8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.134796238244514,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.74259386062622,
|
|
"epoch": 0.003918495297805642,
|
|
"grad_norm": 5.34375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7559,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9174.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742582988739013,
|
|
"epoch": 0.007836990595611285,
|
|
"grad_norm": 5.09375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7343,
|
|
"mean_token_accuracy": 0.00011074197245761753,
|
|
"num_tokens": 17790.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742611026763916,
|
|
"epoch": 0.011755485893416929,
|
|
"grad_norm": 5.03125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.6946,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 25850.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742605590820313,
|
|
"epoch": 0.01567398119122257,
|
|
"grad_norm": 5.1875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6748,
|
|
"mean_token_accuracy": 0.001098086079582572,
|
|
"num_tokens": 35194.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742543792724609,
|
|
"epoch": 0.019592476489028215,
|
|
"grad_norm": 4.53125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5631,
|
|
"mean_token_accuracy": 0.008560778619721533,
|
|
"num_tokens": 44218.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.74220838546753,
|
|
"epoch": 0.023510971786833857,
|
|
"grad_norm": 4.09375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4525,
|
|
"mean_token_accuracy": 0.03252750001847744,
|
|
"num_tokens": 53397.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740808391571045,
|
|
"epoch": 0.0274294670846395,
|
|
"grad_norm": 3.421875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3079,
|
|
"mean_token_accuracy": 0.0437689334154129,
|
|
"num_tokens": 62749.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.736591911315918,
|
|
"epoch": 0.03134796238244514,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1743,
|
|
"mean_token_accuracy": 0.043372325226664545,
|
|
"num_tokens": 71721.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.729655265808105,
|
|
"epoch": 0.03526645768025078,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0405,
|
|
"mean_token_accuracy": 0.046825164929032326,
|
|
"num_tokens": 79844.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.72356595993042,
|
|
"epoch": 0.03918495297805643,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9917,
|
|
"mean_token_accuracy": 0.04553030654788017,
|
|
"num_tokens": 88866.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.719326305389405,
|
|
"epoch": 0.04310344827586207,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9336,
|
|
"mean_token_accuracy": 0.044035762920975684,
|
|
"num_tokens": 97918.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.715408420562744,
|
|
"epoch": 0.047021943573667714,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8702,
|
|
"mean_token_accuracy": 0.04544390179216862,
|
|
"num_tokens": 107043.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.710628414154053,
|
|
"epoch": 0.050940438871473356,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8037,
|
|
"mean_token_accuracy": 0.04611521698534489,
|
|
"num_tokens": 116000.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.704075908660888,
|
|
"epoch": 0.054858934169279,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7686,
|
|
"mean_token_accuracy": 0.04174055978655815,
|
|
"num_tokens": 124559.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.700677013397216,
|
|
"epoch": 0.05877742946708464,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.6568,
|
|
"mean_token_accuracy": 0.04639651067554951,
|
|
"num_tokens": 132868.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.695802783966064,
|
|
"epoch": 0.06269592476489028,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6523,
|
|
"mean_token_accuracy": 0.04210274666547775,
|
|
"num_tokens": 141286.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.688793563842774,
|
|
"epoch": 0.06661442006269593,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5499,
|
|
"mean_token_accuracy": 0.05031363070011139,
|
|
"num_tokens": 150406.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.679227447509765,
|
|
"epoch": 0.07053291536050156,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4859,
|
|
"mean_token_accuracy": 0.056176548823714256,
|
|
"num_tokens": 158770.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.663717269897461,
|
|
"epoch": 0.07445141065830721,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4023,
|
|
"mean_token_accuracy": 0.06409536376595497,
|
|
"num_tokens": 167763.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.639647197723388,
|
|
"epoch": 0.07836990595611286,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.286,
|
|
"mean_token_accuracy": 0.06052683852612972,
|
|
"num_tokens": 177306.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.605971527099609,
|
|
"epoch": 0.0822884012539185,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1915,
|
|
"mean_token_accuracy": 0.06200857125222683,
|
|
"num_tokens": 186014.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.545482730865478,
|
|
"epoch": 0.08620689655172414,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0547,
|
|
"mean_token_accuracy": 0.06552336104214192,
|
|
"num_tokens": 194122.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.471019172668457,
|
|
"epoch": 0.09012539184952978,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9639,
|
|
"mean_token_accuracy": 0.061758537963032725,
|
|
"num_tokens": 203097.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.420811939239503,
|
|
"epoch": 0.09404388714733543,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8529,
|
|
"mean_token_accuracy": 0.06423232220113277,
|
|
"num_tokens": 211413.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.383054542541505,
|
|
"epoch": 0.09796238244514106,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7779,
|
|
"mean_token_accuracy": 0.06270350776612758,
|
|
"num_tokens": 220550.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.31447114944458,
|
|
"epoch": 0.10188087774294671,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6079,
|
|
"mean_token_accuracy": 0.06446850374341011,
|
|
"num_tokens": 229197.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.19693603515625,
|
|
"epoch": 0.10579937304075235,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.5715,
|
|
"mean_token_accuracy": 0.05803979113698006,
|
|
"num_tokens": 238479.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.129131984710693,
|
|
"epoch": 0.109717868338558,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4073,
|
|
"mean_token_accuracy": 0.06696602776646614,
|
|
"num_tokens": 246318.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.014921092987061,
|
|
"epoch": 0.11363636363636363,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.3497,
|
|
"mean_token_accuracy": 0.06019376777112484,
|
|
"num_tokens": 254783.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.837921333312988,
|
|
"epoch": 0.11755485893416928,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.1727,
|
|
"mean_token_accuracy": 0.06220055930316448,
|
|
"num_tokens": 263416.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.704282855987548,
|
|
"epoch": 0.12147335423197492,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1265,
|
|
"mean_token_accuracy": 0.06537272743880748,
|
|
"num_tokens": 271930.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.534114360809326,
|
|
"epoch": 0.12539184952978055,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9819,
|
|
"mean_token_accuracy": 0.06310658231377601,
|
|
"num_tokens": 280546.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.289366626739502,
|
|
"epoch": 0.12931034482758622,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.9162,
|
|
"mean_token_accuracy": 0.0649850957095623,
|
|
"num_tokens": 288813.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.108893775939942,
|
|
"epoch": 0.13322884012539185,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8379,
|
|
"mean_token_accuracy": 0.06351686343550682,
|
|
"num_tokens": 297966.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.892046165466308,
|
|
"epoch": 0.1371473354231975,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8598,
|
|
"mean_token_accuracy": 0.06292634233832359,
|
|
"num_tokens": 307135.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.728424072265625,
|
|
"epoch": 0.14106583072100312,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7429,
|
|
"mean_token_accuracy": 0.06494242362678052,
|
|
"num_tokens": 315546.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.619221019744874,
|
|
"epoch": 0.14498432601880878,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7168,
|
|
"mean_token_accuracy": 0.06593996658921242,
|
|
"num_tokens": 323930.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.409439945220948,
|
|
"epoch": 0.14890282131661442,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.5689,
|
|
"mean_token_accuracy": 0.0733262188732624,
|
|
"num_tokens": 332419.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.326526832580566,
|
|
"epoch": 0.15282131661442006,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.654,
|
|
"mean_token_accuracy": 0.0696730900555849,
|
|
"num_tokens": 341362.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.3056715965271,
|
|
"epoch": 0.15673981191222572,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.5645,
|
|
"mean_token_accuracy": 0.07249549068510533,
|
|
"num_tokens": 349395.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.24736156463623,
|
|
"epoch": 0.16065830721003135,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5705,
|
|
"mean_token_accuracy": 0.06626113764941692,
|
|
"num_tokens": 358413.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.157195377349854,
|
|
"epoch": 0.164576802507837,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.6532,
|
|
"mean_token_accuracy": 0.07263443917036057,
|
|
"num_tokens": 366489.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.1927809715271,
|
|
"epoch": 0.16849529780564262,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.5231,
|
|
"mean_token_accuracy": 0.06818853206932544,
|
|
"num_tokens": 375335.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.176179027557373,
|
|
"epoch": 0.1724137931034483,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.5638,
|
|
"mean_token_accuracy": 0.07268193662166596,
|
|
"num_tokens": 384276.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.140285873413086,
|
|
"epoch": 0.17633228840125392,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.5991,
|
|
"mean_token_accuracy": 0.06890415549278259,
|
|
"num_tokens": 393571.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.1433687210083,
|
|
"epoch": 0.18025078369905956,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.5228,
|
|
"mean_token_accuracy": 0.07579943463206291,
|
|
"num_tokens": 401865.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.072039985656739,
|
|
"epoch": 0.1841692789968652,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.4982,
|
|
"mean_token_accuracy": 0.07117367573082448,
|
|
"num_tokens": 410518.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.076621770858765,
|
|
"epoch": 0.18808777429467086,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.5051,
|
|
"mean_token_accuracy": 0.06969159096479416,
|
|
"num_tokens": 419769.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.030782747268677,
|
|
"epoch": 0.1920062695924765,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.4391,
|
|
"mean_token_accuracy": 0.07655345052480697,
|
|
"num_tokens": 428204.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.057575464248657,
|
|
"epoch": 0.19592476489028213,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.5263,
|
|
"mean_token_accuracy": 0.07384133897721767,
|
|
"num_tokens": 436594.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.054854202270509,
|
|
"epoch": 0.19984326018808776,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.3139,
|
|
"mean_token_accuracy": 0.07670355625450612,
|
|
"num_tokens": 444645.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.953717613220215,
|
|
"epoch": 0.20376175548589343,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3561,
|
|
"mean_token_accuracy": 0.07812742255628109,
|
|
"num_tokens": 453016.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.9555809020996096,
|
|
"epoch": 0.20768025078369906,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.4192,
|
|
"mean_token_accuracy": 0.07524299174547196,
|
|
"num_tokens": 462116.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.957012748718261,
|
|
"epoch": 0.2115987460815047,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3694,
|
|
"mean_token_accuracy": 0.07774586156010628,
|
|
"num_tokens": 470807.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.9492302417755125,
|
|
"epoch": 0.21551724137931033,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.4561,
|
|
"mean_token_accuracy": 0.07432891242206097,
|
|
"num_tokens": 479592.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.016344738006591,
|
|
"epoch": 0.219435736677116,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.3803,
|
|
"mean_token_accuracy": 0.07607875801622868,
|
|
"num_tokens": 488107.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.870184087753296,
|
|
"epoch": 0.22335423197492163,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.3792,
|
|
"mean_token_accuracy": 0.07687489986419678,
|
|
"num_tokens": 496775.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.973387956619263,
|
|
"epoch": 0.22727272727272727,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.4096,
|
|
"mean_token_accuracy": 0.07644881159067154,
|
|
"num_tokens": 505415.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.946991682052612,
|
|
"epoch": 0.23119122257053293,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.3915,
|
|
"mean_token_accuracy": 0.07934824042022229,
|
|
"num_tokens": 514447.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.891976928710937,
|
|
"epoch": 0.23510971786833856,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.381,
|
|
"mean_token_accuracy": 0.07889064475893974,
|
|
"num_tokens": 522998.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.92128734588623,
|
|
"epoch": 0.2390282131661442,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.3459,
|
|
"mean_token_accuracy": 0.08099666759371757,
|
|
"num_tokens": 531542.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.863611841201783,
|
|
"epoch": 0.24294670846394983,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2927,
|
|
"mean_token_accuracy": 0.07772218361496926,
|
|
"num_tokens": 540143.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.924414491653442,
|
|
"epoch": 0.2468652037617555,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.3679,
|
|
"mean_token_accuracy": 0.08110597431659698,
|
|
"num_tokens": 549156.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.803961706161499,
|
|
"epoch": 0.2507836990595611,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2252,
|
|
"mean_token_accuracy": 0.08031321838498115,
|
|
"num_tokens": 557522.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.805857229232788,
|
|
"epoch": 0.2547021943573668,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2879,
|
|
"mean_token_accuracy": 0.08254150673747063,
|
|
"num_tokens": 566650.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.8912803649902346,
|
|
"epoch": 0.25862068965517243,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2408,
|
|
"mean_token_accuracy": 0.08324127942323685,
|
|
"num_tokens": 576219.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.849798917770386,
|
|
"epoch": 0.26253918495297807,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.3416,
|
|
"mean_token_accuracy": 0.08195492923259735,
|
|
"num_tokens": 584304.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.75479941368103,
|
|
"epoch": 0.2664576802507837,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2574,
|
|
"mean_token_accuracy": 0.08678329288959503,
|
|
"num_tokens": 593163.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.7592004299163815,
|
|
"epoch": 0.27037617554858934,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1773,
|
|
"mean_token_accuracy": 0.08335737958550453,
|
|
"num_tokens": 602077.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.792924070358277,
|
|
"epoch": 0.274294670846395,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.2812,
|
|
"mean_token_accuracy": 0.08297529369592667,
|
|
"num_tokens": 610009.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.816137742996216,
|
|
"epoch": 0.2782131661442006,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.3247,
|
|
"mean_token_accuracy": 0.07837995067238808,
|
|
"num_tokens": 619282.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.692404460906983,
|
|
"epoch": 0.28213166144200624,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.2457,
|
|
"mean_token_accuracy": 0.08888485580682755,
|
|
"num_tokens": 628138.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.819083976745605,
|
|
"epoch": 0.28605015673981193,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.1219,
|
|
"mean_token_accuracy": 0.09141435995697975,
|
|
"num_tokens": 637021.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.734615421295166,
|
|
"epoch": 0.28996865203761757,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.2772,
|
|
"mean_token_accuracy": 0.08298610299825668,
|
|
"num_tokens": 646703.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.772846841812134,
|
|
"epoch": 0.2938871473354232,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.1535,
|
|
"mean_token_accuracy": 0.08612042218446732,
|
|
"num_tokens": 655616.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.662965631484985,
|
|
"epoch": 0.29780564263322884,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1548,
|
|
"mean_token_accuracy": 0.09385274946689606,
|
|
"num_tokens": 663783.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.72055869102478,
|
|
"epoch": 0.3017241379310345,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.1499,
|
|
"mean_token_accuracy": 0.09183024540543556,
|
|
"num_tokens": 671855.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.686996603012085,
|
|
"epoch": 0.3056426332288401,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.1691,
|
|
"mean_token_accuracy": 0.08594455868005753,
|
|
"num_tokens": 680981.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.710857486724853,
|
|
"epoch": 0.30956112852664575,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1032,
|
|
"mean_token_accuracy": 0.09159721955657005,
|
|
"num_tokens": 689294.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.561884641647339,
|
|
"epoch": 0.31347962382445144,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.055,
|
|
"mean_token_accuracy": 0.09060370847582817,
|
|
"num_tokens": 697703.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.728968143463135,
|
|
"epoch": 0.31739811912225707,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.1907,
|
|
"mean_token_accuracy": 0.08998276144266129,
|
|
"num_tokens": 706792.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.705010414123535,
|
|
"epoch": 0.3213166144200627,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.0954,
|
|
"mean_token_accuracy": 0.09226684123277665,
|
|
"num_tokens": 715864.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.541502141952515,
|
|
"epoch": 0.32523510971786834,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.1394,
|
|
"mean_token_accuracy": 0.0916991800069809,
|
|
"num_tokens": 723921.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.679799509048462,
|
|
"epoch": 0.329153605015674,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.1503,
|
|
"mean_token_accuracy": 0.08857458978891372,
|
|
"num_tokens": 732797.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.634342527389526,
|
|
"epoch": 0.3330721003134796,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.0126,
|
|
"mean_token_accuracy": 0.0977866668254137,
|
|
"num_tokens": 741248.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.6327766418457035,
|
|
"epoch": 0.33699059561128525,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0961,
|
|
"mean_token_accuracy": 0.09004503935575485,
|
|
"num_tokens": 749766.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.5497727394104,
|
|
"epoch": 0.3409090909090909,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.067,
|
|
"mean_token_accuracy": 0.09081325829029083,
|
|
"num_tokens": 758695.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.6185346126556395,
|
|
"epoch": 0.3448275862068966,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.0591,
|
|
"mean_token_accuracy": 0.08728412315249442,
|
|
"num_tokens": 767624.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.5867267608642575,
|
|
"epoch": 0.3487460815047022,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0721,
|
|
"mean_token_accuracy": 0.09007409512996674,
|
|
"num_tokens": 776616.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.504705095291138,
|
|
"epoch": 0.35266457680250785,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.0826,
|
|
"mean_token_accuracy": 0.09070678725838661,
|
|
"num_tokens": 786147.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.628587007522583,
|
|
"epoch": 0.3565830721003135,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.1945,
|
|
"mean_token_accuracy": 0.09022908806800842,
|
|
"num_tokens": 795213.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.5562244892120365,
|
|
"epoch": 0.3605015673981191,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.0011,
|
|
"mean_token_accuracy": 0.09484408944845199,
|
|
"num_tokens": 803118.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.582311534881592,
|
|
"epoch": 0.36442006269592475,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.9727,
|
|
"mean_token_accuracy": 0.10151090547442436,
|
|
"num_tokens": 811358.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.470248889923096,
|
|
"epoch": 0.3683385579937304,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.971,
|
|
"mean_token_accuracy": 0.09536491632461548,
|
|
"num_tokens": 819653.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.467239570617676,
|
|
"epoch": 0.3722570532915361,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9984,
|
|
"mean_token_accuracy": 0.0897679604589939,
|
|
"num_tokens": 828462.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.6044769287109375,
|
|
"epoch": 0.3761755485893417,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.0876,
|
|
"mean_token_accuracy": 0.09487015083432197,
|
|
"num_tokens": 836204.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.44920597076416,
|
|
"epoch": 0.38009404388714735,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.9686,
|
|
"mean_token_accuracy": 0.09797411188483238,
|
|
"num_tokens": 845032.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.5414040088653564,
|
|
"epoch": 0.384012539184953,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.9796,
|
|
"mean_token_accuracy": 0.09801916852593422,
|
|
"num_tokens": 853324.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.508725357055664,
|
|
"epoch": 0.3879310344827586,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.9449,
|
|
"mean_token_accuracy": 0.10975441187620164,
|
|
"num_tokens": 861640.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.553517627716064,
|
|
"epoch": 0.39184952978056425,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.116,
|
|
"mean_token_accuracy": 0.09244248941540718,
|
|
"num_tokens": 870758.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.39184952978056425,
|
|
"eval_entropy": 7.418469055678493,
|
|
"eval_loss": 7.0935492515563965,
|
|
"eval_mean_token_accuracy": 0.09673699742370798,
|
|
"eval_num_tokens": 870758.0,
|
|
"eval_runtime": 2.2092,
|
|
"eval_samples_per_second": 1865.802,
|
|
"eval_steps_per_second": 233.565,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.561475372314453,
|
|
"epoch": 0.3957680250783699,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.0205,
|
|
"mean_token_accuracy": 0.0910447046160698,
|
|
"num_tokens": 879968.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.430616950988769,
|
|
"epoch": 0.3996865203761755,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.9955,
|
|
"mean_token_accuracy": 0.09514781311154366,
|
|
"num_tokens": 888804.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.443049621582031,
|
|
"epoch": 0.4036050156739812,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9991,
|
|
"mean_token_accuracy": 0.09540855661034583,
|
|
"num_tokens": 898593.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.458410596847534,
|
|
"epoch": 0.40752351097178685,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.9981,
|
|
"mean_token_accuracy": 0.093653604388237,
|
|
"num_tokens": 906873.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.480202293395996,
|
|
"epoch": 0.4114420062695925,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.9857,
|
|
"mean_token_accuracy": 0.10073858648538589,
|
|
"num_tokens": 915080.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.492116975784302,
|
|
"epoch": 0.4153605015673981,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.9989,
|
|
"mean_token_accuracy": 0.09387330785393715,
|
|
"num_tokens": 923741.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.394177532196045,
|
|
"epoch": 0.41927899686520376,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8859,
|
|
"mean_token_accuracy": 0.09964613541960717,
|
|
"num_tokens": 932455.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.348754358291626,
|
|
"epoch": 0.4231974921630094,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8926,
|
|
"mean_token_accuracy": 0.09797894805669785,
|
|
"num_tokens": 941064.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.362092924118042,
|
|
"epoch": 0.427115987460815,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.9206,
|
|
"mean_token_accuracy": 0.09599274471402168,
|
|
"num_tokens": 950261.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.476526355743408,
|
|
"epoch": 0.43103448275862066,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.9869,
