|
|
"eval_samples_per_second": 1799.022,
|
||
|
|
"eval_steps_per_second": 225.198,
|
||
|
|
"step": 1000
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 5,
|
||
|
|
"max_steps": 9240,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 10,
|
||
|
|
"save_steps": 500,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": false
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 517821876633600.0,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|