|
|
"mean_token_accuracy": 0.09676681011915207,
|
|
"num_tokens": 959151.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.451156759262085,
|
|
"epoch": 0.43495297805642635,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.0111,
|
|
"mean_token_accuracy": 0.09589274004101753,
|
|
"num_tokens": 968441.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.319547843933106,
|
|
"epoch": 0.438871473354232,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8728,
|
|
"mean_token_accuracy": 0.09684987366199493,
|
|
"num_tokens": 977058.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.395937252044678,
|
|
"epoch": 0.4427899686520376,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.8911,
|
|
"mean_token_accuracy": 0.09753472357988358,
|
|
"num_tokens": 986531.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.321355772018433,
|
|
"epoch": 0.44670846394984326,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7391,
|
|
"mean_token_accuracy": 0.10372037217020988,
|
|
"num_tokens": 995081.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.329747819900513,
|
|
"epoch": 0.4506269592476489,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8174,
|
|
"mean_token_accuracy": 0.09918634667992592,
|
|
"num_tokens": 1003459.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.2806806564331055,
|
|
"epoch": 0.45454545454545453,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.8646,
|
|
"mean_token_accuracy": 0.09958613365888595,
|
|
"num_tokens": 1012420.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.450819778442383,
|
|
"epoch": 0.45846394984326017,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.9327,
|
|
"mean_token_accuracy": 0.09935399070382119,
|
|
"num_tokens": 1021198.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.32666244506836,
|
|
"epoch": 0.46238244514106586,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.9064,
|
|
"mean_token_accuracy": 0.10413050577044487,
|
|
"num_tokens": 1030023.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.313800811767578,
|
|
"epoch": 0.4663009404388715,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7756,
|
|
"mean_token_accuracy": 0.10411769896745682,
|
|
"num_tokens": 1038824.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.428035593032837,
|
|
"epoch": 0.4702194357366771,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.0354,
|
|
"mean_token_accuracy": 0.10027154311537742,
|
|
"num_tokens": 1048207.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.444586324691772,
|
|
"epoch": 0.47413793103448276,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.003,
|
|
"mean_token_accuracy": 0.0982631802558899,
|
|
"num_tokens": 1057354.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.326791095733642,
|
|
"epoch": 0.4780564263322884,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.9402,
|
|
"mean_token_accuracy": 0.09771996289491654,
|
|
"num_tokens": 1066825.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.27696385383606,
|
|
"epoch": 0.48197492163009403,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.8595,
|
|
"mean_token_accuracy": 0.09698087573051453,
|
|
"num_tokens": 1075752.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.259910535812378,
|
|
"epoch": 0.48589341692789967,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7766,
|
|
"mean_token_accuracy": 0.10016454383730888,
|
|
"num_tokens": 1084826.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.3767705917358395,
|
|
"epoch": 0.4898119122257053,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.9169,
|
|
"mean_token_accuracy": 0.09827893227338791,
|
|
"num_tokens": 1093425.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.34305853843689,
|
|
"epoch": 0.493730407523511,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.9157,
|
|
"mean_token_accuracy": 0.09747867435216903,
|
|
"num_tokens": 1102619.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.221006727218628,
|
|
"epoch": 0.49764890282131663,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.8191,
|
|
"mean_token_accuracy": 0.10153490155935288,
|
|
"num_tokens": 1111337.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.207894706726075,
|
|
"epoch": 0.5015673981191222,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.655,
|
|
"mean_token_accuracy": 0.10878187417984009,
|
|
"num_tokens": 1119381.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.210852241516113,
|
|
"epoch": 0.5054858934169278,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7069,
|
|
"mean_token_accuracy": 0.10850767269730568,
|
|
"num_tokens": 1127825.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.236933660507202,
|
|
"epoch": 0.5094043887147336,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7371,
|
|
"mean_token_accuracy": 0.10238605439662933,
|
|
"num_tokens": 1136750.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.227217006683349,
|
|
"epoch": 0.5133228840125392,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.783,
|
|
"mean_token_accuracy": 0.10508323460817337,
|
|
"num_tokens": 1145739.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.235933113098144,
|
|
"epoch": 0.5172413793103449,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.8375,
|
|
"mean_token_accuracy": 0.10406330898404122,
|
|
"num_tokens": 1154415.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.198075199127198,
|
|
"epoch": 0.5211598746081505,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7247,
|
|
"mean_token_accuracy": 0.1038141705095768,
|
|
"num_tokens": 1162480.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.198743104934692,
|
|
"epoch": 0.5250783699059561,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.808,
|
|
"mean_token_accuracy": 0.0985616073012352,
|
|
"num_tokens": 1170774.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.186748123168945,
|
|
"epoch": 0.5289968652037618,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.743,
|
|
"mean_token_accuracy": 0.10074765086174012,
|
|
"num_tokens": 1179629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.185740804672241,
|
|
"epoch": 0.5329153605015674,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.8976,
|
|
"mean_token_accuracy": 0.10007678046822548,
|
|
"num_tokens": 1189111.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.288842678070068,
|
|
"epoch": 0.536833855799373,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.8946,
|
|
"mean_token_accuracy": 0.09661665931344032,
|
|
"num_tokens": 1198827.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.352398300170899,
|
|
"epoch": 0.5407523510971787,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7966,
|
|
"mean_token_accuracy": 0.10527415350079536,
|
|
"num_tokens": 1207089.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.145380878448487,
|
|
"epoch": 0.5446708463949843,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.7822,
|
|
"mean_token_accuracy": 0.10294175297021865,
|
|
"num_tokens": 1216509.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.118182611465454,
|
|
"epoch": 0.54858934169279,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6116,
|
|
"mean_token_accuracy": 0.11104098409414291,
|
|
"num_tokens": 1225684.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.15866265296936,
|
|
"epoch": 0.5525078369905956,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.657,
|
|
"mean_token_accuracy": 0.1122810646891594,
|
|
"num_tokens": 1233675.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.050290060043335,
|
|
"epoch": 0.5564263322884012,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6596,
|
|
"mean_token_accuracy": 0.10829117894172668,
|
|
"num_tokens": 1242147.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.144255828857422,
|
|
"epoch": 0.5603448275862069,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.7579,
|
|
"mean_token_accuracy": 0.1056147575378418,
|
|
"num_tokens": 1251127.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.173990964889526,
|
|
"epoch": 0.5642633228840125,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.7293,
|
|
"mean_token_accuracy": 0.1077524796128273,
|
|
"num_tokens": 1260589.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.268736267089844,
|
|
"epoch": 0.5681818181818182,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.7872,
|
|
"mean_token_accuracy": 0.10429466664791107,
|
|
"num_tokens": 1269281.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.0816668510437015,
|
|
"epoch": 0.5721003134796239,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6605,
|
|
"mean_token_accuracy": 0.10658104047179222,
|
|
"num_tokens": 1278033.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.192685079574585,
|
|
"epoch": 0.5760188087774295,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.7278,
|
|
"mean_token_accuracy": 0.1095947191119194,
|
|
"num_tokens": 1285847.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.141543245315551,
|
|
"epoch": 0.5799373040752351,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.7272,
|
|
"mean_token_accuracy": 0.1035495713353157,
|
|
"num_tokens": 1294770.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.305365896224975,
|
|
"epoch": 0.5838557993730408,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.9019,
|
|
"mean_token_accuracy": 0.09915386661887168,
|
|
"num_tokens": 1304282.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.114371252059937,
|
|
"epoch": 0.5877742946708464,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7476,
|
|
"mean_token_accuracy": 0.10341449677944184,
|
|
"num_tokens": 1313218.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.149020004272461,
|
|
"epoch": 0.591692789968652,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.7097,
|
|
"mean_token_accuracy": 0.10334419459104538,
|
|
"num_tokens": 1322093.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.016652154922485,
|
|
"epoch": 0.5956112852664577,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.7458,
|
|
"mean_token_accuracy": 0.10744681730866432,
|
|
"num_tokens": 1330718.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.10248818397522,
|
|
"epoch": 0.5995297805642633,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6178,
|
|
"mean_token_accuracy": 0.10776162892580032,
|
|
"num_tokens": 1339672.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.099648666381836,
|
|
"epoch": 0.603448275862069,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.6488,
|
|
"mean_token_accuracy": 0.1087779700756073,
|
|
"num_tokens": 1348587.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.018001747131348,
|
|
"epoch": 0.6073667711598746,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.6294,
|
|
"mean_token_accuracy": 0.11382819563150406,
|
|
"num_tokens": 1357597.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.176814222335816,
|
|
"epoch": 0.6112852664576802,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.7204,
|
|
"mean_token_accuracy": 0.10900540873408318,
|
|
"num_tokens": 1366246.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.034693956375122,
|
|
"epoch": 0.6152037617554859,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.6284,
|
|
"mean_token_accuracy": 0.10845897570252419,
|
|
"num_tokens": 1374664.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.943607664108276,
|
|
"epoch": 0.6191222570532915,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.642,
|
|
"mean_token_accuracy": 0.10764897763729095,
|
|
"num_tokens": 1382765.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.017871522903443,
|
|
"epoch": 0.6230407523510971,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.6123,
|
|
"mean_token_accuracy": 0.1094035878777504,
|
|
"num_tokens": 1391708.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.066233396530151,
|
|
"epoch": 0.6269592476489029,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.761,
|
|
"mean_token_accuracy": 0.09915644004940986,
|
|
"num_tokens": 1401373.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.068959426879883,
|
|
"epoch": 0.6308777429467085,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5598,
|
|
"mean_token_accuracy": 0.11302306056022644,
|
|
"num_tokens": 1409702.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.03672308921814,
|
|
"epoch": 0.6347962382445141,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.77,
|
|
"mean_token_accuracy": 0.10150963887572288,
|
|
"num_tokens": 1419596.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.083284759521485,
|
|
"epoch": 0.6387147335423198,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.7146,
|
|
"mean_token_accuracy": 0.1047218032181263,
|
|
"num_tokens": 1428437.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.074973106384277,
|
|
"epoch": 0.6426332288401254,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.6087,
|
|
"mean_token_accuracy": 0.10951325818896293,
|
|
"num_tokens": 1436764.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.933263826370239,
|
|
"epoch": 0.646551724137931,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6592,
|
|
"mean_token_accuracy": 0.11087424829602241,
|
|
"num_tokens": 1445755.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.154396057128906,
|
|
"epoch": 0.6504702194357367,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.7092,
|
|
"mean_token_accuracy": 0.10701614245772362,
|
|
"num_tokens": 1454550.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.98425703048706,
|
|
"epoch": 0.6543887147335423,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.5981,
|
|
"mean_token_accuracy": 0.11420469135046005,
|
|
"num_tokens": 1463039.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.988760995864868,
|
|
"epoch": 0.658307210031348,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.6625,
|
|
"mean_token_accuracy": 0.10574135035276414,
|
|
"num_tokens": 1472029.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.009801864624023,
|
|
"epoch": 0.6622257053291536,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.6865,
|
|
"mean_token_accuracy": 0.10918803215026855,
|
|
"num_tokens": 1481588.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.000622129440307,
|
|
"epoch": 0.6661442006269592,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5815,
|
|
"mean_token_accuracy": 0.1104504257440567,
|
|
"num_tokens": 1489915.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.095453310012817,
|
|
"epoch": 0.6700626959247649,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.6823,
|
|
"mean_token_accuracy": 0.10836535915732384,
|
|
"num_tokens": 1497920.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.9481316089630125,
|
|
"epoch": 0.6739811912225705,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5112,
|
|
"mean_token_accuracy": 0.1121792584657669,
|
|
"num_tokens": 1506965.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.970028448104858,
|
|
"epoch": 0.6778996865203761,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.6225,
|
|
"mean_token_accuracy": 0.11289523839950562,
|
|
"num_tokens": 1516028.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.008283472061157,
|
|
"epoch": 0.6818181818181818,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.6409,
|
|
"mean_token_accuracy": 0.10912753567099572,
|
|
"num_tokens": 1524579.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.899019956588745,
|
|
"epoch": 0.6857366771159875,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.5487,
|
|
"mean_token_accuracy": 0.11380631253123283,
|
|
"num_tokens": 1534285.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.978461980819702,
|
|
"epoch": 0.6896551724137931,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5658,
|
|
"mean_token_accuracy": 0.11228583827614784,
|
|
"num_tokens": 1543701.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.062817907333374,
|
|
"epoch": 0.6935736677115988,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.8259,
|
|
"mean_token_accuracy": 0.09782676100730896,
|
|
"num_tokens": 1552715.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.756741905212403,
|
|
"epoch": 0.6974921630094044,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4375,
|
|
"mean_token_accuracy": 0.11271021217107773,
|
|
"num_tokens": 1561523.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.057762336730957,
|
|
"epoch": 0.70141065830721,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.6532,
|
|
"mean_token_accuracy": 0.11261685714125633,
|
|
"num_tokens": 1570057.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.8740034103393555,
|
|
"epoch": 0.7053291536050157,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5135,
|
|
"mean_token_accuracy": 0.11441917642951012,
|
|
"num_tokens": 1578795.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.84917516708374,
|
|
"epoch": 0.7092476489028213,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5937,
|
|
"mean_token_accuracy": 0.11051025837659836,
|
|
"num_tokens": 1588108.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.046537971496582,
|
|
"epoch": 0.713166144200627,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5419,
|
|
"mean_token_accuracy": 0.1160262867808342,
|
|
"num_tokens": 1596545.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.86083025932312,
|
|
"epoch": 0.7170846394984326,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.5415,
|
|
"mean_token_accuracy": 0.11502676978707313,
|
|
"num_tokens": 1605098.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.83633713722229,
|
|
"epoch": 0.7210031347962382,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5629,
|
|
"mean_token_accuracy": 0.11907657384872436,
|
|
"num_tokens": 1613760.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.9062807083129885,
|
|
"epoch": 0.7249216300940439,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.5931,
|
|
"mean_token_accuracy": 0.11272144094109535,
|
|
"num_tokens": 1622345.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.0440794944763185,
|
|
"epoch": 0.7288401253918495,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.5803,
|
|
"mean_token_accuracy": 0.11651012524962426,
|
|
"num_tokens": 1630706.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.893813562393189,
|
|
"epoch": 0.7327586206896551,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.6478,
|
|
"mean_token_accuracy": 0.10516302213072777,
|
|
"num_tokens": 1639880.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.890382957458496,
|
|
"epoch": 0.7366771159874608,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.4892,
|
|
"mean_token_accuracy": 0.11007987186312676,
|
|
"num_tokens": 1648325.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.773949003219604,
|
|
"epoch": 0.7405956112852664,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4465,
|
|
"mean_token_accuracy": 0.11900387927889824,
|
|
"num_tokens": 1656951.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.8440714359283445,
|
|
"epoch": 0.7445141065830722,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.5745,
|
|
"mean_token_accuracy": 0.11141012907028199,
|
|
"num_tokens": 1665832.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.896161079406738,
|
|
"epoch": 0.7484326018808778,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5042,
|
|
"mean_token_accuracy": 0.11246282234787941,
|
|
"num_tokens": 1675116.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.777887010574341,
|
|
"epoch": 0.7523510971786834,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5073,
|
|
"mean_token_accuracy": 0.11693843528628349,
|
|
"num_tokens": 1684195.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.849985933303833,
|
|
"epoch": 0.7562695924764891,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4641,
|
|
"mean_token_accuracy": 0.11287508308887481,
|
|
"num_tokens": 1693025.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.709165954589844,
|
|
"epoch": 0.7601880877742947,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4612,
|
|
"mean_token_accuracy": 0.11437803655862808,
|
|
"num_tokens": 1701210.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.903396844863892,
|
|
"epoch": 0.7641065830721003,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.553,
|
|
"mean_token_accuracy": 0.10555337145924568,
|
|
"num_tokens": 1711026.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.918910503387451,
|
|
"epoch": 0.768025078369906,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.57,
|
|
"mean_token_accuracy": 0.11464387774467469,
|
|
"num_tokens": 1719441.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.7402503967285154,
|
|
"epoch": 0.7719435736677116,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.536,
|
|
"mean_token_accuracy": 0.11087964922189712,
|
|
"num_tokens": 1728300.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.8907630443573,
|
|
"epoch": 0.7758620689655172,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4406,
|
|
"mean_token_accuracy": 0.11694614291191101,
|
|
"num_tokens": 1736364.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.68822078704834,
|
|
"epoch": 0.7797805642633229,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2699,
|
|
"mean_token_accuracy": 0.1223867692053318,
|
|
"num_tokens": 1745043.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.7981610774993895,
|
|
"epoch": 0.7836990595611285,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.11675246804952621,
|
|
"num_tokens": 1754008.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.7836990595611285,
|
|
"eval_entropy": 6.654667783153149,
|
|
"eval_loss": 6.574846267700195,
|
|
"eval_mean_token_accuracy": 0.11710607061205908,
|
|
"eval_num_tokens": 1754008.0,
|
|
"eval_runtime": 2.3667,
|
|
"eval_samples_per_second": 1741.632,
|
|
"eval_steps_per_second": 218.021,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.649208927154541,
|
|
"epoch": 0.7876175548589341,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999998713243189,
|
|
"loss": 6.4846,
|
|
"mean_token_accuracy": 0.11015921831130981,
|
|
"num_tokens": 1763592.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.839963436126709,
|
|
"epoch": 0.7915360501567398,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999993485796164,
|
|
"loss": 6.3468,
|
|
"mean_token_accuracy": 0.11887188702821731,
|
|
"num_tokens": 1773111.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.739857482910156,
|
|
"epoch": 0.7954545454545454,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999984237245962,
|
|
"loss": 6.4146,
|
|
"mean_token_accuracy": 0.12405953705310821,
|
|
"num_tokens": 1781230.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.8473914623260494,
|
|
"epoch": 0.799373040752351,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999970967609109,
|
|
"loss": 6.6375,
|
|
"mean_token_accuracy": 0.11473575830459595,
|
|
"num_tokens": 1790324.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.818466901779175,
|
|
"epoch": 0.8032915360501567,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999953676909322,
|
|
"loss": 6.488,
|
|
"mean_token_accuracy": 0.11022145599126816,
|
|
"num_tokens": 1798849.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.805118274688721,
|
|
"epoch": 0.8072100313479624,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004999932365177502,
|
|
"loss": 6.3716,
|
|
"mean_token_accuracy": 0.1193474993109703,
|
|
"num_tokens": 1808019.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.7946457862854,
|
|
"epoch": 0.8111285266457681,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999907032451735,
|
|
"loss": 6.4944,
|
|
"mean_token_accuracy": 0.11334498226642609,
|
|
"num_tokens": 1816550.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.711936664581299,
|
|
"epoch": 0.8150470219435737,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999877678777296,
|
|
"loss": 6.3788,
|
|
"mean_token_accuracy": 0.12799450531601905,
|
|
"num_tokens": 1825238.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.681297397613525,
|
|
"epoch": 0.8189655172413793,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999844304206645,
|
|
"loss": 6.346,
|
|
"mean_token_accuracy": 0.12015214636921882,
|
|
"num_tokens": 1833645.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.777402448654175,
|
|
"epoch": 0.822884012539185,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999806908799428,
|
|
"loss": 6.4284,
|
|
"mean_token_accuracy": 0.1182809017598629,
|
|
"num_tokens": 1842892.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.647558307647705,
|
|
"epoch": 0.8268025078369906,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999765492622474,
|
|
"loss": 6.4544,
|
|
"mean_token_accuracy": 0.1192950263619423,
|
|
"num_tokens": 1851400.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.854276514053344,
|
|
"epoch": 0.8307210031347962,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999720055749804,
|
|
"loss": 6.4078,
|
|
"mean_token_accuracy": 0.1215155504643917,
|
|
"num_tokens": 1860088.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.762131643295288,
|
|
"epoch": 0.8346394984326019,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999670598262619,
|
|
"loss": 6.4539,
|
|
"mean_token_accuracy": 0.12213057726621628,
|
|
"num_tokens": 1868558.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.676323461532593,
|
|
"epoch": 0.8385579937304075,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999617120249308,
|
|
"loss": 6.4381,
|
|
"mean_token_accuracy": 0.11797192841768264,
|
|
"num_tokens": 1877624.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.867647266387939,
|
|
"epoch": 0.8424764890282131,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999559621805445,
|
|
"loss": 6.5615,
|
|
"mean_token_accuracy": 0.11477635577321052,
|
|
"num_tokens": 1886505.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.589630508422852,
|
|
"epoch": 0.8463949843260188,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999498103033788,
|
|
"loss": 6.2977,
|
|
"mean_token_accuracy": 0.12378846853971481,
|
|
"num_tokens": 1895701.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.666740560531617,
|
|
"epoch": 0.8503134796238244,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999432564044284,
|
|
"loss": 6.2686,
|
|
"mean_token_accuracy": 0.12306156605482102,
|
|
"num_tokens": 1904640.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.720833110809326,
|
|
"epoch": 0.85423197492163,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999363004954058,
|
|
"loss": 6.481,
|
|
"mean_token_accuracy": 0.11864836812019348,
|
|
"num_tokens": 1914216.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.69180383682251,
|
|
"epoch": 0.8581504702194357,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.4162,
|
|
"mean_token_accuracy": 0.12137454971671105,
|
|
"num_tokens": 1923280.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.79173903465271,
|
|
"epoch": 0.8620689655172413,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999211826975884,
|
|
"loss": 6.4387,
|
|
"mean_token_accuracy": 0.12172688096761704,
|
|
"num_tokens": 1931403.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.808387517929077,
|
|
"epoch": 0.8659874608150471,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999130208358114,
|
|
"loss": 6.4484,
|
|
"mean_token_accuracy": 0.1191784381866455,
|
|
"num_tokens": 1940774.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.769060564041138,
|
|
"epoch": 0.8699059561128527,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999044570179983,
|
|
"loss": 6.5459,
|
|
"mean_token_accuracy": 0.1176558330655098,
|
|
"num_tokens": 1950045.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.803036499023437,
|
|
"epoch": 0.8738244514106583,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998954912594541,
|
|
"loss": 6.3989,
|
|
"mean_token_accuracy": 0.1201501503586769,
|
|
"num_tokens": 1959303.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.66001410484314,
|
|
"epoch": 0.877742946708464,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998861235762018,
|
|
"loss": 6.382,
|
|
"mean_token_accuracy": 0.11791504472494126,
|
|
"num_tokens": 1968583.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.7807149410247805,
|
|
"epoch": 0.8816614420062696,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998763539849832,
|
|
"loss": 6.4873,
|
|
"mean_token_accuracy": 0.11481102481484413,
|
|
"num_tokens": 1977489.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.625271320343018,
|
|
"epoch": 0.8855799373040752,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998661825032579,
|
|
"loss": 6.2531,
|
|
"mean_token_accuracy": 0.12325010523200035,
|
|
"num_tokens": 1985308.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.662072563171387,
|
|
"epoch": 0.8894984326018809,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998556091492043,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.11545633301138877,
|
|
"num_tokens": 1993930.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.648345518112182,
|
|
"epoch": 0.8934169278996865,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998446339417184,
|
|
"loss": 6.4231,
|
|
"mean_token_accuracy": 0.11234267503023147,
|
|
"num_tokens": 2003802.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.76238489151001,
|
|
"epoch": 0.8973354231974922,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998332569004149,
|
|
"loss": 6.3499,
|
|
"mean_token_accuracy": 0.1253760874271393,
|
|
"num_tokens": 2011957.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.612128019332886,
|
|
"epoch": 0.9012539184952978,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998214780456263,
|
|
"loss": 6.3359,
|
|
"mean_token_accuracy": 0.12428404837846756,
|
|
"num_tokens": 2020741.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.711992835998535,
|
|
"epoch": 0.9051724137931034,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998092973984033,
|
|
"loss": 6.407,
|
|
"mean_token_accuracy": 0.11499650552868843,
|
|
"num_tokens": 2029804.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.695094060897827,
|
|
"epoch": 0.9090909090909091,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997967149805147,
|
|
"loss": 6.4652,
|
|
"mean_token_accuracy": 0.11959076821804046,
|
|
"num_tokens": 2038732.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.733592510223389,
|
|
"epoch": 0.9130094043887147,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004997837308144474,
|
|
"loss": 6.3533,
|
|
"mean_token_accuracy": 0.12914359271526338,
|
|
"num_tokens": 2046362.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.622471714019776,
|
|
"epoch": 0.9169278996865203,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997703449234062,
|
|
"loss": 6.3397,
|
|
"mean_token_accuracy": 0.11818950101733208,
|
|
"num_tokens": 2056018.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.590902185440063,
|
|
"epoch": 0.920846394984326,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997565573313139,
|
|
"loss": 6.3139,
|
|
"mean_token_accuracy": 0.11889233589172363,
|
|
"num_tokens": 2065077.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.753049659729004,
|
|
"epoch": 0.9247648902821317,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997423680628109,
|
|
"loss": 6.3166,
|
|
"mean_token_accuracy": 0.12146413177251816,
|
|
"num_tokens": 2073405.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.4639513969421385,
|
|
"epoch": 0.9286833855799373,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499727777143256,
|
|
"loss": 6.2857,
|
|
"mean_token_accuracy": 0.12157903984189034,
|
|
"num_tokens": 2081891.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.8065098285675045,
|
|
"epoch": 0.932601880877743,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997127845987255,
|
|
"loss": 6.3537,
|
|
"mean_token_accuracy": 0.1274058498442173,
|
|
"num_tokens": 2090732.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.566525173187256,
|
|
"epoch": 0.9365203761755486,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996973904560135,
|
|
"loss": 6.2968,
|
|
"mean_token_accuracy": 0.128861191123724,
|
|
"num_tokens": 2099352.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.673365211486816,
|
|
"epoch": 0.9404388714733543,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004996815947426317,
|
|
"loss": 6.3191,
|
|
"mean_token_accuracy": 0.1260666698217392,
|
|
"num_tokens": 2107365.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.596097421646118,
|
|
"epoch": 0.9443573667711599,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004996653974868097,
|
|
"loss": 6.1833,
|
|
"mean_token_accuracy": 0.12711638659238816,
|
|
"num_tokens": 2116192.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.5889317989349365,
|
|
"epoch": 0.9482758620689655,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996487987174946,
|
|
"loss": 6.2961,
|
|
"mean_token_accuracy": 0.11961350440979004,
|
|
"num_tokens": 2124410.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.608324480056763,
|
|
"epoch": 0.9521943573667712,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996317984643512,
|
|
"loss": 6.3123,
|
|
"mean_token_accuracy": 0.12350876927375794,
|
|
"num_tokens": 2132799.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.701848888397217,
|
|
"epoch": 0.9561128526645768,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996143967577615,
|
|
"loss": 6.4849,
|
|
"mean_token_accuracy": 0.11310579478740693,
|
|
"num_tokens": 2141589.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.725940990447998,
|
|
"epoch": 0.9600313479623824,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004995965936288254,
|
|
"loss": 6.3573,
|
|
"mean_token_accuracy": 0.12382125556468963,
|
|
"num_tokens": 2150704.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.522915983200074,
|
|
"epoch": 0.9639498432601881,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004995783891093597,
|
|
"loss": 6.2705,
|
|
"mean_token_accuracy": 0.1250034339725971,
|
|
"num_tokens": 2159587.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.510832786560059,
|
|
"epoch": 0.9678683385579937,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499559783231899,
|
|
"loss": 6.2626,
|
|
"mean_token_accuracy": 0.11797072142362594,
|
|
"num_tokens": 2169071.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.600325727462769,
|
|
"epoch": 0.9717868338557993,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995407760296952,
|
|
"loss": 6.3472,
|
|
"mean_token_accuracy": 0.11733860597014427,
|
|
"num_tokens": 2178248.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.62357530593872,
|
|
"epoch": 0.975705329153605,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995213675367169,
|
|
"loss": 6.2451,
|
|
"mean_token_accuracy": 0.11946402341127396,
|
|
"num_tokens": 2187520.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.519882535934448,
|
|
"epoch": 0.9796238244514106,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995015577876504,
|
|
"loss": 6.3099,
|
|
"mean_token_accuracy": 0.11913644820451737,
|
|
"num_tokens": 2196861.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.558454179763794,
|
|
"epoch": 0.9835423197492164,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499481346817899,
|
|
"loss": 6.2324,
|
|
"mean_token_accuracy": 0.12462019845843315,
|
|
"num_tokens": 2205523.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.541610050201416,
|
|
"epoch": 0.987460815047022,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994607346635829,
|
|
"loss": 6.1878,
|
|
"mean_token_accuracy": 0.13304450660943984,
|
|
"num_tokens": 2214291.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.681296253204346,
|
|
"epoch": 0.9913793103448276,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994397213615394,
|
|
"loss": 6.3197,
|
|
"mean_token_accuracy": 0.12296603471040726,
|
|
"num_tokens": 2223093.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.531461000442505,
|
|
"epoch": 0.9952978056426333,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004994183069493225,
|
|
"loss": 6.2564,
|
|
"mean_token_accuracy": 0.12638100013136863,
|
|
"num_tokens": 2231551.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.491727447509765,
|
|
"epoch": 0.9992163009404389,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993964914652034,
|
|
"loss": 6.2405,
|
|
"mean_token_accuracy": 0.1313079461455345,
|
|
"num_tokens": 2240650.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.658728313446045,
|
|
"epoch": 1.0031347962382444,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004993742749481699,
|
|
"loss": 6.186,
|
|
"mean_token_accuracy": 0.12531767934560775,
|
|
"num_tokens": 2249392.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.395848560333252,
|
|
"epoch": 1.0070532915360502,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004993516574379265,
|
|
"loss": 5.9971,
|
|
"mean_token_accuracy": 0.1336176484823227,
|
|
"num_tokens": 2258100.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.498657894134522,
|
|
"epoch": 1.0109717868338557,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004993286389748943,
|
|
"loss": 6.0709,
|
|
"mean_token_accuracy": 0.1330066442489624,
|
|
"num_tokens": 2266661.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.504871654510498,
|
|
"epoch": 1.0148902821316614,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993052196002112,
|
|
"loss": 6.1585,
|
|
"mean_token_accuracy": 0.12869242057204247,
|
|
"num_tokens": 2275475.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.493157625198364,
|
|
"epoch": 1.0188087774294672,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004992813993557312,
|
|
"loss": 6.0818,
|
|
"mean_token_accuracy": 0.132894304394722,
|
|
"num_tokens": 2284077.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.4438941955566404,
|
|
"epoch": 1.0227272727272727,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499257178284025,
|
|
"loss": 6.0829,
|
|
"mean_token_accuracy": 0.12662639245390891,
|
|
"num_tokens": 2293029.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.4581663608551025,
|
|
"epoch": 1.0266457680250785,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004992325564283797,
|
|
"loss": 6.0023,
|
|
"mean_token_accuracy": 0.14118103608489035,
|
|
"num_tokens": 2301574.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.411208295822144,
|
|
"epoch": 1.030564263322884,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004992075338327984,
|
|
"loss": 6.0123,
|
|
"mean_token_accuracy": 0.1320878393948078,
|
|
"num_tokens": 2310366.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.492961502075195,
|
|
"epoch": 1.0344827586206897,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991821105420006,
|
|
"loss": 6.1267,
|
|
"mean_token_accuracy": 0.1311240755021572,
|
|
"num_tokens": 2318943.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.596848249435425,
|
|
"epoch": 1.0384012539184952,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004991562866014219,
|
|
"loss": 6.129,
|
|
"mean_token_accuracy": 0.13712250739336013,
|
|
"num_tokens": 2327173.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.508216285705567,
|
|
"epoch": 1.042319749216301,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.0855,
|
|
"mean_token_accuracy": 0.1303306058049202,
|
|
"num_tokens": 2336194.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.420964479446411,
|
|
"epoch": 1.0462382445141065,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004991034369562438,
|
|
"loss": 6.148,
|
|
"mean_token_accuracy": 0.13162081241607665,
|
|
"num_tokens": 2344721.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.457345342636108,
|
|
"epoch": 1.0501567398119123,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004990764113460953,
|
|
"loss": 6.0766,
|
|
"mean_token_accuracy": 0.13223855197429657,
|
|
"num_tokens": 2353406.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.553745222091675,
|
|
"epoch": 1.0540752351097178,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004990489852750675,
|
|
"loss": 6.2511,
|
|
"mean_token_accuracy": 0.12270163372159004,
|
|
"num_tokens": 2361896.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.4831219673156735,
|
|
"epoch": 1.0579937304075235,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004990211587921751,
|
|
"loss": 6.0737,
|
|
"mean_token_accuracy": 0.1310164101421833,
|
|
"num_tokens": 2371861.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.373387718200684,
|
|
"epoch": 1.061912225705329,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004989929319471487,
|
|
"loss": 5.9984,
|
|
"mean_token_accuracy": 0.13476544693112374,
|
|
"num_tokens": 2381555.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.396108531951905,
|
|
"epoch": 1.0658307210031348,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004989643047904341,
|
|
"loss": 6.0114,
|
|
"mean_token_accuracy": 0.1410997711122036,
|
|
"num_tokens": 2389851.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.388179016113281,
|
|
"epoch": 1.0697492163009406,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004989352773731928,
|
|
"loss": 5.9311,
|
|
"mean_token_accuracy": 0.14064377471804618,
|
|
"num_tokens": 2397764.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.448402547836304,
|
|
"epoch": 1.073667711598746,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004989058497473014,
|
|
"loss": 6.0836,
|
|
"mean_token_accuracy": 0.1399063892662525,
|
|
"num_tokens": 2406688.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.525130748748779,
|
|
"epoch": 1.0775862068965518,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004988760219653518,
|
|
"loss": 6.1549,
|
|
"mean_token_accuracy": 0.131688591837883,
|
|
"num_tokens": 2415475.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.424018096923828,
|
|
"epoch": 1.0815047021943573,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004988457940806513,
|
|
"loss": 6.0973,
|
|
"mean_token_accuracy": 0.1294810250401497,
|
|
"num_tokens": 2425719.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.418878507614136,
|
|
"epoch": 1.085423197492163,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004988151661472218,
|
|
"loss": 6.0063,
|
|
"mean_token_accuracy": 0.13849975615739823,
|
|
"num_tokens": 2435138.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.45744514465332,
|
|
"epoch": 1.0893416927899686,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004987841382198006,
|
|
"loss": 6.1368,
|
|
"mean_token_accuracy": 0.13256047889590264,
|
|
"num_tokens": 2443976.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.278638410568237,
|
|
"epoch": 1.0932601880877744,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987527103538394,
|
|
"loss": 6.0461,
|
|
"mean_token_accuracy": 0.13556990772485733,
|
|
"num_tokens": 2453287.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.505991792678833,
|
|
"epoch": 1.09717868338558,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000498720882605505,
|
|
"loss": 6.1381,
|
|
"mean_token_accuracy": 0.1309444561600685,
|
|
"num_tokens": 2462054.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.364629411697388,
|
|
"epoch": 1.1010971786833856,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004986886550316788,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.13807333409786224,
|
|
"num_tokens": 2470333.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.3567602157592775,
|
|
"epoch": 1.1050156739811912,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004986560276899565,
|
|
"loss": 6.0596,
|
|
"mean_token_accuracy": 0.13362243846058847,
|
|
"num_tokens": 2479621.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.439157104492187,
|
|
"epoch": 1.108934169278997,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004986230006386488,
|
|
"loss": 6.038,
|
|
"mean_token_accuracy": 0.1371801607310772,
|
|
"num_tokens": 2488020.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.429386949539184,
|
|
"epoch": 1.1128526645768024,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004985895739367802,
|
|
"loss": 6.0645,
|
|
"mean_token_accuracy": 0.13344691395759584,
|
|
"num_tokens": 2497779.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.466728544235229,
|
|
"epoch": 1.1167711598746082,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004985557476440895,
|
|
"loss": 6.0183,
|
|
"mean_token_accuracy": 0.13338489159941674,
|
|
"num_tokens": 2506791.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.395694780349731,
|
|
"epoch": 1.1206896551724137,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00049852152182103,
|
|
"loss": 6.1286,
|
|
"mean_token_accuracy": 0.13055293411016464,
|
|
"num_tokens": 2515576.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.488246631622315,
|
|
"epoch": 1.1246081504702194,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004984868965287686,
|
|
"loss": 6.0423,
|
|
"mean_token_accuracy": 0.1341018334031105,
|
|
"num_tokens": 2523933.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.3448042392730715,
|
|
"epoch": 1.1285266457680252,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004984518718291864,
|
|
"loss": 5.983,
|
|
"mean_token_accuracy": 0.13408639580011367,
|
|
"num_tokens": 2532453.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.4296407222747805,
|
|
"epoch": 1.1324451410658307,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004984164477848783,
|
|
"loss": 6.0353,
|
|
"mean_token_accuracy": 0.12955902740359307,
|
|
"num_tokens": 2541149.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.364090204238892,
|
|
"epoch": 1.1363636363636362,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004983806244591528,
|
|
"loss": 6.0056,
|
|
"mean_token_accuracy": 0.1397395424544811,
|
|
"num_tokens": 2550302.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.3148603439331055,
|
|
"epoch": 1.140282131661442,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004983444019160318,
|
|
"loss": 5.8909,
|
|
"mean_token_accuracy": 0.1413780689239502,
|
|
"num_tokens": 2558507.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.429317855834961,
|
|
"epoch": 1.1442006269592477,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004983077802202511,
|
|
"loss": 6.1431,
|
|
"mean_token_accuracy": 0.1290089689195156,
|
|
"num_tokens": 2567487.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.423747301101685,
|
|
"epoch": 1.1481191222570533,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004982707594372595,
|
|
"loss": 6.0036,
|
|
"mean_token_accuracy": 0.1317966252565384,
|
|
"num_tokens": 2576232.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.29382700920105,
|
|
"epoch": 1.152037617554859,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000498233339633219,
|
|
"loss": 5.9924,
|
|
"mean_token_accuracy": 0.13600054606795312,
|
|
"num_tokens": 2585081.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.3161386013031,
|
|
"epoch": 1.1559561128526645,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000498195520875005,
|
|
"loss": 5.9778,
|
|
"mean_token_accuracy": 0.13530001565814018,
|
|
"num_tokens": 2593890.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.392177295684815,
|
|
"epoch": 1.1598746081504703,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004981573032302059,
|
|
"loss": 6.0645,
|
|
"mean_token_accuracy": 0.13251638934016227,
|
|
"num_tokens": 2602011.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.3406764507293705,
|
|
"epoch": 1.1637931034482758,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004981186867671225,
|
|
"loss": 5.9532,
|
|
"mean_token_accuracy": 0.142037995159626,
|
|
"num_tokens": 2610366.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.236144065856934,
|
|
"epoch": 1.1677115987460815,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004980796715547691,
|
|
"loss": 5.958,
|
|
"mean_token_accuracy": 0.13757709935307502,
|
|
"num_tokens": 2619737.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.418478155136109,
|
|
"epoch": 1.171630094043887,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004980402576628717,
|
|
"loss": 5.9928,
|
|
"mean_token_accuracy": 0.13964883610606194,
|
|
"num_tokens": 2628176.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.250125694274902,
|
|
"epoch": 1.1755485893416928,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004980004451618697,
|
|
"loss": 5.9964,
|
|
"mean_token_accuracy": 0.13738328740000724,
|
|
"num_tokens": 2636195.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.1755485893416928,
|
|
"eval_entropy": 6.177109536274459,
|
|
"eval_loss": 6.257157802581787,
|
|
"eval_mean_token_accuracy": 0.1324432172574276,
|
|
"eval_num_tokens": 2636195.0,
|
|
"eval_runtime": 2.1574,
|
|
"eval_samples_per_second": 1910.589,
|
|
"eval_steps_per_second": 239.171,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.4703504085540775,
|
|
"epoch": 1.1794670846394983,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004979602341229144,
|
|
"loss": 6.0796,
|
|
"mean_token_accuracy": 0.13138044998049736,
|
|
"num_tokens": 2645035.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.379116153717041,
|
|
"epoch": 1.183385579937304,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004979196246178694,
|
|
"loss": 6.0416,
|
|
"mean_token_accuracy": 0.1322546310722828,
|
|
"num_tokens": 2653239.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.251059198379517,
|
|
"epoch": 1.1873040752351098,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004978786167193105,
|
|
"loss": 6.0399,
|
|
"mean_token_accuracy": 0.13632970154285431,
|
|
"num_tokens": 2662532.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.3726804733276365,
|
|
"epoch": 1.1912225705329154,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978372105005254,
|
|
"loss": 6.0364,
|
|
"mean_token_accuracy": 0.13286485522985458,
|
|
"num_tokens": 2671867.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.302795553207398,
|
|
"epoch": 1.1951410658307209,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004977954060355139,
|
|
"loss": 5.9888,
|
|
"mean_token_accuracy": 0.13450938537716867,
|
|
"num_tokens": 2680693.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.4019904136657715,
|
|
"epoch": 1.1990595611285266,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004977532033989871,
|
|
"loss": 6.0748,
|
|
"mean_token_accuracy": 0.13225645050406457,
|
|
"num_tokens": 2689228.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.3064361095428465,
|
|
"epoch": 1.2029780564263324,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004977106026663681,
|
|
"loss": 5.9495,
|
|
"mean_token_accuracy": 0.1334281750023365,
|
|
"num_tokens": 2698483.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.389675569534302,
|
|
"epoch": 1.206896551724138,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976676039137914,
|
|
"loss": 6.0646,
|
|
"mean_token_accuracy": 0.12965597957372665,
|
|
"num_tokens": 2706920.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.350861549377441,
|
|
"epoch": 1.2108150470219436,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976242072181026,
|
|
"loss": 5.9682,
|
|
"mean_token_accuracy": 0.14265039190649986,
|
|
"num_tokens": 2715954.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.26219539642334,
|
|
"epoch": 1.2147335423197492,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004975804126568587,
|
|
"loss": 5.9622,
|
|
"mean_token_accuracy": 0.1418926827609539,
|
|
"num_tokens": 2724890.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.404929876327515,
|
|
"epoch": 1.218652037617555,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004975362203083277,
|
|
"loss": 6.0071,
|
|
"mean_token_accuracy": 0.13822806254029274,
|
|
"num_tokens": 2733688.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.361732912063599,
|
|
"epoch": 1.2225705329153604,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004974916302514886,
|
|
"loss": 6.009,
|
|
"mean_token_accuracy": 0.13254594281315804,
|
|
"num_tokens": 2743364.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.254259157180786,
|
|
"epoch": 1.2264890282131662,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.049,
|
|
"mean_token_accuracy": 0.12958745285868645,
|
|
"num_tokens": 2754276.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.3881830215454105,
|
|
"epoch": 1.2304075235109717,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004974012573323545,
|
|
"loss": 5.8374,
|
|
"mean_token_accuracy": 0.14611872881650925,
|
|
"num_tokens": 2762680.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.158905410766602,
|
|
"epoch": 1.2343260188087775,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004973554746315709,
|
|
"loss": 5.9507,
|
|
"mean_token_accuracy": 0.14377902895212175,
|
|
"num_tokens": 2771161.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.413244533538818,
|
|
"epoch": 1.238244514106583,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004973092945455009,
|
|
"loss": 5.9282,
|
|
"mean_token_accuracy": 0.14490692168474198,
|
|
"num_tokens": 2779321.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.223971366882324,
|
|
"epoch": 1.2421630094043887,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004972627171566757,
|
|
"loss": 6.0696,
|
|
"mean_token_accuracy": 0.1372699737548828,
|
|
"num_tokens": 2788425.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.326494312286377,
|
|
"epoch": 1.2460815047021945,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004972157425483368,
|
|
"loss": 5.9409,
|
|
"mean_token_accuracy": 0.14271166324615478,
|
|
"num_tokens": 2796747.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.238678884506226,
|
|
"epoch": 1.25,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004971683708044353,
|
|
"loss": 5.9965,
|
|
"mean_token_accuracy": 0.13859018310904503,
|
|
"num_tokens": 2805335.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.392447566986084,
|
|
"epoch": 1.2539184952978055,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004971206020096323,
|
|
"loss": 5.8832,
|
|
"mean_token_accuracy": 0.14730977714061738,
|
|
"num_tokens": 2813350.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.330147838592529,
|
|
"epoch": 1.2578369905956113,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004970724362492984,
|
|
"loss": 6.055,
|
|
"mean_token_accuracy": 0.13747652769088745,
|
|
"num_tokens": 2821922.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.251412439346313,
|
|
"epoch": 1.261755485893417,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004970238736095135,
|
|
"loss": 5.9803,
|
|
"mean_token_accuracy": 0.14070260748267174,
|
|
"num_tokens": 2830832.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.382120895385742,
|
|
"epoch": 1.2656739811912225,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004969749141770671,
|
|
"loss": 5.9625,
|
|
"mean_token_accuracy": 0.14103439301252366,
|
|
"num_tokens": 2839699.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.262205171585083,
|
|
"epoch": 1.2695924764890283,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004969255580394575,
|
|
"loss": 6.0318,
|
|
"mean_token_accuracy": 0.13988178670406343,
|
|
"num_tokens": 2848293.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.339599609375,
|
|
"epoch": 1.2735109717868338,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000496875805284892,
|
|
"loss": 5.9703,
|
|
"mean_token_accuracy": 0.14059794023633004,
|
|
"num_tokens": 2856468.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.290027809143067,
|
|
"epoch": 1.2774294670846396,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004968256560022871,
|
|
"loss": 5.9114,
|
|
"mean_token_accuracy": 0.1416413113474846,
|
|
"num_tokens": 2865366.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.232825565338135,
|
|
"epoch": 1.281347962382445,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004967751102812676,
|
|
"loss": 5.9728,
|
|
"mean_token_accuracy": 0.14043197259306908,
|
|
"num_tokens": 2873520.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.297229909896851,
|
|
"epoch": 1.2852664576802508,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004967241682121669,
|
|
"loss": 6.0343,
|
|
"mean_token_accuracy": 0.13920704945921897,
|
|
"num_tokens": 2883324.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.26034722328186,
|
|
"epoch": 1.2891849529780564,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004966728298860267,
|
|
"loss": 5.9641,
|
|
"mean_token_accuracy": 0.1391127072274685,
|
|
"num_tokens": 2892200.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.288340997695923,
|
|
"epoch": 1.293103448275862,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004966210953945969,
|
|
"loss": 5.9395,
|
|
"mean_token_accuracy": 0.138271501660347,
|
|
"num_tokens": 2900597.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.338475894927979,
|
|
"epoch": 1.2970219435736676,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004965689648303355,
|
|
"loss": 5.9778,
|
|
"mean_token_accuracy": 0.1388529971241951,
|
|
"num_tokens": 2909869.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.21885986328125,
|
|
"epoch": 1.3009404388714734,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004965164382864083,
|
|
"loss": 5.9989,
|
|
"mean_token_accuracy": 0.13749246522784234,
|
|
"num_tokens": 2919166.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.422362565994263,
|
|
"epoch": 1.3048589341692791,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004964635158566886,
|
|
"loss": 6.005,
|
|
"mean_token_accuracy": 0.13960996866226197,
|
|
"num_tokens": 2927884.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.208298921585083,
|
|
"epoch": 1.3087774294670846,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004964101976357574,
|
|
"loss": 5.9381,
|
|
"mean_token_accuracy": 0.1448947124183178,
|
|
"num_tokens": 2936737.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.331356334686279,
|
|
"epoch": 1.3126959247648902,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000496356483718903,
|
|
"loss": 6.0109,
|
|
"mean_token_accuracy": 0.13827889934182166,
|
|
"num_tokens": 2944203.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.379136466979981,
|
|
"epoch": 1.316614420062696,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004963023742021211,
|
|
"loss": 6.0467,
|
|
"mean_token_accuracy": 0.13718675822019577,
|
|
"num_tokens": 2952639.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.240991640090942,
|
|
"epoch": 1.3205329153605017,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000496247869182114,
|
|
"loss": 5.9536,
|
|
"mean_token_accuracy": 0.13835291489958762,
|
|
"num_tokens": 2961482.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.2600264072418215,
|
|
"epoch": 1.3244514106583072,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004961929687562909,
|
|
"loss": 5.9726,
|
|
"mean_token_accuracy": 0.13861743956804276,
|
|
"num_tokens": 2970927.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.229094076156616,
|
|
"epoch": 1.328369905956113,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004961376730227681,
|
|
"loss": 5.9134,
|
|
"mean_token_accuracy": 0.14266664236783982,
|
|
"num_tokens": 2979329.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.23995304107666,
|
|
"epoch": 1.3322884012539185,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004960819820803675,
|
|
"loss": 5.8961,
|
|
"mean_token_accuracy": 0.14467830434441567,
|
|
"num_tokens": 2986894.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.2356210231781,
|
|
"epoch": 1.3362068965517242,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004960258960286185,
|
|
"loss": 5.8805,
|
|
"mean_token_accuracy": 0.14453305155038834,
|
|
"num_tokens": 2994721.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.30308198928833,
|
|
"epoch": 1.3401253918495297,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004959694149677553,
|
|
"loss": 6.0497,
|
|
"mean_token_accuracy": 0.13306479826569556,
|
|
"num_tokens": 3003165.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.354442501068116,
|
|
"epoch": 1.3440438871473355,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004959125389987193,
|
|
"loss": 6.0155,
|
|
"mean_token_accuracy": 0.13234709352254867,
|
|
"num_tokens": 3012178.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.222462844848633,
|
|
"epoch": 1.347962382445141,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004958552682231567,
|
|
"loss": 5.9087,
|
|
"mean_token_accuracy": 0.14164854660630227,
|
|
"num_tokens": 3021685.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.25830979347229,
|
|
"epoch": 1.3518808777429467,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004957976027434199,
|
|
"loss": 5.9688,
|
|
"mean_token_accuracy": 0.13602245822548867,
|
|
"num_tokens": 3030667.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.260446071624756,
|
|
"epoch": 1.3557993730407523,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004957395426625663,
|
|
"loss": 6.025,
|
|
"mean_token_accuracy": 0.13554641157388686,
|
|
"num_tokens": 3039501.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.239696407318116,
|
|
"epoch": 1.359717868338558,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004956810880843587,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.13451322689652442,
|
|
"num_tokens": 3048547.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.344391298294068,
|
|
"epoch": 1.3636363636363638,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000495622239113265,
|
|
"loss": 5.92,
|
|
"mean_token_accuracy": 0.14174419417977333,
|
|
"num_tokens": 3056521.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.136912250518799,
|
|
"epoch": 1.3675548589341693,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004955629958544577,
|
|
"loss": 5.9009,
|
|
"mean_token_accuracy": 0.13801625072956086,
|
|
"num_tokens": 3066007.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.331682205200195,
|
|
"epoch": 1.3714733542319748,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004955033584138143,
|
|
"loss": 6.0903,
|
|
"mean_token_accuracy": 0.13032267466187478,
|
|
"num_tokens": 3075111.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.204483461380005,
|
|
"epoch": 1.3753918495297806,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004954433268979164,
|
|
"loss": 5.8777,
|
|
"mean_token_accuracy": 0.1512053370475769,
|
|
"num_tokens": 3083766.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.212308073043824,
|
|
"epoch": 1.3793103448275863,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004953829014140502,
|
|
"loss": 5.9519,
|
|
"mean_token_accuracy": 0.14036413207650184,
|
|
"num_tokens": 3092103.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.220862245559692,
|
|
"epoch": 1.3832288401253918,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004953220820702057,
|
|
"loss": 5.9547,
|
|
"mean_token_accuracy": 0.13751930817961694,
|
|
"num_tokens": 3101286.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.402812719345093,
|
|
"epoch": 1.3871473354231976,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004952608689750771,
|
|
"loss": 5.9525,
|
|
"mean_token_accuracy": 0.14038907513022422,
|
|
"num_tokens": 3110554.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.145372581481934,
|
|
"epoch": 1.391065830721003,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004951992622380619,
|
|
"loss": 5.8312,
|
|
"mean_token_accuracy": 0.14609851241111754,
|
|
"num_tokens": 3119265.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.242946434020996,
|
|
"epoch": 1.3949843260188088,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004951372619692615,
|
|
"loss": 5.8717,
|
|
"mean_token_accuracy": 0.1432626023888588,
|
|
"num_tokens": 3127962.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.264250564575195,
|
|
"epoch": 1.3989028213166144,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004950748682794804,
|
|
"loss": 5.9898,
|
|
"mean_token_accuracy": 0.13833442330360413,
|
|
"num_tokens": 3137265.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.144586849212646,
|
|
"epoch": 1.40282131661442,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004950120812802262,
|
|
"loss": 5.8398,
|
|
"mean_token_accuracy": 0.1428191475570202,
|
|
"num_tokens": 3146458.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.2710494041442875,
|
|
"epoch": 1.4067398119122256,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004949489010837095,
|
|
"loss": 5.9742,
|
|
"mean_token_accuracy": 0.13895326256752014,
|
|
"num_tokens": 3155557.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.2062944889068605,
|
|
"epoch": 1.4106583072100314,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 5.8685,
|
|
"mean_token_accuracy": 0.14544058069586754,
|
|
"num_tokens": 3163478.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.247825574874878,
|
|
"epoch": 1.414576802507837,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000494821361551244,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.14252912029623985,
|
|
"num_tokens": 3172632.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.200651741027832,
|
|
"epoch": 1.4184952978056427,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004947570024432291,
|
|
"loss": 5.974,
|
|
"mean_token_accuracy": 0.14242844879627228,
|
|
"num_tokens": 3181325.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.27247805595398,
|
|
"epoch": 1.4224137931034484,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004946922505938189,
|
|
"loss": 6.0187,
|
|
"mean_token_accuracy": 0.1474633678793907,
|
|
"num_tokens": 3189783.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.279434061050415,
|
|
"epoch": 1.426332288401254,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004946271061187354,
|
|
"loss": 5.8715,
|
|
"mean_token_accuracy": 0.13832092061638832,
|
|
"num_tokens": 3198605.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.1585523128509525,
|
|
"epoch": 1.4302507836990594,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004945615691344025,
|
|
"loss": 5.9755,
|
|
"mean_token_accuracy": 0.1351933255791664,
|
|
"num_tokens": 3208061.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.231250381469726,
|
|
"epoch": 1.4341692789968652,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004944956397579453,
|
|
"loss": 5.8805,
|
|
"mean_token_accuracy": 0.14801415875554086,
|
|
"num_tokens": 3216093.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.1551989078521725,
|
|
"epoch": 1.438087774294671,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004944293181071902,
|
|
"loss": 5.8674,
|
|
"mean_token_accuracy": 0.15063873678445816,
|
|
"num_tokens": 3224446.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.286295795440674,
|
|
"epoch": 1.4420062695924765,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004943626043006649,
|
|
"loss": 5.9823,
|
|
"mean_token_accuracy": 0.14128997549414635,
|
|
"num_tokens": 3233398.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.2087263584136965,
|
|
"epoch": 1.4459247648902822,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000494295498457598,
|
|
"loss": 5.9846,
|
|
"mean_token_accuracy": 0.13925029039382936,
|
|
"num_tokens": 3243391.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.208963346481323,
|
|
"epoch": 1.4498432601880877,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004942280006979179,
|
|
"loss": 5.9253,
|
|
"mean_token_accuracy": 0.14278757125139235,
|
|
"num_tokens": 3251954.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.229836082458496,
|
|
"epoch": 1.4537617554858935,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004941601111422546,
|
|
"loss": 5.9383,
|
|
"mean_token_accuracy": 0.14045246094465255,
|
|
"num_tokens": 3261121.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.122421455383301,
|
|
"epoch": 1.457680250783699,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004940918299119375,
|
|
"loss": 5.8503,
|
|
"mean_token_accuracy": 0.14445059597492219,
|
|
"num_tokens": 3270179.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.0559203147888185,
|
|
"epoch": 1.4615987460815048,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004940231571289962,
|
|
"loss": 5.8619,
|
|
"mean_token_accuracy": 0.14289901107549668,
|
|
"num_tokens": 3279749.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.2980797290802,
|
|
"epoch": 1.4655172413793103,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004939540929161603,
|
|
"loss": 5.8807,
|
|
"mean_token_accuracy": 0.14523373246192933,
|
|
"num_tokens": 3288458.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.144744491577148,
|
|
"epoch": 1.469435736677116,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004938846373968586,
|
|
"loss": 5.8635,
|
|
"mean_token_accuracy": 0.142127887904644,
|
|
"num_tokens": 3297231.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.21766209602356,
|
|
"epoch": 1.4733542319749215,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004938147906952194,
|
|
"loss": 5.9983,
|
|
"mean_token_accuracy": 0.1394887737929821,
|
|
"num_tokens": 3306915.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.2247779846191404,
|
|
"epoch": 1.4772727272727273,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00049374455293607,
|
|
"loss": 5.9451,
|
|
"mean_token_accuracy": 0.136509021371603,
|
|
"num_tokens": 3315985.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.283927154541016,
|
|
"epoch": 1.481191222570533,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004936739242449369,
|
|
"loss": 5.9402,
|
|
"mean_token_accuracy": 0.1394059270620346,
|
|
"num_tokens": 3324924.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.20784740447998,
|
|
"epoch": 1.4851097178683386,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004936029047480447,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.14035279899835587,
|
|
"num_tokens": 3333464.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.179001522064209,
|
|
"epoch": 1.489028213166144,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004935314945723171,
|
|
"loss": 5.8552,
|
|
"mean_token_accuracy": 0.1488176554441452,
|
|
"num_tokens": 3342441.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.2065064907073975,
|
|
"epoch": 1.4929467084639498,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004934596938453754,
|
|
"loss": 5.8868,
|
|
"mean_token_accuracy": 0.14286348223686218,
|
|
"num_tokens": 3351338.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.233264207839966,
|
|
"epoch": 1.4968652037617556,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004933875026955391,
|
|
"loss": 5.8533,
|
|
"mean_token_accuracy": 0.14092985838651656,
|
|
"num_tokens": 3360356.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.208340358734131,
|
|
"epoch": 1.500783699059561,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004933149212518258,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.1386915609240532,
|
|
"num_tokens": 3369485.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.215372371673584,
|
|
"epoch": 1.5047021943573666,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004932419496439501,
|
|
"loss": 5.8937,
|
|
"mean_token_accuracy": 0.14970971941947936,
|
|
"num_tokens": 3378355.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.136859512329101,
|
|
"epoch": 1.5086206896551724,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000493168588002324,
|
|
"loss": 5.8095,
|
|
"mean_token_accuracy": 0.1455121897161007,
|
|
"num_tokens": 3386729.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.288068962097168,
|
|
"epoch": 1.5125391849529781,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004930948364580569,
|
|
"loss": 6.04,
|
|
"mean_token_accuracy": 0.1318528264760971,
|
|
"num_tokens": 3395822.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.158831930160522,
|
|
"epoch": 1.5164576802507836,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004930206951429546,
|
|
"loss": 5.9377,
|
|
"mean_token_accuracy": 0.1419134981930256,
|
|
"num_tokens": 3404671.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.203008460998535,
|
|
"epoch": 1.5203761755485894,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004929461641895197,
|
|
"loss": 5.8854,
|
|
"mean_token_accuracy": 0.1436059854924679,
|
|
"num_tokens": 3413289.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.188502645492553,
|
|
"epoch": 1.5242946708463951,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000492871243730951,
|
|
"loss": 5.9214,
|
|
"mean_token_accuracy": 0.14071950167417527,
|
|
"num_tokens": 3421960.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.19671664237976,
|
|
"epoch": 1.5282131661442007,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004927959339011437,
|
|
"loss": 5.9194,
|
|
"mean_token_accuracy": 0.13918522149324417,
|
|
"num_tokens": 3431958.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.057013893127442,
|
|
"epoch": 1.5321316614420062,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004927202348346885,
|
|
"loss": 5.7281,
|
|
"mean_token_accuracy": 0.15232588797807695,
|
|
"num_tokens": 3441438.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.1603470802307125,
|
|
"epoch": 1.536050156739812,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000492644146666872,
|
|
"loss": 5.8508,
|
|
"mean_token_accuracy": 0.14812538623809815,
|
|
"num_tokens": 3450176.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.09682993888855,
|
|
"epoch": 1.5399686520376177,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004925676695336761,
|
|
"loss": 5.8415,
|
|
"mean_token_accuracy": 0.14344199746847153,
|
|
"num_tokens": 3458685.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.182942724227905,
|
|
"epoch": 1.5438871473354232,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004924908035717777,
|
|
"loss": 5.9859,
|
|
"mean_token_accuracy": 0.13745419234037398,
|
|
"num_tokens": 3468362.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.222579002380371,
|
|
"epoch": 1.5478056426332287,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000492413548918549,
|
|
"loss": 5.9162,
|
|
"mean_token_accuracy": 0.14725780487060547,
|
|
"num_tokens": 3476791.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.145301818847656,
|
|
"epoch": 1.5517241379310345,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004923359057120563,
|
|
"loss": 5.8287,
|
|
"mean_token_accuracy": 0.13777379170060158,
|
|
"num_tokens": 3486486.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.0842972755432125,
|
|
"epoch": 1.5556426332288402,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004922578740910608,
|
|
"loss": 5.7895,
|
|
"mean_token_accuracy": 0.1460484504699707,
|
|
"num_tokens": 3496145.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.085639333724975,
|
|
"epoch": 1.5595611285266457,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004921794541950177,
|
|
"loss": 5.7723,
|
|
"mean_token_accuracy": 0.15120832473039628,
|
|
"num_tokens": 3504644.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.1082745552062985,
|
|
"epoch": 1.5634796238244513,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004921006461640758,
|
|
"loss": 5.806,
|
|
"mean_token_accuracy": 0.13911434262990952,
|
|
"num_tokens": 3513126.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.024177742004395,
|
|
"epoch": 1.567398119122257,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000492021450139078,
|
|
"loss": 5.7685,
|
|
"mean_token_accuracy": 0.14868831634521484,
|
|
"num_tokens": 3521505.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.567398119122257,
|
|
"eval_entropy": 5.94463034470876,
|
|
"eval_loss": 6.05142068862915,
|
|
"eval_mean_token_accuracy": 0.14222837957127613,
|
|
"eval_num_tokens": 3521505.0,
|
|
"eval_runtime": 2.3528,
|
|
"eval_samples_per_second": 1751.969,
|
|
"eval_steps_per_second": 219.315,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.21664867401123,
|
|
"epoch": 1.5713166144200628,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004919418662615605,
|
|
"loss": 5.9982,
|
|
"mean_token_accuracy": 0.13932664021849633,
|
|
"num_tokens": 3530008.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.23062653541565,
|
|
"epoch": 1.5752351097178683,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004918618946737525,
|
|
"loss": 5.8992,
|
|
"mean_token_accuracy": 0.1416752226650715,
|
|
"num_tokens": 3538778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.187751579284668,
|
|
"epoch": 1.579153605015674,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004917815355185762,
|
|
"loss": 5.8629,
|
|
"mean_token_accuracy": 0.14034502878785132,
|
|
"num_tokens": 3547108.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.170332574844361,
|
|
"epoch": 1.5830721003134798,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004917007889396464,
|
|
"loss": 5.8958,
|
|
"mean_token_accuracy": 0.13810683488845826,
|
|
"num_tokens": 3556094.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.113199138641358,
|
|
"epoch": 1.5869905956112853,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004916196550812706,
|
|
"loss": 5.7672,
|
|
"mean_token_accuracy": 0.15340544283390045,
|
|
"num_tokens": 3564222.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.000951814651489,
|
|
"epoch": 1.5909090909090908,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004915381340884477,
|
|
"loss": 5.7646,
|
|
"mean_token_accuracy": 0.14757658094167708,
|
|
"num_tokens": 3573123.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.1683708190917965,
|
|
"epoch": 1.5948275862068966,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 5.807,
|
|
"mean_token_accuracy": 0.1480908825993538,
|
|
"num_tokens": 3581462.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.03673357963562,
|
|
"epoch": 1.5987460815047023,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004913739312829181,
|
|
"loss": 5.8201,
|
|
"mean_token_accuracy": 0.14760260954499244,
|
|
"num_tokens": 3590718.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.099112033843994,
|
|
"epoch": 1.6026645768025078,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004912912497636683,
|
|
"loss": 5.8394,
|
|
"mean_token_accuracy": 0.1434948369860649,
|
|
"num_tokens": 3599890.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.237199831008911,
|
|
"epoch": 1.6065830721003134,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004912081816968853,
|
|
"loss": 5.9261,
|
|
"mean_token_accuracy": 0.14199856221675872,
|
|
"num_tokens": 3607927.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.158766651153565,
|
|
"epoch": 1.6105015673981191,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000491124727231025,
|
|
"loss": 5.7923,
|
|
"mean_token_accuracy": 0.15161944478750228,
|
|
"num_tokens": 3615935.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.0325727462768555,
|
|
"epoch": 1.6144200626959249,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004910408865152343,
|
|
"loss": 5.8137,
|
|
"mean_token_accuracy": 0.14737543165683747,
|
|
"num_tokens": 3624475.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.20059061050415,
|
|
"epoch": 1.6183385579937304,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004909566596993498,
|
|
"loss": 6.0119,
|
|
"mean_token_accuracy": 0.13500457927584647,
|
|
"num_tokens": 3633995.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.167231750488281,
|
|
"epoch": 1.622257053291536,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004908720469338988,
|
|
"loss": 5.8802,
|
|
"mean_token_accuracy": 0.1439649850130081,
|
|
"num_tokens": 3643121.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.128624773025512,
|
|
"epoch": 1.6261755485893417,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004907870483700979,
|
|
"loss": 5.8155,
|
|
"mean_token_accuracy": 0.14720201641321182,
|
|
"num_tokens": 3652285.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.079375267028809,
|
|
"epoch": 1.6300940438871474,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004907016641598534,
|
|
"loss": 5.8845,
|
|
"mean_token_accuracy": 0.138520397990942,
|
|
"num_tokens": 3661441.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.1116255760192875,
|
|
"epoch": 1.634012539184953,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004906158944557607,
|
|
"loss": 5.7766,
|
|
"mean_token_accuracy": 0.14789269417524337,
|
|
"num_tokens": 3669364.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.161786890029907,
|
|
"epoch": 1.6379310344827587,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000490529739411104,
|
|
"loss": 5.8637,
|
|
"mean_token_accuracy": 0.14427998960018157,
|
|
"num_tokens": 3677848.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.072094297409057,
|
|
"epoch": 1.6418495297805644,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004904431991798565,
|
|
"loss": 5.7405,
|
|
"mean_token_accuracy": 0.15071966126561165,
|
|
"num_tokens": 3686617.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.140974855422973,
|
|
"epoch": 1.64576802507837,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004903562739166797,
|
|
"loss": 5.7636,
|
|
"mean_token_accuracy": 0.14302895069122315,
|
|
"num_tokens": 3695106.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.993181848526001,
|
|
"epoch": 1.6496865203761755,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004902689637769229,
|
|
"loss": 5.6811,
|
|
"mean_token_accuracy": 0.15598931014537812,
|
|
"num_tokens": 3703167.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.092042684555054,
|
|
"epoch": 1.6536050156739812,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004901812689166237,
|
|
"loss": 5.8199,
|
|
"mean_token_accuracy": 0.14301452562212943,
|
|
"num_tokens": 3711597.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.0771400928497314,
|
|
"epoch": 1.657523510971787,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004900931894925069,
|
|
"loss": 5.9083,
|
|
"mean_token_accuracy": 0.1459727555513382,
|
|
"num_tokens": 3720277.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.055560636520386,
|
|
"epoch": 1.6614420062695925,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004900047256619849,
|
|
"loss": 5.8713,
|
|
"mean_token_accuracy": 0.14490682035684585,
|
|
"num_tokens": 3729832.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.126559925079346,
|
|
"epoch": 1.665360501567398,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004899158775831566,
|
|
"loss": 5.8105,
|
|
"mean_token_accuracy": 0.15142747312784194,
|
|
"num_tokens": 3738390.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.06225266456604,
|
|
"epoch": 1.6692789968652038,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004898266454148081,
|
|
"loss": 5.8769,
|
|
"mean_token_accuracy": 0.14729567617177963,
|
|
"num_tokens": 3747046.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.109605741500855,
|
|
"epoch": 1.6731974921630095,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004897370293164119,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.14484280049800874,
|
|
"num_tokens": 3755818.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.241956520080566,
|
|
"epoch": 1.677115987460815,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004896470294481262,
|
|
"loss": 5.945,
|
|
"mean_token_accuracy": 0.14419927522540094,
|
|
"num_tokens": 3764123.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.041977167129517,
|
|
"epoch": 1.6810344827586206,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004895566459707954,
|
|
"loss": 5.9984,
|
|
"mean_token_accuracy": 0.13666256815195083,
|
|
"num_tokens": 3773105.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.2252052307128904,
|
|
"epoch": 1.6849529780564263,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004894658790459498,
|
|
"loss": 5.8398,
|
|
"mean_token_accuracy": 0.1510275974869728,
|
|
"num_tokens": 3781417.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.213212680816651,
|
|
"epoch": 1.688871473354232,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004893747288358041,
|
|
"loss": 5.9243,
|
|
"mean_token_accuracy": 0.14902326017618178,
|
|
"num_tokens": 3789657.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.0916680812835695,
|
|
"epoch": 1.6927899686520376,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000489283195503259,
|
|
"loss": 5.9402,
|
|
"mean_token_accuracy": 0.14173602908849717,
|
|
"num_tokens": 3798700.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.136766576766968,
|
|
"epoch": 1.6967084639498433,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000489191279211899,
|
|
"loss": 5.781,
|
|
"mean_token_accuracy": 0.15388026386499404,
|
|
"num_tokens": 3807299.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.0979077339172365,
|
|
"epoch": 1.700626959247649,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004890989801259935,
|
|
"loss": 5.8945,
|
|
"mean_token_accuracy": 0.14848074018955232,
|
|
"num_tokens": 3816292.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.1879682540893555,
|
|
"epoch": 1.7045454545454546,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000489006298410496,
|
|
"loss": 5.7973,
|
|
"mean_token_accuracy": 0.14656740576028823,
|
|
"num_tokens": 3825070.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.100113868713379,
|
|
"epoch": 1.70846394984326,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004889132342310438,
|
|
"loss": 5.7876,
|
|
"mean_token_accuracy": 0.1485414281487465,
|
|
"num_tokens": 3834016.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.107239866256714,
|
|
"epoch": 1.7123824451410659,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004888197877539577,
|
|
"loss": 5.839,
|
|
"mean_token_accuracy": 0.1498752199113369,
|
|
"num_tokens": 3842172.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.068277835845947,
|
|
"epoch": 1.7163009404388716,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004887259591462417,
|
|
"loss": 5.7337,
|
|
"mean_token_accuracy": 0.15146998018026353,
|
|
"num_tokens": 3851109.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.074309825897217,
|
|
"epoch": 1.7202194357366771,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004886317485755825,
|
|
"loss": 5.8602,
|
|
"mean_token_accuracy": 0.1496201664209366,
|
|
"num_tokens": 3859656.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.063158750534058,
|
|
"epoch": 1.7241379310344827,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004885371562103498,
|
|
"loss": 5.715,
|
|
"mean_token_accuracy": 0.15688403695821762,
|
|
"num_tokens": 3868698.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.0098330020904545,
|
|
"epoch": 1.7280564263322884,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004884421822195957,
|
|
"loss": 5.8125,
|
|
"mean_token_accuracy": 0.14759075343608857,
|
|
"num_tokens": 3877474.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.1693117141723635,
|
|
"epoch": 1.7319749216300941,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004883468267730538,
|
|
"loss": 5.7782,
|
|
"mean_token_accuracy": 0.14859344214200973,
|
|
"num_tokens": 3886328.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.113722562789917,
|
|
"epoch": 1.7358934169278997,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048825109004114006,
|
|
"loss": 5.7635,
|
|
"mean_token_accuracy": 0.15487073063850404,
|
|
"num_tokens": 3894424.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.032921504974365,
|
|
"epoch": 1.7398119122257052,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004881549721949513,
|
|
"loss": 5.8464,
|
|
"mean_token_accuracy": 0.14409356713294982,
|
|
"num_tokens": 3903746.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.051635980606079,
|
|
"epoch": 1.743730407523511,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004880584734062658,
|
|
"loss": 5.8322,
|
|
"mean_token_accuracy": 0.14758656471967696,
|
|
"num_tokens": 3912033.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.979352283477783,
|
|
"epoch": 1.7476489028213167,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004879615938475425,
|
|
"loss": 5.7211,
|
|
"mean_token_accuracy": 0.14839933067560196,
|
|
"num_tokens": 3920388.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.038036680221557,
|
|
"epoch": 1.7515673981191222,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004878643336919209,
|
|
"loss": 5.9264,
|
|
"mean_token_accuracy": 0.14375157058238983,
|
|
"num_tokens": 3929926.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.258039426803589,
|
|
"epoch": 1.7554858934169277,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004877666931132206,
|
|
"loss": 5.905,
|
|
"mean_token_accuracy": 0.13925225138664246,
|
|
"num_tokens": 3938696.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.053710460662842,
|
|
"epoch": 1.7594043887147337,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004876686722859413,
|
|
"loss": 5.7441,
|
|
"mean_token_accuracy": 0.15748699605464936,
|
|
"num_tokens": 3947537.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.096884107589721,
|
|
"epoch": 1.7633228840125392,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000487570271385262,
|
|
"loss": 5.7367,
|
|
"mean_token_accuracy": 0.1488718293607235,
|
|
"num_tokens": 3956539.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.038178634643555,
|
|
"epoch": 1.7672413793103448,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004874714905870411,
|
|
"loss": 5.7245,
|
|
"mean_token_accuracy": 0.1485554739832878,
|
|
"num_tokens": 3964927.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.008406734466552,
|
|
"epoch": 1.7711598746081505,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004873723300678159,
|
|
"loss": 5.8556,
|
|
"mean_token_accuracy": 0.14758724644780158,
|
|
"num_tokens": 3973565.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.149889707565308,
|
|
"epoch": 1.7750783699059562,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048727279000480226,
|
|
"loss": 5.8708,
|
|
"mean_token_accuracy": 0.14537112414836884,
|
|
"num_tokens": 3982683.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.130380153656006,
|
|
"epoch": 1.7789968652037618,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.782,
|
|
"mean_token_accuracy": 0.1462070681154728,
|
|
"num_tokens": 3991462.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.820031452178955,
|
|
"epoch": 1.7829153605015673,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004870725719596648,
|
|
"loss": 5.5506,
|
|
"mean_token_accuracy": 0.15967189222574235,
|
|
"num_tokens": 3999821.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.128752088546753,
|
|
"epoch": 1.786833855799373,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004869718943353631,
|
|
"loss": 5.8093,
|
|
"mean_token_accuracy": 0.14672327786684036,
|
|
"num_tokens": 4008655.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.08288459777832,
|
|
"epoch": 1.7907523510971788,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048687083788291656,
|
|
"loss": 5.8593,
|
|
"mean_token_accuracy": 0.14925219714641572,
|
|
"num_tokens": 4017317.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.93970217704773,
|
|
"epoch": 1.7946708463949843,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048676940278292953,
|
|
"loss": 5.6694,
|
|
"mean_token_accuracy": 0.1536447986960411,
|
|
"num_tokens": 4025407.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.037428998947144,
|
|
"epoch": 1.7985893416927898,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048666758921668286,
|
|
"loss": 5.6893,
|
|
"mean_token_accuracy": 0.15228061527013778,
|
|
"num_tokens": 4034310.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.041277456283569,
|
|
"epoch": 1.8025078369905956,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048656539736613403,
|
|
"loss": 5.838,
|
|
"mean_token_accuracy": 0.14865779280662536,
|
|
"num_tokens": 4042326.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.9669349670410154,
|
|
"epoch": 1.8064263322884013,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004864628274139164,
|
|
"loss": 5.7103,
|
|
"mean_token_accuracy": 0.1561269626021385,
|
|
"num_tokens": 4051248.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.061789512634277,
|
|
"epoch": 1.8103448275862069,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004863598795433391,
|
|
"loss": 5.8197,
|
|
"mean_token_accuracy": 0.14619538933038712,
|
|
"num_tokens": 4060227.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.179299306869507,
|
|
"epoch": 1.8142633228840124,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048625655393838666,
|
|
"loss": 5.9148,
|
|
"mean_token_accuracy": 0.1406283162534237,
|
|
"num_tokens": 4069700.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.098657703399658,
|
|
"epoch": 1.8181818181818183,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004861528507837186,
|
|
"loss": 5.7929,
|
|
"mean_token_accuracy": 0.15217663645744323,
|
|
"num_tokens": 4079363.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.126735830307007,
|
|
"epoch": 1.8221003134796239,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004860487702646695,
|
|
"loss": 5.8705,
|
|
"mean_token_accuracy": 0.14706421196460723,
|
|
"num_tokens": 4087944.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.996777725219727,
|
|
"epoch": 1.8260188087774294,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004859443125672479,
|
|
"loss": 5.5717,
|
|
"mean_token_accuracy": 0.1707485556602478,
|
|
"num_tokens": 4096431.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.9904883861541744,
|
|
"epoch": 1.8299373040752351,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004858394778781368,
|
|
"loss": 5.7307,
|
|
"mean_token_accuracy": 0.1449565328657627,
|
|
"num_tokens": 4104894.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.016371631622315,
|
|
"epoch": 1.8338557993730409,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004857342663846927,
|
|
"loss": 5.7798,
|
|
"mean_token_accuracy": 0.14995129629969597,
|
|
"num_tokens": 4112878.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.056232166290283,
|
|
"epoch": 1.8377742946708464,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004856286782749456,
|
|
"loss": 5.7905,
|
|
"mean_token_accuracy": 0.15299687385559083,
|
|
"num_tokens": 4122501.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.989445781707763,
|
|
"epoch": 1.841692789968652,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004855227137375986,
|
|
"loss": 5.7562,
|
|
"mean_token_accuracy": 0.14394884705543518,
|
|
"num_tokens": 4131301.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.029977178573608,
|
|
"epoch": 1.8456112852664577,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004854163729620275,
|
|
"loss": 5.8024,
|
|
"mean_token_accuracy": 0.1524229481816292,
|
|
"num_tokens": 4140076.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.027350330352784,
|
|
"epoch": 1.8495297805642634,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004853096561382805,
|
|
"loss": 5.7119,
|
|
"mean_token_accuracy": 0.15929625928401947,
|
|
"num_tokens": 4148728.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.964261531829834,
|
|
"epoch": 1.853448275862069,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004852025634570779,
|
|
"loss": 5.8484,
|
|
"mean_token_accuracy": 0.14873012974858285,
|
|
"num_tokens": 4157520.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.116676235198975,
|
|
"epoch": 1.8573667711598745,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004850950951098116,
|
|
"loss": 5.6647,
|
|
"mean_token_accuracy": 0.1596071884036064,
|
|
"num_tokens": 4165547.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.964800882339477,
|
|
"epoch": 1.8612852664576802,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004849872512885451,
|
|
"loss": 5.8123,
|
|
"mean_token_accuracy": 0.14799876362085343,
|
|
"num_tokens": 4174656.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.023153924942017,
|
|
"epoch": 1.865203761755486,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004848790321860127,
|
|
"loss": 5.8453,
|
|
"mean_token_accuracy": 0.15013710036873817,
|
|
"num_tokens": 4183452.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.080140924453735,
|
|
"epoch": 1.8691222570532915,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048477043799561946,
|
|
"loss": 5.7387,
|
|
"mean_token_accuracy": 0.14741700068116187,
|
|
"num_tokens": 4192306.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.98345947265625,
|
|
"epoch": 1.873040752351097,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004846614689114409,
|
|
"loss": 5.7805,
|
|
"mean_token_accuracy": 0.1505447432398796,
|
|
"num_tokens": 4201201.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.007718801498413,
|
|
"epoch": 1.876959247648903,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004845521251282223,
|
|
"loss": 5.6965,
|
|
"mean_token_accuracy": 0.1586468607187271,
|
|
"num_tokens": 4209374.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.9288530349731445,
|
|
"epoch": 1.8808777429467085,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004844424068413789,
|
|
"loss": 5.669,
|
|
"mean_token_accuracy": 0.1509244903922081,
|
|
"num_tokens": 4217258.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.010879802703857,
|
|
"epoch": 1.884796238244514,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048433231424699504,
|
|
"loss": 5.6986,
|
|
"mean_token_accuracy": 0.15514398068189622,
|
|
"num_tokens": 4226484.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.014545488357544,
|
|
"epoch": 1.8887147335423198,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048422184754182384,
|
|
"loss": 5.7378,
|
|
"mean_token_accuracy": 0.1522963598370552,
|
|
"num_tokens": 4235546.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.065006399154663,
|
|
"epoch": 1.8926332288401255,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004841110069232875,
|
|
"loss": 5.8718,
|
|
"mean_token_accuracy": 0.15006418973207475,
|
|
"num_tokens": 4244549.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.004025554656982,
|
|
"epoch": 1.896551724137931,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048399979258947597,
|
|
"loss": 5.7847,
|
|
"mean_token_accuracy": 0.15459518432617186,
|
|
"num_tokens": 4252918.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.012329864501953,
|
|
"epoch": 1.9004702194357366,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004838882047391474,
|
|
"loss": 5.7637,
|
|
"mean_token_accuracy": 0.1527097150683403,
|
|
"num_tokens": 4261674.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.015922689437867,
|
|
"epoch": 1.9043887147335423,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048377624357172724,
|
|
"loss": 5.6576,
|
|
"mean_token_accuracy": 0.1551213540136814,
|
|
"num_tokens": 4269986.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.988532638549804,
|
|
"epoch": 1.908307210031348,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048366390928730843,
|
|
"loss": 5.7694,
|
|
"mean_token_accuracy": 0.1541934445500374,
|
|
"num_tokens": 4278131.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.0779579162597654,
|
|
"epoch": 1.9122257053291536,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004835512020866504,
|
|
"loss": 5.7708,
|
|
"mean_token_accuracy": 0.15191547721624374,
|
|
"num_tokens": 4287214.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.048527002334595,
|
|
"epoch": 1.9161442006269591,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048343812217117925,
|
|
"loss": 5.6931,
|
|
"mean_token_accuracy": 0.15481389611959456,
|
|
"num_tokens": 4295949.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.00555386543274,
|
|
"epoch": 1.9200626959247649,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048332466974298723,
|
|
"loss": 5.6941,
|
|
"mean_token_accuracy": 0.1512192964553833,
|
|
"num_tokens": 4305014.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.952016401290893,
|
|
"epoch": 1.9239811912225706,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048321084500483203,
|
|
"loss": 5.6598,
|
|
"mean_token_accuracy": 0.158599616587162,
|
|
"num_tokens": 4313560.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.989430856704712,
|
|
"epoch": 1.9278996865203761,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000483096648160137,
|
|
"loss": 5.7628,
|
|
"mean_token_accuracy": 0.15148875415325164,
|
|
"num_tokens": 4323280.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.022593927383423,
|
|
"epoch": 1.9318181818181817,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048298207941299047,
|
|
"loss": 5.7954,
|
|
"mean_token_accuracy": 0.15118271857500076,
|
|
"num_tokens": 4331757.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.891229438781738,
|
|
"epoch": 1.9357366771159876,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048286713896814536,
|
|
"loss": 5.7222,
|
|
"mean_token_accuracy": 0.15107436180114747,
|
|
"num_tokens": 4341453.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.082182884216309,
|
|
"epoch": 1.9396551724137931,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048275182703101877,
|
|
"loss": 5.6774,
|
|
"mean_token_accuracy": 0.1568118989467621,
|
|
"num_tokens": 4349095.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.953945541381836,
|
|
"epoch": 1.9435736677115987,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048263614380769193,
|
|
"loss": 5.7753,
|
|
"mean_token_accuracy": 0.14955031126737595,
|
|
"num_tokens": 4357913.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.984019470214844,
|
|
"epoch": 1.9474921630094044,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048252008950490957,
|
|
"loss": 5.7351,
|
|
"mean_token_accuracy": 0.1511574998497963,
|
|
"num_tokens": 4366586.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.010700225830078,
|
|
"epoch": 1.9514106583072102,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048240366433007935,
|
|
"loss": 5.6973,
|
|
"mean_token_accuracy": 0.15562127381563187,
|
|
"num_tokens": 4375393.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.01687388420105,
|
|
"epoch": 1.9553291536050157,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048228686849127213,
|
|
"loss": 5.7641,
|
|
"mean_token_accuracy": 0.15305927991867066,
|
|
"num_tokens": 4383747.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.010976028442383,
|
|
"epoch": 1.9592476489028212,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004821697021972209,
|
|
"loss": 5.7533,
|
|
"mean_token_accuracy": 0.14949061274528502,
|
|
"num_tokens": 4392085.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 1.9592476489028212,
|
|
"eval_entropy": 5.852172873740972,
|
|
"eval_loss": 5.90216064453125,
|
|
"eval_mean_token_accuracy": 0.150472123248055,
|
|
"eval_num_tokens": 4392085.0,
|
|
"eval_runtime": 2.153,
|
|
"eval_samples_per_second": 1914.513,
|
|
"eval_steps_per_second": 239.663,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.998786163330078,
|
|
"epoch": 1.963166144200627,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.7507,
|
|
"mean_token_accuracy": 0.15378087013959885,
|
|
"num_tokens": 4400843.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.924379634857178,
|
|
"epoch": 1.9670846394984327,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004819342590816288,
|
|
"loss": 5.6964,
|
|
"mean_token_accuracy": 0.14832123517990112,
|
|
"num_tokens": 4409126.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.043420886993408,
|
|
"epoch": 1.9710031347962382,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004818159826808631,
|
|
"loss": 5.8401,
|
|
"mean_token_accuracy": 0.15090397149324417,
|
|
"num_tokens": 4418401.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.983637428283691,
|
|
"epoch": 1.9749216300940438,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004816973366664026,
|
|
"loss": 5.7692,
|
|
"mean_token_accuracy": 0.15161943435668945,
|
|
"num_tokens": 4428141.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.050208282470703,
|
|
"epoch": 1.9788401253918495,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004815783212502871,
|
|
"loss": 5.7798,
|
|
"mean_token_accuracy": 0.14357385337352752,
|
|
"num_tokens": 4437158.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.038489580154419,
|
|
"epoch": 1.9827586206896552,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048145893664521645,
|
|
"loss": 5.837,
|
|
"mean_token_accuracy": 0.14841558784246445,
|
|
"num_tokens": 4445901.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.000403785705567,
|
|
"epoch": 1.9866771159874608,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048133918306455023,
|
|
"loss": 5.6521,
|
|
"mean_token_accuracy": 0.16284386068582535,
|
|
"num_tokens": 4454200.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.967909860610962,
|
|
"epoch": 1.9905956112852663,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004812190607223075,
|
|
"loss": 5.775,
|
|
"mean_token_accuracy": 0.1452760264277458,
|
|
"num_tokens": 4463484.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.84949746131897,
|
|
"epoch": 1.9945141065830723,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048109856983316655,
|
|
"loss": 5.5909,
|
|
"mean_token_accuracy": 0.16197559535503386,
|
|
"num_tokens": 4472659.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.051226043701172,
|
|
"epoch": 1.9984326018808778,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000480977710612464,
|
|
"loss": 5.6749,
|
|
"mean_token_accuracy": 0.14808449968695642,
|
|
"num_tokens": 4481355.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.960011577606201,
|
|
"epoch": 2.0023510971786833,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004808564832761948,
|
|
"loss": 5.5495,
|
|
"mean_token_accuracy": 0.15930166691541672,
|
|
"num_tokens": 4490005.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.927367639541626,
|
|
"epoch": 2.006269592476489,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004807348880410119,
|
|
"loss": 5.3638,
|
|
"mean_token_accuracy": 0.16506589353084564,
|
|
"num_tokens": 4498645.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.892033481597901,
|
|
"epoch": 2.010188087774295,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004806129251242258,
|
|
"loss": 5.4898,
|
|
"mean_token_accuracy": 0.1605222076177597,
|
|
"num_tokens": 4509675.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.883535718917846,
|
|
"epoch": 2.0141065830721003,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048049059474380393,
|
|
"loss": 5.3214,
|
|
"mean_token_accuracy": 0.17261001616716384,
|
|
"num_tokens": 4518463.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.9985833168029785,
|
|
"epoch": 2.018025078369906,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048036789711837047,
|
|
"loss": 5.4597,
|
|
"mean_token_accuracy": 0.1614847019314766,
|
|
"num_tokens": 4527120.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.957412052154541,
|
|
"epoch": 2.0219435736677114,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048024483246720607,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.1617726743221283,
|
|
"num_tokens": 4535962.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.988943815231323,
|
|
"epoch": 2.0258620689655173,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004801214010102472,
|
|
"loss": 5.4533,
|
|
"mean_token_accuracy": 0.15809701830148698,
|
|
"num_tokens": 4545233.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.842477607727051,
|
|
"epoch": 2.029780564263323,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004799976029680858,
|
|
"loss": 5.3138,
|
|
"mean_token_accuracy": 0.16294387280941008,
|
|
"num_tokens": 4554658.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.966088390350341,
|
|
"epoch": 2.0336990595611284,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004798734385619691,
|
|
"loss": 5.5188,
|
|
"mean_token_accuracy": 0.1634453907608986,
|
|
"num_tokens": 4563964.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.87786169052124,
|
|
"epoch": 2.0376175548589344,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000479748908013799,
|
|
"loss": 5.3772,
|
|
"mean_token_accuracy": 0.17351995557546615,
|
|
"num_tokens": 4572865.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.8549463748931885,
|
|
"epoch": 2.04153605015674,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004796240115461319,
|
|
"loss": 5.344,
|
|
"mean_token_accuracy": 0.17395910620689392,
|
|
"num_tokens": 4581021.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.8385937213897705,
|
|
"epoch": 2.0454545454545454,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004794987493821779,
|
|
"loss": 5.4213,
|
|
"mean_token_accuracy": 0.16788042485713958,
|
|
"num_tokens": 4590779.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.889391136169434,
|
|
"epoch": 2.049373040752351,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047937312174580084,
|
|
"loss": 5.4206,
|
|
"mean_token_accuracy": 0.16239412873983383,
|
|
"num_tokens": 4599831.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.0305015563964846,
|
|
"epoch": 2.053291536050157,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004792471288615177,
|
|
"loss": 5.5657,
|
|
"mean_token_accuracy": 0.1568538099527359,
|
|
"num_tokens": 4609264.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.931199312210083,
|
|
"epoch": 2.0572100313479624,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004791207709544981,
|
|
"loss": 5.4693,
|
|
"mean_token_accuracy": 0.16482812762260438,
|
|
"num_tokens": 4617885.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.819200706481934,
|
|
"epoch": 2.061128526645768,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047899404825056424,
|
|
"loss": 5.3852,
|
|
"mean_token_accuracy": 0.168488447368145,
|
|
"num_tokens": 4626536.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.881595706939697,
|
|
"epoch": 2.0650470219435735,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004788669609761901,
|
|
"loss": 5.4785,
|
|
"mean_token_accuracy": 0.16281253546476365,
|
|
"num_tokens": 4635563.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.860688543319702,
|
|
"epoch": 2.0689655172413794,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047873950935850107,
|
|
"loss": 5.3875,
|
|
"mean_token_accuracy": 0.16267673224210738,
|
|
"num_tokens": 4644822.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.907549715042114,
|
|
"epoch": 2.072884012539185,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004786116936252742,
|
|
"loss": 5.3622,
|
|
"mean_token_accuracy": 0.17064258903265,
|
|
"num_tokens": 4652709.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.801628589630127,
|
|
"epoch": 2.0768025078369905,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004784835140049367,
|
|
"loss": 5.4024,
|
|
"mean_token_accuracy": 0.1627717688679695,
|
|
"num_tokens": 4662014.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.945126247406006,
|
|
"epoch": 2.080721003134796,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004783549707265663,
|
|
"loss": 5.3535,
|
|
"mean_token_accuracy": 0.16518666893243789,
|
|
"num_tokens": 4670618.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.870398330688476,
|
|
"epoch": 2.084639498432602,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047822606401989084,
|
|
"loss": 5.5182,
|
|
"mean_token_accuracy": 0.15221553966403006,
|
|
"num_tokens": 4679937.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.8566924095153805,
|
|
"epoch": 2.0885579937304075,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004780967941152873,
|
|
"loss": 5.3182,
|
|
"mean_token_accuracy": 0.17954363077878951,
|
|
"num_tokens": 4688006.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.8244866847991945,
|
|
"epoch": 2.092476489028213,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004779671612437822,
|
|
"loss": 5.2892,
|
|
"mean_token_accuracy": 0.17375630885362625,
|
|
"num_tokens": 4697106.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.754194021224976,
|
|
"epoch": 2.096394984326019,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047783716563705035,
|
|
"loss": 5.4238,
|
|
"mean_token_accuracy": 0.17014558762311935,
|
|
"num_tokens": 4705582.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.8972570419311525,
|
|
"epoch": 2.1003134796238245,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000477706807527415,
|
|
"loss": 5.3503,
|
|
"mean_token_accuracy": 0.17016672492027282,
|
|
"num_tokens": 4713640.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.896171998977661,
|
|
"epoch": 2.10423197492163,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004775760871478472,
|
|
"loss": 5.4766,
|
|
"mean_token_accuracy": 0.1626301109790802,
|
|
"num_tokens": 4722729.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.83009819984436,
|
|
"epoch": 2.1081504702194356,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047744500473196564,
|
|
"loss": 5.3784,
|
|
"mean_token_accuracy": 0.16960960477590561,
|
|
"num_tokens": 4732099.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.9093749046325685,
|
|
"epoch": 2.1120689655172415,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047731356051403556,
|
|
"loss": 5.4022,
|
|
"mean_token_accuracy": 0.1599876195192337,
|
|
"num_tokens": 4741766.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.737287712097168,
|
|
"epoch": 2.115987460815047,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004771817547289693,
|
|
"loss": 5.3155,
|
|
"mean_token_accuracy": 0.16314125210046768,
|
|
"num_tokens": 4750890.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.828125953674316,
|
|
"epoch": 2.1199059561128526,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004770495876123251,
|
|
"loss": 5.4238,
|
|
"mean_token_accuracy": 0.16328950524330138,
|
|
"num_tokens": 4760063.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.899077892303467,
|
|
"epoch": 2.123824451410658,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004769170594003071,
|
|
"loss": 5.4325,
|
|
"mean_token_accuracy": 0.17012127488851547,
|
|
"num_tokens": 4768928.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.765718746185303,
|
|
"epoch": 2.127742946708464,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047678417032976457,
|
|
"loss": 5.3971,
|
|
"mean_token_accuracy": 0.1617774859070778,
|
|
"num_tokens": 4778129.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.842560338973999,
|
|
"epoch": 2.1316614420062696,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004766509206381919,
|
|
"loss": 5.3984,
|
|
"mean_token_accuracy": 0.1708651602268219,
|
|
"num_tokens": 4786759.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.796217775344848,
|
|
"epoch": 2.135579937304075,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004765173105637279,
|
|
"loss": 5.3985,
|
|
"mean_token_accuracy": 0.1680765986442566,
|
|
"num_tokens": 4795355.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.749894285202027,
|
|
"epoch": 2.139498432601881,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047638334034515547,
|
|
"loss": 5.3541,
|
|
"mean_token_accuracy": 0.16741562783718109,
|
|
"num_tokens": 4804030.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.745327663421631,
|
|
"epoch": 2.1434169278996866,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047624901022190106,
|
|
"loss": 5.2963,
|
|
"mean_token_accuracy": 0.18802160918712615,
|
|
"num_tokens": 4812779.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.878271293640137,
|
|
"epoch": 2.147335423197492,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.4387,
|
|
"mean_token_accuracy": 0.16737519353628158,
|
|
"num_tokens": 4822292.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.934808540344238,
|
|
"epoch": 2.1512539184952977,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004759792712222679,
|
|
"loss": 5.463,
|
|
"mean_token_accuracy": 0.166291281580925,
|
|
"num_tokens": 4830524.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.913129043579102,
|
|
"epoch": 2.1551724137931036,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004758438628279565,
|
|
"loss": 5.4851,
|
|
"mean_token_accuracy": 0.169268399477005,
|
|
"num_tokens": 4839328.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.783896112442017,
|
|
"epoch": 2.159090909090909,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047570809549309697,
|
|
"loss": 5.3277,
|
|
"mean_token_accuracy": 0.17981158941984177,
|
|
"num_tokens": 4847711.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.931724786758423,
|
|
"epoch": 2.1630094043887147,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004755719694603275,
|
|
"loss": 5.4747,
|
|
"mean_token_accuracy": 0.16213576793670653,
|
|
"num_tokens": 4856706.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.8394874095916744,
|
|
"epoch": 2.16692789968652,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004754354849729274,
|
|
"loss": 5.5019,
|
|
"mean_token_accuracy": 0.16769517064094544,
|
|
"num_tokens": 4865250.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.815733146667481,
|
|
"epoch": 2.170846394984326,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047529864227481653,
|
|
"loss": 5.4207,
|
|
"mean_token_accuracy": 0.16898391246795655,
|
|
"num_tokens": 4873826.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.929620552062988,
|
|
"epoch": 2.1747648902821317,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000475161441610555,
|
|
"loss": 5.3931,
|
|
"mean_token_accuracy": 0.16994017511606216,
|
|
"num_tokens": 4882631.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.823086500167847,
|
|
"epoch": 2.1786833855799372,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004750238832253427,
|
|
"loss": 5.4463,
|
|
"mean_token_accuracy": 0.16315830498933792,
|
|
"num_tokens": 4891007.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.806562614440918,
|
|
"epoch": 2.1826018808777428,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004748859673650187,
|
|
"loss": 5.3455,
|
|
"mean_token_accuracy": 0.1706724837422371,
|
|
"num_tokens": 4899692.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.780480670928955,
|
|
"epoch": 2.1865203761755487,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047474769427606115,
|
|
"loss": 5.4647,
|
|
"mean_token_accuracy": 0.1650555893778801,
|
|
"num_tokens": 4909354.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.849972772598266,
|
|
"epoch": 2.1904388714733543,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004746090642055863,
|
|
"loss": 5.4049,
|
|
"mean_token_accuracy": 0.16903691589832306,
|
|
"num_tokens": 4918400.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.9036431312561035,
|
|
"epoch": 2.19435736677116,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047447007740134857,
|
|
"loss": 5.5212,
|
|
"mean_token_accuracy": 0.15624323338270188,
|
|
"num_tokens": 4926852.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.83280234336853,
|
|
"epoch": 2.1982758620689653,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047433073411174,
|
|
"loss": 5.4155,
|
|
"mean_token_accuracy": 0.166817806661129,
|
|
"num_tokens": 4935717.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.782654237747193,
|
|
"epoch": 2.2021943573667713,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004741910345857896,
|
|
"loss": 5.3343,
|
|
"mean_token_accuracy": 0.16697602868080139,
|
|
"num_tokens": 4945389.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.858935832977295,
|
|
"epoch": 2.206112852664577,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047405097907316315,
|
|
"loss": 5.4654,
|
|
"mean_token_accuracy": 0.16850972026586533,
|
|
"num_tokens": 4953670.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.793374490737915,
|
|
"epoch": 2.2100313479623823,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004739105678241625,
|
|
"loss": 5.4747,
|
|
"mean_token_accuracy": 0.16831177920103074,
|
|
"num_tokens": 4962383.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.83435435295105,
|
|
"epoch": 2.2139498432601883,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004737698010897253,
|
|
"loss": 5.2798,
|
|
"mean_token_accuracy": 0.17088344246149062,
|
|
"num_tokens": 4970672.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.8983100891113285,
|
|
"epoch": 2.217868338557994,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004736286791214245,
|
|
"loss": 5.4757,
|
|
"mean_token_accuracy": 0.16149610430002212,
|
|
"num_tokens": 4979303.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.844316291809082,
|
|
"epoch": 2.2217868338557993,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047348720217146807,
|
|
"loss": 5.3265,
|
|
"mean_token_accuracy": 0.17194976657629013,
|
|
"num_tokens": 4988322.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.685111141204834,
|
|
"epoch": 2.225705329153605,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047334537049269806,
|
|
"loss": 5.409,
|
|
"mean_token_accuracy": 0.17122900635004043,
|
|
"num_tokens": 4997250.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.876538801193237,
|
|
"epoch": 2.229623824451411,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004732031843385909,
|
|
"loss": 5.4817,
|
|
"mean_token_accuracy": 0.16331940442323684,
|
|
"num_tokens": 5005459.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.81519718170166,
|
|
"epoch": 2.2335423197492164,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004730606439632562,
|
|
"loss": 5.2391,
|
|
"mean_token_accuracy": 0.187631319463253,
|
|
"num_tokens": 5013704.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.6843428134918215,
|
|
"epoch": 2.237460815047022,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004729177496214367,
|
|
"loss": 5.3822,
|
|
"mean_token_accuracy": 0.1647112935781479,
|
|
"num_tokens": 5022545.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.770077419281006,
|
|
"epoch": 2.2413793103448274,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047277450156850767,
|
|
"loss": 5.368,
|
|
"mean_token_accuracy": 0.17814216613769532,
|
|
"num_tokens": 5031133.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.815965747833252,
|
|
"epoch": 2.2452978056426334,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004726309000604768,
|
|
"loss": 5.3904,
|
|
"mean_token_accuracy": 0.16778234392404556,
|
|
"num_tokens": 5039424.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.885768032073974,
|
|
"epoch": 2.249216300940439,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004724869453539832,
|
|
"loss": 5.5892,
|
|
"mean_token_accuracy": 0.15788995027542113,
|
|
"num_tokens": 5048868.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.919712972640991,
|
|
"epoch": 2.2531347962382444,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004723426377062972,
|
|
"loss": 5.4274,
|
|
"mean_token_accuracy": 0.16562917232513427,
|
|
"num_tokens": 5057532.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.835101652145386,
|
|
"epoch": 2.2570532915360504,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047219797737532,
|
|
"loss": 5.4371,
|
|
"mean_token_accuracy": 0.1600351244211197,
|
|
"num_tokens": 5066088.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.766459894180298,
|
|
"epoch": 2.260971786833856,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047205296461958314,
|
|
"loss": 5.3319,
|
|
"mean_token_accuracy": 0.16788178980350493,
|
|
"num_tokens": 5074968.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.730514430999756,
|
|
"epoch": 2.2648902821316614,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047190759969824785,
|
|
"loss": 5.3604,
|
|
"mean_token_accuracy": 0.17571904212236406,
|
|
"num_tokens": 5083605.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.762352228164673,
|
|
"epoch": 2.268808777429467,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047176188287110485,
|
|
"loss": 5.3771,
|
|
"mean_token_accuracy": 0.1716836303472519,
|
|
"num_tokens": 5092182.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.925917482376098,
|
|
"epoch": 2.2727272727272725,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004716158143985737,
|
|
"loss": 5.5465,
|
|
"mean_token_accuracy": 0.15876618176698684,
|
|
"num_tokens": 5100910.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.854732799530029,
|
|
"epoch": 2.2766457680250785,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047146939454170245,
|
|
"loss": 5.4199,
|
|
"mean_token_accuracy": 0.1700390562415123,
|
|
"num_tokens": 5109760.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.836867237091065,
|
|
"epoch": 2.280564263322884,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004713226235621672,
|
|
"loss": 5.4516,
|
|
"mean_token_accuracy": 0.16734010577201844,
|
|
"num_tokens": 5119193.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.75097393989563,
|
|
"epoch": 2.2844827586206895,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004711755017222714,
|
|
"loss": 5.3661,
|
|
"mean_token_accuracy": 0.1686698392033577,
|
|
"num_tokens": 5127480.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.928559589385986,
|
|
"epoch": 2.2884012539184955,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047102802928494563,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.1568053625524044,
|
|
"num_tokens": 5136648.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.803340721130371,
|
|
"epoch": 2.292319749216301,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004708802065137471,
|
|
"loss": 5.318,
|
|
"mean_token_accuracy": 0.17390057891607286,
|
|
"num_tokens": 5144708.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.740973472595215,
|
|
"epoch": 2.2962382445141065,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004707320336728591,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.16566374748945237,
|
|
"num_tokens": 5153388.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.936216545104981,
|
|
"epoch": 2.300156739811912,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004705835110270904,
|
|
"loss": 5.4254,
|
|
"mean_token_accuracy": 0.16602192521095277,
|
|
"num_tokens": 5161874.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.847784757614136,
|
|
"epoch": 2.304075235109718,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047043463884187517,
|
|
"loss": 5.4186,
|
|
"mean_token_accuracy": 0.16694424748420716,
|
|
"num_tokens": 5170284.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.733828592300415,
|
|
"epoch": 2.3079937304075235,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047028541738327207,
|
|
"loss": 5.3988,
|
|
"mean_token_accuracy": 0.16293196380138397,
|
|
"num_tokens": 5179547.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.848221302032471,
|
|
"epoch": 2.311912225705329,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004701358469179641,
|
|
"loss": 5.4078,
|
|
"mean_token_accuracy": 0.17330763339996338,
|
|
"num_tokens": 5188256.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.854952335357666,
|
|
"epoch": 2.3158307210031346,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004699859277132578,
|
|
"loss": 5.5443,
|
|
"mean_token_accuracy": 0.1712660476565361,
|
|
"num_tokens": 5197464.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.81339955329895,
|
|
"epoch": 2.3197492163009406,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046983566003708336,
|
|
"loss": 5.327,
|
|
"mean_token_accuracy": 0.1743195742368698,
|
|
"num_tokens": 5205989.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.778117513656616,
|
|
"epoch": 2.323667711598746,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00046968504415799325,
|
|
"loss": 5.3756,
|
|
"mean_token_accuracy": 0.17124260514974593,
|
|
"num_tokens": 5214490.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.757774209976196,
|
|
"epoch": 2.3275862068965516,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004695340803451625,
|
|
"loss": 5.438,
|
|
"mean_token_accuracy": 0.16912826895713806,
|
|
"num_tokens": 5223335.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.750115585327149,
|
|
"epoch": 2.3315047021943576,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.3602,
|
|
"mean_token_accuracy": 0.1715103030204773,
|
|
"num_tokens": 5231665.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.754814624786377,
|
|
"epoch": 2.335423197492163,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004692311099980878,
|
|
"loss": 5.3576,
|
|
"mean_token_accuracy": 0.17187349796295165,
|
|
"num_tokens": 5239810.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.727986288070679,
|
|
"epoch": 2.3393416927899686,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046907910400530097,
|
|
"loss": 5.3117,
|
|
"mean_token_accuracy": 0.17412266731262208,
|
|
"num_tokens": 5247525.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.645954179763794,
|
|
"epoch": 2.343260188087774,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004689267511616868,
|
|
"loss": 5.3688,
|
|
"mean_token_accuracy": 0.16983558386564254,
|
|
"num_tokens": 5255528.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.775051403045654,
|
|
"epoch": 2.34717868338558,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046877405173952465,
|
|
"loss": 5.3348,
|
|
"mean_token_accuracy": 0.17650954872369767,
|
|
"num_tokens": 5263944.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.813045358657837,
|
|
"epoch": 2.3510971786833856,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000468621006011713,
|
|
"loss": 5.4391,
|
|
"mean_token_accuracy": 0.16434687674045562,
|
|
"num_tokens": 5271724.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.3510971786833856,
|
|
"eval_entropy": 5.565989548845809,
|
|
"eval_loss": 5.846439361572266,
|
|
"eval_mean_token_accuracy": 0.1547132217751224,
|
|
"eval_num_tokens": 5271724.0,
|
|
"eval_runtime": 2.1661,
|
|
"eval_samples_per_second": 1902.976,
|
|
"eval_steps_per_second": 238.218,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.796149730682373,
|
|
"epoch": 2.355015673981191,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046846761425176943,
|
|
"loss": 5.4593,
|
|
"mean_token_accuracy": 0.16462171897292138,
|
|
"num_tokens": 5281199.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.924826192855835,
|
|
"epoch": 2.3589341692789967,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004683138767338299,
|
|
"loss": 5.4733,
|
|
"mean_token_accuracy": 0.1625735267996788,
|
|
"num_tokens": 5289782.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.67628846168518,
|
|
"epoch": 2.3628526645768027,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004681597937326483,
|
|
"loss": 5.2823,
|
|
"mean_token_accuracy": 0.17548982352018355,
|
|
"num_tokens": 5297922.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.723129463195801,
|
|
"epoch": 2.366771159874608,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004680053655235959,
|
|
"loss": 5.4173,
|
|
"mean_token_accuracy": 0.1753725901246071,
|
|
"num_tokens": 5306178.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.771388530731201,
|
|
"epoch": 2.3706896551724137,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004678505923826609,
|
|
"loss": 5.3725,
|
|
"mean_token_accuracy": 0.16879830956459047,
|
|
"num_tokens": 5314936.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.765290832519531,
|
|
"epoch": 2.3746081504702197,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046769547458644817,
|
|
"loss": 5.4249,
|
|
"mean_token_accuracy": 0.16534564942121505,
|
|
"num_tokens": 5323549.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.782919788360596,
|
|
"epoch": 2.378526645768025,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004675400124121782,
|
|
"loss": 5.4184,
|
|
"mean_token_accuracy": 0.16575070172548295,
|
|
"num_tokens": 5332310.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.904462051391602,
|
|
"epoch": 2.3824451410658307,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046738420613768716,
|
|
"loss": 5.539,
|
|
"mean_token_accuracy": 0.1612643212080002,
|
|
"num_tokens": 5340757.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.879794454574585,
|
|
"epoch": 2.3863636363636362,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046722805604142614,
|
|
"loss": 5.5393,
|
|
"mean_token_accuracy": 0.16443511545658113,
|
|
"num_tokens": 5349433.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.787813568115235,
|
|
"epoch": 2.3902821316614418,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046707156240246076,
|
|
"loss": 5.4817,
|
|
"mean_token_accuracy": 0.1583324208855629,
|
|
"num_tokens": 5358943.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.785201740264893,
|
|
"epoch": 2.3942006269592477,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046691472550047027,
|
|
"loss": 5.392,
|
|
"mean_token_accuracy": 0.16532289534807204,
|
|
"num_tokens": 5367782.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.740094423294067,
|
|
"epoch": 2.3981191222570533,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046675754561574783,
|
|
"loss": 5.2838,
|
|
"mean_token_accuracy": 0.17282020002603532,
|
|
"num_tokens": 5376186.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.824817800521851,
|
|
"epoch": 2.402037617554859,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046660002302919933,
|
|
"loss": 5.5255,
|
|
"mean_token_accuracy": 0.15993946492671968,
|
|
"num_tokens": 5384888.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.841602325439453,
|
|
"epoch": 2.4059561128526648,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004664421580223433,
|
|
"loss": 5.4044,
|
|
"mean_token_accuracy": 0.171414378285408,
|
|
"num_tokens": 5393394.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.851276826858521,
|
|
"epoch": 2.4098746081504703,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046628395087730995,
|
|
"loss": 5.4846,
|
|
"mean_token_accuracy": 0.164702008664608,
|
|
"num_tokens": 5402901.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.8428614139556885,
|
|
"epoch": 2.413793103448276,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004661254018768411,
|
|
"loss": 5.4559,
|
|
"mean_token_accuracy": 0.1624145142734051,
|
|
"num_tokens": 5412704.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.788400888442993,
|
|
"epoch": 2.4177115987460813,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004659665113042897,
|
|
"loss": 5.412,
|
|
"mean_token_accuracy": 0.17654994875192642,
|
|
"num_tokens": 5420863.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.70422854423523,
|
|
"epoch": 2.4216300940438873,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004658072794436187,
|
|
"loss": 5.3842,
|
|
"mean_token_accuracy": 0.17131802141666413,
|
|
"num_tokens": 5430002.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.75596137046814,
|
|
"epoch": 2.425548589341693,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046564770657940146,
|
|
"loss": 5.3728,
|
|
"mean_token_accuracy": 0.17176300138235093,
|
|
"num_tokens": 5438424.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.755931997299195,
|
|
"epoch": 2.4294670846394983,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004654877929968205,
|
|
"loss": 5.386,
|
|
"mean_token_accuracy": 0.1691512793302536,
|
|
"num_tokens": 5446970.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.817753076553345,
|
|
"epoch": 2.433385579937304,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004653275389816673,
|
|
"loss": 5.443,
|
|
"mean_token_accuracy": 0.1643589124083519,
|
|
"num_tokens": 5456689.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.729093456268311,
|
|
"epoch": 2.43730407523511,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046516694482034174,
|
|
"loss": 5.3563,
|
|
"mean_token_accuracy": 0.17465108633041382,
|
|
"num_tokens": 5465269.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.762671279907226,
|
|
"epoch": 2.4412225705329154,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046500601079985164,
|
|
"loss": 5.4191,
|
|
"mean_token_accuracy": 0.16873616129159927,
|
|
"num_tokens": 5473930.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.753766679763794,
|
|
"epoch": 2.445141065830721,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004648447372078123,
|
|
"loss": 5.4511,
|
|
"mean_token_accuracy": 0.1727587401866913,
|
|
"num_tokens": 5482885.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.683706140518188,
|
|
"epoch": 2.449059561128527,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004646831243324457,
|
|
"loss": 5.377,
|
|
"mean_token_accuracy": 0.1691886603832245,
|
|
"num_tokens": 5491489.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.814060354232788,
|
|
"epoch": 2.4529780564263324,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004645211724625802,
|
|
"loss": 5.4484,
|
|
"mean_token_accuracy": 0.16421406716108322,
|
|
"num_tokens": 5499694.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.692279481887818,
|
|
"epoch": 2.456896551724138,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046435888188765015,
|
|
"loss": 5.3385,
|
|
"mean_token_accuracy": 0.1722370281815529,
|
|
"num_tokens": 5508037.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.7511965274810795,
|
|
"epoch": 2.4608150470219434,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004641962528976951,
|
|
"loss": 5.3234,
|
|
"mean_token_accuracy": 0.17271196842193604,
|
|
"num_tokens": 5516644.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.739976596832276,
|
|
"epoch": 2.4647335423197494,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004640332857833593,
|
|
"loss": 5.3853,
|
|
"mean_token_accuracy": 0.16930657923221587,
|
|
"num_tokens": 5524789.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.729832649230957,
|
|
"epoch": 2.468652037617555,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046386998083589156,
|
|
"loss": 5.4152,
|
|
"mean_token_accuracy": 0.16972128450870513,
|
|
"num_tokens": 5533739.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.9425112247467045,
|
|
"epoch": 2.4725705329153604,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004637063383471442,
|
|
"loss": 5.4405,
|
|
"mean_token_accuracy": 0.16971082389354705,
|
|
"num_tokens": 5542735.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.704774808883667,
|
|
"epoch": 2.476489028213166,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046354235860957287,
|
|
"loss": 5.28,
|
|
"mean_token_accuracy": 0.17443583756685258,
|
|
"num_tokens": 5551272.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.728410339355468,
|
|
"epoch": 2.480407523510972,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004633780419162361,
|
|
"loss": 5.4731,
|
|
"mean_token_accuracy": 0.16237824112176896,
|
|
"num_tokens": 5559868.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.779831266403198,
|
|
"epoch": 2.4843260188087775,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046321338856079435,
|
|
"loss": 5.2715,
|
|
"mean_token_accuracy": 0.17919818311929703,
|
|
"num_tokens": 5568276.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.759677124023438,
|
|
"epoch": 2.488244514106583,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046304839883750987,
|
|
"loss": 5.3741,
|
|
"mean_token_accuracy": 0.16535062193870545,
|
|
"num_tokens": 5576462.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.759877061843872,
|
|
"epoch": 2.492163009404389,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000462883073041246,
|
|
"loss": 5.4988,
|
|
"mean_token_accuracy": 0.16511365175247192,
|
|
"num_tokens": 5586044.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.63417682647705,
|
|
"epoch": 2.4960815047021945,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004627174114674669,
|
|
"loss": 5.2968,
|
|
"mean_token_accuracy": 0.175435633957386,
|
|
"num_tokens": 5594626.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.817134284973145,
|
|
"epoch": 2.5,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004625514144122365,
|
|
"loss": 5.3545,
|
|
"mean_token_accuracy": 0.17124097496271135,
|
|
"num_tokens": 5603268.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.73789644241333,
|
|
"epoch": 2.5039184952978055,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004623850821722185,
|
|
"loss": 5.3293,
|
|
"mean_token_accuracy": 0.1691139817237854,
|
|
"num_tokens": 5612053.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.739892101287841,
|
|
"epoch": 2.507836990595611,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004622184150446756,
|
|
"loss": 5.5934,
|
|
"mean_token_accuracy": 0.1566196158528328,
|
|
"num_tokens": 5621297.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.793266296386719,
|
|
"epoch": 2.511755485893417,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046205141332746904,
|
|
"loss": 5.408,
|
|
"mean_token_accuracy": 0.1692503049969673,
|
|
"num_tokens": 5629683.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.688299226760864,
|
|
"epoch": 2.5156739811912225,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 5.3424,
|
|
"mean_token_accuracy": 0.169652496278286,
|
|
"num_tokens": 5638707.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.8154294967651365,
|
|
"epoch": 2.519592476489028,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004617164073184987,
|
|
"loss": 5.3757,
|
|
"mean_token_accuracy": 0.17186844795942308,
|
|
"num_tokens": 5648241.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.728038120269775,
|
|
"epoch": 2.523510971786834,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046154840362544496,
|
|
"loss": 5.425,
|
|
"mean_token_accuracy": 0.16608259677886963,
|
|
"num_tokens": 5656743.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.758688020706177,
|
|
"epoch": 2.5274294670846396,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004613800665401466,
|
|
"loss": 5.4053,
|
|
"mean_token_accuracy": 0.16440549939870835,
|
|
"num_tokens": 5666418.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.728449201583862,
|
|
"epoch": 2.531347962382445,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004612113963634493,
|
|
"loss": 5.3603,
|
|
"mean_token_accuracy": 0.16875675171613694,
|
|
"num_tokens": 5675404.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.71140718460083,
|
|
"epoch": 2.535266457680251,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004610423933967938,
|
|
"loss": 5.446,
|
|
"mean_token_accuracy": 0.15746726989746093,
|
|
"num_tokens": 5684699.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.8733460903167725,
|
|
"epoch": 2.5391849529780566,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000460873057942216,
|
|
"loss": 5.4584,
|
|
"mean_token_accuracy": 0.16523150354623795,
|
|
"num_tokens": 5693219.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.830907583236694,
|
|
"epoch": 2.543103448275862,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004607033903023458,
|
|
"loss": 5.4811,
|
|
"mean_token_accuracy": 0.16551702916622163,
|
|
"num_tokens": 5702005.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.830226945877075,
|
|
"epoch": 2.5470219435736676,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046053339078040674,
|
|
"loss": 5.4576,
|
|
"mean_token_accuracy": 0.16286366581916809,
|
|
"num_tokens": 5710715.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.779508304595947,
|
|
"epoch": 2.550940438871473,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004603630596802155,
|
|
"loss": 5.3716,
|
|
"mean_token_accuracy": 0.1688416823744774,
|
|
"num_tokens": 5719444.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.661671829223633,
|
|
"epoch": 2.554858934169279,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004601923973061814,
|
|
"loss": 5.3671,
|
|
"mean_token_accuracy": 0.16215841919183732,
|
|
"num_tokens": 5727755.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.717129802703857,
|
|
"epoch": 2.5587774294670846,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046002140396330575,
|
|
"loss": 5.2939,
|
|
"mean_token_accuracy": 0.17248321771621705,
|
|
"num_tokens": 5736933.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.820798110961914,
|
|
"epoch": 2.56269592476489,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045985007995718154,
|
|
"loss": 5.468,
|
|
"mean_token_accuracy": 0.16719975471496581,
|
|
"num_tokens": 5745831.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.7877552032470705,
|
|
"epoch": 2.566614420062696,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004596784255939923,
|
|
"loss": 5.5034,
|
|
"mean_token_accuracy": 0.16323807686567307,
|
|
"num_tokens": 5755419.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.732455062866211,
|
|
"epoch": 2.5705329153605017,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004595064411805123,
|
|
"loss": 5.3888,
|
|
"mean_token_accuracy": 0.17277422398328782,
|
|
"num_tokens": 5764008.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.719626760482788,
|
|
"epoch": 2.574451410658307,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004593341270241057,
|
|
"loss": 5.2951,
|
|
"mean_token_accuracy": 0.17218087464571,
|
|
"num_tokens": 5772129.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.696467590332031,
|
|
"epoch": 2.5783699059561127,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004591614834327257,
|
|
"loss": 5.4252,
|
|
"mean_token_accuracy": 0.16468510031700134,
|
|
"num_tokens": 5781079.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.725289011001587,
|
|
"epoch": 2.5822884012539182,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045898851071491444,
|
|
"loss": 5.3781,
|
|
"mean_token_accuracy": 0.17338531464338303,
|
|
"num_tokens": 5790444.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.675624036788941,
|
|
"epoch": 2.586206896551724,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004588152091798022,
|
|
"loss": 5.2774,
|
|
"mean_token_accuracy": 0.16970853209495546,
|
|
"num_tokens": 5799828.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.779384756088257,
|
|
"epoch": 2.5901253918495297,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004586415791371069,
|
|
"loss": 5.5329,
|
|
"mean_token_accuracy": 0.16314616203308105,
|
|
"num_tokens": 5808765.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.839154291152954,
|
|
"epoch": 2.5940438871473352,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004584676208971336,
|
|
"loss": 5.3227,
|
|
"mean_token_accuracy": 0.17356155961751937,
|
|
"num_tokens": 5817281.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.686545133590698,
|
|
"epoch": 2.597962382445141,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045829333477077384,
|
|
"loss": 5.3542,
|
|
"mean_token_accuracy": 0.1818431496620178,
|
|
"num_tokens": 5825741.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.760312747955322,
|
|
"epoch": 2.6018808777429467,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004581187210695053,
|
|
"loss": 5.3443,
|
|
"mean_token_accuracy": 0.17580293416976928,
|
|
"num_tokens": 5834466.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.727016830444336,
|
|
"epoch": 2.6057993730407523,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000457943780105391,
|
|
"loss": 5.3624,
|
|
"mean_token_accuracy": 0.17589896470308303,
|
|
"num_tokens": 5843463.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.736898040771484,
|
|
"epoch": 2.6097178683385582,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045776851219107856,
|
|
"loss": 5.4222,
|
|
"mean_token_accuracy": 0.16509513407945633,
|
|
"num_tokens": 5851885.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.770832252502442,
|
|
"epoch": 2.6136363636363638,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045759291763980035,
|
|
"loss": 5.3971,
|
|
"mean_token_accuracy": 0.16737177222967148,
|
|
"num_tokens": 5861537.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.685166120529175,
|
|
"epoch": 2.6175548589341693,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045741699676537227,
|
|
"loss": 5.3476,
|
|
"mean_token_accuracy": 0.1710105285048485,
|
|
"num_tokens": 5870332.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.7655970573425295,
|
|
"epoch": 2.621473354231975,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045724074988219343,
|
|
"loss": 5.5431,
|
|
"mean_token_accuracy": 0.16379230171442033,
|
|
"num_tokens": 5879240.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.693886756896973,
|
|
"epoch": 2.6253918495297803,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045706417730524565,
|
|
"loss": 5.2831,
|
|
"mean_token_accuracy": 0.1666434794664383,
|
|
"num_tokens": 5887961.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.779108667373658,
|
|
"epoch": 2.6293103448275863,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004568872793500927,
|
|
"loss": 5.5151,
|
|
"mean_token_accuracy": 0.17167288511991502,
|
|
"num_tokens": 5896551.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.685304641723633,
|
|
"epoch": 2.633228840125392,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045671005633287986,
|
|
"loss": 5.3075,
|
|
"mean_token_accuracy": 0.17264989167451858,
|
|
"num_tokens": 5905485.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.717501974105835,
|
|
"epoch": 2.6371473354231973,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004565325085703336,
|
|
"loss": 5.3352,
|
|
"mean_token_accuracy": 0.17089089304208754,
|
|
"num_tokens": 5915166.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.748154067993164,
|
|
"epoch": 2.6410658307210033,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004563546363797602,
|
|
"loss": 5.399,
|
|
"mean_token_accuracy": 0.17310827523469924,
|
|
"num_tokens": 5923782.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.753880453109741,
|
|
"epoch": 2.644984326018809,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004561764400790465,
|
|
"loss": 5.4945,
|
|
"mean_token_accuracy": 0.16915059536695481,
|
|
"num_tokens": 5931904.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.77569613456726,
|
|
"epoch": 2.6489028213166144,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045599791998665796,
|
|
"loss": 5.3764,
|
|
"mean_token_accuracy": 0.17558474987745284,
|
|
"num_tokens": 5940067.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.727733182907104,
|
|
"epoch": 2.6528213166144203,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004558190764216389,
|
|
"loss": 5.3844,
|
|
"mean_token_accuracy": 0.17443183064460754,
|
|
"num_tokens": 5948663.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.776641941070556,
|
|
"epoch": 2.656739811912226,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004556399097036119,
|
|
"loss": 5.3814,
|
|
"mean_token_accuracy": 0.17264174818992614,
|
|
"num_tokens": 5957224.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.766242837905883,
|
|
"epoch": 2.6606583072100314,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004554604201527768,
|
|
"loss": 5.4622,
|
|
"mean_token_accuracy": 0.17051523327827453,
|
|
"num_tokens": 5965604.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.748430061340332,
|
|
"epoch": 2.664576802507837,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045528060808991075,
|
|
"loss": 5.2908,
|
|
"mean_token_accuracy": 0.17768048495054245,
|
|
"num_tokens": 5974309.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.685353755950928,
|
|
"epoch": 2.6684952978056424,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004551004738363669,
|
|
"loss": 5.3064,
|
|
"mean_token_accuracy": 0.17040223330259324,
|
|
"num_tokens": 5982315.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.671670150756836,
|
|
"epoch": 2.6724137931034484,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045492001771407434,
|
|
"loss": 5.3392,
|
|
"mean_token_accuracy": 0.1792501240968704,
|
|
"num_tokens": 5990515.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.774185132980347,
|
|
"epoch": 2.676332288401254,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004547392400455374,
|
|
"loss": 5.5019,
|
|
"mean_token_accuracy": 0.16954251676797866,
|
|
"num_tokens": 6000431.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.763256311416626,
|
|
"epoch": 2.6802507836990594,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004545581411538353,
|
|
"loss": 5.4222,
|
|
"mean_token_accuracy": 0.1638216942548752,
|
|
"num_tokens": 6009022.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.770555830001831,
|
|
"epoch": 2.6841692789968654,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045437672136262083,
|
|
"loss": 5.3644,
|
|
"mean_token_accuracy": 0.1686328426003456,
|
|
"num_tokens": 6019330.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.684002304077149,
|
|
"epoch": 2.688087774294671,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004541949809961208,
|
|
"loss": 5.3644,
|
|
"mean_token_accuracy": 0.18156362622976302,
|
|
"num_tokens": 6027492.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.735467100143433,
|
|
"epoch": 2.6920062695924765,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004540129203791346,
|
|
"loss": 5.3763,
|
|
"mean_token_accuracy": 0.1766294062137604,
|
|
"num_tokens": 6036467.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.70538363456726,
|
|
"epoch": 2.695924764890282,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045383053983703413,
|
|
"loss": 5.3395,
|
|
"mean_token_accuracy": 0.17558803260326386,
|
|
"num_tokens": 6045711.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.719822549819947,
|
|
"epoch": 2.6998432601880875,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 5.4198,
|
|
"mean_token_accuracy": 0.16597308069467545,
|
|
"num_tokens": 6054626.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.762411260604859,
|
|
"epoch": 2.7037617554858935,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045346482028183583,
|
|
"loss": 5.4969,
|
|
"mean_token_accuracy": 0.16687073707580566,
|
|
"num_tokens": 6064118.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.756428527832031,
|
|
"epoch": 2.707680250783699,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045328148192233823,
|
|
"loss": 5.3907,
|
|
"mean_token_accuracy": 0.17539616227149962,
|
|
"num_tokens": 6072723.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.739053058624267,
|
|
"epoch": 2.7115987460815045,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004530978249449254,
|
|
"loss": 5.3667,
|
|
"mean_token_accuracy": 0.17589126527309418,
|
|
"num_tokens": 6081535.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.679334306716919,
|
|
"epoch": 2.7155172413793105,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004529138496778222,
|
|
"loss": 5.3304,
|
|
"mean_token_accuracy": 0.17083911299705506,
|
|
"num_tokens": 6090524.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.757283496856689,
|
|
"epoch": 2.719435736677116,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004527295564498222,
|
|
"loss": 5.4162,
|
|
"mean_token_accuracy": 0.1700947180390358,
|
|
"num_tokens": 6099870.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.696364450454712,
|
|
"epoch": 2.7233542319749215,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004525449455902874,
|
|
"loss": 5.3768,
|
|
"mean_token_accuracy": 0.17267284393310547,
|
|
"num_tokens": 6107760.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.717883825302124,
|
|
"epoch": 2.7272727272727275,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004523600174291473,
|
|
"loss": 5.4409,
|
|
"mean_token_accuracy": 0.1661365106701851,
|
|
"num_tokens": 6117135.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.7076338768005375,
|
|
"epoch": 2.731191222570533,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004521747722968985,
|
|
"loss": 5.371,
|
|
"mean_token_accuracy": 0.17119020819664002,
|
|
"num_tokens": 6125711.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.805576276779175,
|
|
"epoch": 2.7351097178683386,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045198921052460396,
|
|
"loss": 5.4547,
|
|
"mean_token_accuracy": 0.16351006776094437,
|
|
"num_tokens": 6134623.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.784835577011108,
|
|
"epoch": 2.739028213166144,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004518033324438928,
|
|
"loss": 5.4354,
|
|
"mean_token_accuracy": 0.1682366296648979,
|
|
"num_tokens": 6143228.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.719631385803223,
|
|
"epoch": 2.7429467084639496,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004516171383869593,
|
|
"loss": 5.4159,
|
|
"mean_token_accuracy": 0.170523701608181,
|
|
"num_tokens": 6152021.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.7429467084639496,
|
|
"eval_entropy": 5.503179455912391,
|
|
"eval_loss": 5.778191089630127,
|
|
"eval_mean_token_accuracy": 0.15960793857744268,
|
|
"eval_num_tokens": 6152021.0,
|
|
"eval_runtime": 2.1553,
|
|
"eval_samples_per_second": 1912.502,
|
|
"eval_steps_per_second": 239.411,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.759955358505249,
|
|
"epoch": 2.7468652037617556,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045143062868656234,
|
|
"loss": 5.4934,
|
|
"mean_token_accuracy": 0.16508397907018663,
|
|
"num_tokens": 6161389.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.796530818939209,
|
|
"epoch": 2.750783699059561,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000451243803676025,
|
|
"loss": 5.3883,
|
|
"mean_token_accuracy": 0.16478795558214188,
|
|
"num_tokens": 6170381.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.694801712036133,
|
|
"epoch": 2.7547021943573666,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045105666368923397,
|
|
"loss": 5.3547,
|
|
"mean_token_accuracy": 0.169776614010334,
|
|
"num_tokens": 6179071.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.654093837738037,
|
|
"epoch": 2.7586206896551726,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045086920906063866,
|
|
"loss": 5.3601,
|
|
"mean_token_accuracy": 0.16542256474494935,
|
|
"num_tokens": 6187896.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.69340934753418,
|
|
"epoch": 2.762539184952978,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045068144012525095,
|
|
"loss": 5.3309,
|
|
"mean_token_accuracy": 0.17853569388389587,
|
|
"num_tokens": 6196328.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.663122367858887,
|
|
"epoch": 2.7664576802507836,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045049335721864426,
|
|
"loss": 5.3717,
|
|
"mean_token_accuracy": 0.16646454036235808,
|
|
"num_tokens": 6204703.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.663711929321289,
|
|
"epoch": 2.7703761755485896,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045030496067695336,
|
|
"loss": 5.3511,
|
|
"mean_token_accuracy": 0.1697184756398201,
|
|
"num_tokens": 6213496.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.718380641937256,
|
|
"epoch": 2.774294670846395,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004501162508368733,
|
|
"loss": 5.3191,
|
|
"mean_token_accuracy": 0.17658223509788512,
|
|
"num_tokens": 6222112.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.733105993270874,
|
|
"epoch": 2.7782131661442007,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004499272280356594,
|
|
"loss": 5.4556,
|
|
"mean_token_accuracy": 0.1580726310610771,
|
|
"num_tokens": 6230288.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.682272434234619,
|
|
"epoch": 2.782131661442006,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004497378926111257,
|
|
"loss": 5.2932,
|
|
"mean_token_accuracy": 0.18290714919567108,
|
|
"num_tokens": 6238608.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.775586271286011,
|
|
"epoch": 2.7860501567398117,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004495482449016456,
|
|
"loss": 5.4166,
|
|
"mean_token_accuracy": 0.17035767585039138,
|
|
"num_tokens": 6247902.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.7183362483978275,
|
|
"epoch": 2.7899686520376177,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004493582852461501,
|
|
"loss": 5.4035,
|
|
"mean_token_accuracy": 0.17043172270059587,
|
|
"num_tokens": 6257433.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.659321737289429,
|
|
"epoch": 2.793887147335423,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004491680139841281,
|
|
"loss": 5.3615,
|
|
"mean_token_accuracy": 0.16800241619348527,
|
|
"num_tokens": 6266017.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.749320793151855,
|
|
"epoch": 2.7978056426332287,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000448977431455625,
|
|
"loss": 5.464,
|
|
"mean_token_accuracy": 0.16183699816465377,
|
|
"num_tokens": 6274759.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.729684782028198,
|
|
"epoch": 2.8017241379310347,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044878653800124285,
|
|
"loss": 5.3233,
|
|
"mean_token_accuracy": 0.1744695082306862,
|
|
"num_tokens": 6283699.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.702778005599976,
|
|
"epoch": 2.80564263322884,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004485953339621391,
|
|
"loss": 5.3577,
|
|
"mean_token_accuracy": 0.17059250324964523,
|
|
"num_tokens": 6292562.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.691700839996338,
|
|
"epoch": 2.8095611285266457,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004484038196800265,
|
|
"loss": 5.3224,
|
|
"mean_token_accuracy": 0.17630542814731598,
|
|
"num_tokens": 6300928.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.617705392837524,
|
|
"epoch": 2.8134796238244513,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004482119954971719,
|
|
"loss": 5.2628,
|
|
"mean_token_accuracy": 0.17675767689943314,
|
|
"num_tokens": 6309994.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.725002098083496,
|
|
"epoch": 2.817398119122257,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044801986175639635,
|
|
"loss": 5.3708,
|
|
"mean_token_accuracy": 0.17235179543495177,
|
|
"num_tokens": 6318856.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.6887524127960205,
|
|
"epoch": 2.8213166144200628,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004478274188010741,
|
|
"loss": 5.3739,
|
|
"mean_token_accuracy": 0.17109404951334,
|
|
"num_tokens": 6327274.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.758692073822021,
|
|
"epoch": 2.8252351097178683,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044763466697513173,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.17634273022413255,
|
|
"num_tokens": 6335957.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.75085129737854,
|
|
"epoch": 2.829153605015674,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044744160662304805,
|
|
"loss": 5.4554,
|
|
"mean_token_accuracy": 0.17372822165489196,
|
|
"num_tokens": 6344673.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.671528577804565,
|
|
"epoch": 2.83307210031348,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044724823808985325,
|
|
"loss": 5.3129,
|
|
"mean_token_accuracy": 0.17819716781377792,
|
|
"num_tokens": 6353085.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.6792402267456055,
|
|
"epoch": 2.8369905956112853,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004470545617211283,
|
|
"loss": 5.3573,
|
|
"mean_token_accuracy": 0.16664835661649705,
|
|
"num_tokens": 6362107.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.722587394714355,
|
|
"epoch": 2.840909090909091,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044686057786300423,
|
|
"loss": 5.3579,
|
|
"mean_token_accuracy": 0.172071173787117,
|
|
"num_tokens": 6370381.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.742739725112915,
|
|
"epoch": 2.844827586206897,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044666628686216154,
|
|
"loss": 5.3976,
|
|
"mean_token_accuracy": 0.16450393199920654,
|
|
"num_tokens": 6378790.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.638479137420655,
|
|
"epoch": 2.8487460815047023,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044647168906583,
|
|
"loss": 5.3821,
|
|
"mean_token_accuracy": 0.17199081927537918,
|
|
"num_tokens": 6387997.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.764104747772217,
|
|
"epoch": 2.852664576802508,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044627678482178716,
|
|
"loss": 5.3744,
|
|
"mean_token_accuracy": 0.17393434047698975,
|
|
"num_tokens": 6397458.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.723266649246216,
|
|
"epoch": 2.8565830721003134,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004460815744783587,
|
|
"loss": 5.3375,
|
|
"mean_token_accuracy": 0.17309237867593766,
|
|
"num_tokens": 6405481.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.739022874832154,
|
|
"epoch": 2.860501567398119,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000445886058384417,
|
|
"loss": 5.447,
|
|
"mean_token_accuracy": 0.16460947543382645,
|
|
"num_tokens": 6414087.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.731280946731568,
|
|
"epoch": 2.864420062695925,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004456902368893811,
|
|
"loss": 5.2764,
|
|
"mean_token_accuracy": 0.18395242393016814,
|
|
"num_tokens": 6421934.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.691809463500976,
|
|
"epoch": 2.8683385579937304,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004454941103432158,
|
|
"loss": 5.2975,
|
|
"mean_token_accuracy": 0.18017626851797103,
|
|
"num_tokens": 6431028.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.570631980895996,
|
|
"epoch": 2.872257053291536,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044529767909643093,
|
|
"loss": 5.2526,
|
|
"mean_token_accuracy": 0.17766658067703248,
|
|
"num_tokens": 6439913.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.694689750671387,
|
|
"epoch": 2.876175548589342,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004451009435000811,
|
|
"loss": 5.3346,
|
|
"mean_token_accuracy": 0.17320116609334946,
|
|
"num_tokens": 6449161.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.725321054458618,
|
|
"epoch": 2.8800940438871474,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004449039039057647,
|
|
"loss": 5.4444,
|
|
"mean_token_accuracy": 0.16738225221633912,
|
|
"num_tokens": 6457834.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.6329692840576175,
|
|
"epoch": 2.884012539184953,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 5.2676,
|
|
"mean_token_accuracy": 0.17993104457855225,
|
|
"num_tokens": 6465879.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.649481678009034,
|
|
"epoch": 2.887931034482759,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004445089141323415,
|
|
"loss": 5.2974,
|
|
"mean_token_accuracy": 0.1735406592488289,
|
|
"num_tokens": 6474801.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.68170747756958,
|
|
"epoch": 2.8918495297805644,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044431096465914554,
|
|
"loss": 5.3169,
|
|
"mean_token_accuracy": 0.18297179639339448,
|
|
"num_tokens": 6483532.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.671950244903565,
|
|
"epoch": 2.89576802507837,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044411271259980315,
|
|
"loss": 5.2889,
|
|
"mean_token_accuracy": 0.17287395447492598,
|
|
"num_tokens": 6491372.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.635216426849365,
|
|
"epoch": 2.8996865203761755,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004439141583086231,
|
|
"loss": 5.3057,
|
|
"mean_token_accuracy": 0.16837496757507325,
|
|
"num_tokens": 6500001.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.7364266872406,
|
|
"epoch": 2.903605015673981,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044371530214045395,
|
|
"loss": 5.4227,
|
|
"mean_token_accuracy": 0.17135694921016692,
|
|
"num_tokens": 6507750.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.594133281707764,
|
|
"epoch": 2.907523510971787,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044351614445068413,
|
|
"loss": 5.2788,
|
|
"mean_token_accuracy": 0.17326867282390596,
|
|
"num_tokens": 6517888.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.734909772872925,
|
|
"epoch": 2.9114420062695925,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044331668559524043,
|
|
"loss": 5.4418,
|
|
"mean_token_accuracy": 0.1657553344964981,
|
|
"num_tokens": 6526573.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.667497396469116,
|
|
"epoch": 2.915360501567398,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004431169259305883,
|
|
"loss": 5.3557,
|
|
"mean_token_accuracy": 0.17305893301963807,
|
|
"num_tokens": 6536544.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.736414432525635,
|
|
"epoch": 2.919278996865204,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004429168658137306,
|
|
"loss": 5.3498,
|
|
"mean_token_accuracy": 0.16646120846271514,
|
|
"num_tokens": 6544982.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.6487133502960205,
|
|
"epoch": 2.9231974921630095,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044271650560220746,
|
|
"loss": 5.3823,
|
|
"mean_token_accuracy": 0.17397093623876572,
|
|
"num_tokens": 6553664.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.735430669784546,
|
|
"epoch": 2.927115987460815,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044251584565409464,
|
|
"loss": 5.3703,
|
|
"mean_token_accuracy": 0.1720076635479927,
|
|
"num_tokens": 6562234.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.638150548934936,
|
|
"epoch": 2.9310344827586206,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004423148863280044,
|
|
"loss": 5.3336,
|
|
"mean_token_accuracy": 0.1707358330488205,
|
|
"num_tokens": 6570707.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.703876399993897,
|
|
"epoch": 2.934952978056426,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044211362798308334,
|
|
"loss": 5.3803,
|
|
"mean_token_accuracy": 0.1716936409473419,
|
|
"num_tokens": 6580054.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.713050317764282,
|
|
"epoch": 2.938871473354232,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004419120709790129,
|
|
"loss": 5.4153,
|
|
"mean_token_accuracy": 0.16757915019989014,
|
|
"num_tokens": 6588585.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.782528829574585,
|
|
"epoch": 2.9427899686520376,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044171021567600814,
|
|
"loss": 5.4319,
|
|
"mean_token_accuracy": 0.16953148692846298,
|
|
"num_tokens": 6597290.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.711483097076416,
|
|
"epoch": 2.946708463949843,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044150806243481715,
|
|
"loss": 5.3559,
|
|
"mean_token_accuracy": 0.16957012861967086,
|
|
"num_tokens": 6604966.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.712342119216919,
|
|
"epoch": 2.950626959247649,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004413056116167206,
|
|
"loss": 5.3298,
|
|
"mean_token_accuracy": 0.17532417327165603,
|
|
"num_tokens": 6613588.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.766825151443482,
|
|
"epoch": 2.9545454545454546,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004411028635835309,
|
|
"loss": 5.5459,
|
|
"mean_token_accuracy": 0.15694651901721954,
|
|
"num_tokens": 6622907.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.694950580596924,
|
|
"epoch": 2.95846394984326,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004408998186975917,
|
|
"loss": 5.3345,
|
|
"mean_token_accuracy": 0.17050072848796843,
|
|
"num_tokens": 6632074.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.735626316070556,
|
|
"epoch": 2.962382445141066,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044069647732177696,
|
|
"loss": 5.3537,
|
|
"mean_token_accuracy": 0.17172691971063614,
|
|
"num_tokens": 6640595.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.791773843765259,
|
|
"epoch": 2.9663009404388716,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044049283981949103,
|
|
"loss": 5.4297,
|
|
"mean_token_accuracy": 0.1690357431769371,
|
|
"num_tokens": 6649861.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.775791120529175,
|
|
"epoch": 2.970219435736677,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004402889065546667,
|
|
"loss": 5.4449,
|
|
"mean_token_accuracy": 0.16524050682783126,
|
|
"num_tokens": 6659157.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.712988376617432,
|
|
"epoch": 2.9741379310344827,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044008467789176625,
|
|
"loss": 5.4099,
|
|
"mean_token_accuracy": 0.1672254279255867,
|
|
"num_tokens": 6667979.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.686794567108154,
|
|
"epoch": 2.978056426332288,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004398801541957791,
|
|
"loss": 5.3584,
|
|
"mean_token_accuracy": 0.16950906366109847,
|
|
"num_tokens": 6677239.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.711395311355591,
|
|
"epoch": 2.981974921630094,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004396753358322223,
|
|
"loss": 5.4653,
|
|
"mean_token_accuracy": 0.16263094246387483,
|
|
"num_tokens": 6687317.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.699147701263428,
|
|
"epoch": 2.9858934169278997,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004394702231671396,
|
|
"loss": 5.3577,
|
|
"mean_token_accuracy": 0.1647719219326973,
|
|
"num_tokens": 6695732.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.713489723205567,
|
|
"epoch": 2.989811912225705,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004392648165671004,
|
|
"loss": 5.283,
|
|
"mean_token_accuracy": 0.17886848598718644,
|
|
"num_tokens": 6704272.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.639943695068359,
|
|
"epoch": 2.993730407523511,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004390591163991998,
|
|
"loss": 5.3171,
|
|
"mean_token_accuracy": 0.17430887818336488,
|
|
"num_tokens": 6713355.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.719687986373901,
|
|
"epoch": 2.9976489028213167,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043885312303105725,
|
|
"loss": 5.4111,
|
|
"mean_token_accuracy": 0.16944750100374223,
|
|
"num_tokens": 6721952.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.6525568008422855,
|
|
"epoch": 3.001567398119122,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004386468368308163,
|
|
"loss": 5.2259,
|
|
"mean_token_accuracy": 0.1783342808485031,
|
|
"num_tokens": 6730582.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.7358222007751465,
|
|
"epoch": 3.0054858934169277,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004384402581671438,
|
|
"loss": 5.0352,
|
|
"mean_token_accuracy": 0.18735815584659576,
|
|
"num_tokens": 6739370.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.668080472946167,
|
|
"epoch": 3.0094043887147337,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004382333874092295,
|
|
"loss": 5.0093,
|
|
"mean_token_accuracy": 0.1892308622598648,
|
|
"num_tokens": 6748282.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.728878593444824,
|
|
"epoch": 3.0133228840125392,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043802622492678466,
|
|
"loss": 5.1417,
|
|
"mean_token_accuracy": 0.18341654688119888,
|
|
"num_tokens": 6757622.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.758401250839233,
|
|
"epoch": 3.0172413793103448,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004378187710900426,
|
|
"loss": 5.0651,
|
|
"mean_token_accuracy": 0.18538413792848588,
|
|
"num_tokens": 6765780.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.635838031768799,
|
|
"epoch": 3.0211598746081503,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043761102626975674,
|
|
"loss": 5.0683,
|
|
"mean_token_accuracy": 0.19026492685079574,
|
|
"num_tokens": 6774343.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.62908787727356,
|
|
"epoch": 3.0250783699059562,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004374029908372007,
|
|
"loss": 5.0837,
|
|
"mean_token_accuracy": 0.18939932584762573,
|
|
"num_tokens": 6783765.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.785342741012573,
|
|
"epoch": 3.0289968652037618,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043719466516416774,
|
|
"loss": 5.1176,
|
|
"mean_token_accuracy": 0.18569748252630233,
|
|
"num_tokens": 6792102.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.69318470954895,
|
|
"epoch": 3.0329153605015673,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043698604962296946,
|
|
"loss": 5.0745,
|
|
"mean_token_accuracy": 0.18498681783676146,
|
|
"num_tokens": 6801435.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.627175331115723,
|
|
"epoch": 3.0368338557993733,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043677714458643566,
|
|
"loss": 5.0772,
|
|
"mean_token_accuracy": 0.1850498288869858,
|
|
"num_tokens": 6810409.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.555335474014282,
|
|
"epoch": 3.040752351097179,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043656795042791357,
|
|
"loss": 4.957,
|
|
"mean_token_accuracy": 0.20043091475963593,
|
|
"num_tokens": 6818752.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.626818370819092,
|
|
"epoch": 3.0446708463949843,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004363584675212671,
|
|
"loss": 5.0788,
|
|
"mean_token_accuracy": 0.18593218624591829,
|
|
"num_tokens": 6827880.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.609489059448242,
|
|
"epoch": 3.04858934169279,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004361486962408761,
|
|
"loss": 5.0408,
|
|
"mean_token_accuracy": 0.19073140919208526,
|
|
"num_tokens": 6835865.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.584641170501709,
|
|
"epoch": 3.052507836990596,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004359386369616359,
|
|
"loss": 5.0129,
|
|
"mean_token_accuracy": 0.1990236148238182,
|
|
"num_tokens": 6843851.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.608503246307373,
|
|
"epoch": 3.0564263322884013,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004357282900589565,
|
|
"loss": 5.0579,
|
|
"mean_token_accuracy": 0.19194671213626863,
|
|
"num_tokens": 6852649.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.603591775894165,
|
|
"epoch": 3.060344827586207,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043551765590876183,
|
|
"loss": 5.0087,
|
|
"mean_token_accuracy": 0.19188642650842666,
|
|
"num_tokens": 6861607.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.608678674697876,
|
|
"epoch": 3.0642633228840124,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004353067348874894,
|
|
"loss": 5.048,
|
|
"mean_token_accuracy": 0.18362130373716354,
|
|
"num_tokens": 6870650.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.655573892593384,
|
|
"epoch": 3.0681818181818183,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 5.0795,
|
|
"mean_token_accuracy": 0.18299010246992112,
|
|
"num_tokens": 6879820.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.546123886108399,
|
|
"epoch": 3.072100313479624,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004348840337400233,
|
|
"loss": 5.0028,
|
|
"mean_token_accuracy": 0.19229621440172195,
|
|
"num_tokens": 6888329.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.573421239852905,
|
|
"epoch": 3.0760188087774294,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043467225436926517,
|
|
"loss": 5.0756,
|
|
"mean_token_accuracy": 0.1873006194829941,
|
|
"num_tokens": 6897635.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.6603357791900635,
|
|
"epoch": 3.079937304075235,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004344601896382988,
|
|
"loss": 5.1188,
|
|
"mean_token_accuracy": 0.19175236225128173,
|
|
"num_tokens": 6907275.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.672836446762085,
|
|
"epoch": 3.083855799373041,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043424783992611837,
|
|
"loss": 5.0648,
|
|
"mean_token_accuracy": 0.19108632802963257,
|
|
"num_tokens": 6916377.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.6168616771697994,
|
|
"epoch": 3.0877742946708464,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043403520561222705,
|
|
"loss": 4.9707,
|
|
"mean_token_accuracy": 0.19896020144224166,
|
|
"num_tokens": 6924355.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.559317302703858,
|
|
"epoch": 3.091692789968652,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004338222870766371,
|
|
"loss": 5.0967,
|
|
"mean_token_accuracy": 0.18889932930469513,
|
|
"num_tokens": 6933328.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.665522050857544,
|
|
"epoch": 3.0956112852664575,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043360908469986827,
|
|
"loss": 5.0482,
|
|
"mean_token_accuracy": 0.19506864845752717,
|
|
"num_tokens": 6942189.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.613420104980468,
|
|
"epoch": 3.0995297805642634,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004333955988629478,
|
|
"loss": 5.0323,
|
|
"mean_token_accuracy": 0.1941305860877037,
|
|
"num_tokens": 6951188.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.532266330718994,
|
|
"epoch": 3.103448275862069,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043318182994740945,
|
|
"loss": 4.9883,
|
|
"mean_token_accuracy": 0.20233253240585328,
|
|
"num_tokens": 6958718.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.607457399368286,
|
|
"epoch": 3.1073667711598745,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004329677783352931,
|
|
"loss": 5.0607,
|
|
"mean_token_accuracy": 0.19018178731203078,
|
|
"num_tokens": 6968040.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.6495068073272705,
|
|
"epoch": 3.1112852664576804,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004327534444091436,
|
|
"loss": 5.0641,
|
|
"mean_token_accuracy": 0.1887529894709587,
|
|
"num_tokens": 6976977.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.635713148117065,
|
|
"epoch": 3.115203761755486,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043253882855201037,
|
|
"loss": 5.0398,
|
|
"mean_token_accuracy": 0.19670076072216033,
|
|
"num_tokens": 6985859.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.487248420715332,
|
|
"epoch": 3.1191222570532915,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043232393114744683,
|
|
"loss": 4.9606,
|
|
"mean_token_accuracy": 0.19737394154071808,
|
|
"num_tokens": 6994031.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.489372205734253,
|
|
"epoch": 3.123040752351097,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004321087525795095,
|
|
"loss": 5.0315,
|
|
"mean_token_accuracy": 0.18793469965457915,
|
|
"num_tokens": 7002527.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.627933073043823,
|
|
"epoch": 3.126959247648903,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004318932932327573,
|
|
"loss": 5.1278,
|
|
"mean_token_accuracy": 0.1932612508535385,
|
|
"num_tokens": 7011983.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.650972175598144,
|
|
"epoch": 3.1308777429467085,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000431677553492251,
|
|
"loss": 5.0848,
|
|
"mean_token_accuracy": 0.18824641704559325,
|
|
"num_tokens": 7020819.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.535352087020874,
|
|
"epoch": 3.134796238244514,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043146153374355256,
|
|
"loss": 4.9547,
|
|
"mean_token_accuracy": 0.19972720742225647,
|
|
"num_tokens": 7029257.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.134796238244514,
|
|
"eval_entropy": 5.37750339323236,
|
|
"eval_loss": 5.750268936157227,
|
|
"eval_mean_token_accuracy": 0.16246519099863233,
|
|
"eval_num_tokens": 7029257.0,
|
|
"eval_runtime": 2.1636,
|
|
"eval_samples_per_second": 1905.119,
|
|
"eval_steps_per_second": 238.486,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 12750,
|
|
"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": 1404780923105280.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|