21266 lines
582 KiB
JSON
21266 lines
582 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 8.22884012539185,
|
|
"eval_steps": 500,
|
|
"global_step": 10500,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.585890579223633,
|
|
"epoch": 3.1387147335423196,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043124523437272427,
|
|
"loss": 5.1397,
|
|
"mean_token_accuracy": 0.1832073912024498,
|
|
"num_tokens": 7038942.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.6329512119293215,
|
|
"epoch": 3.1426332288401255,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004310286557663282,
|
|
"loss": 5.1038,
|
|
"mean_token_accuracy": 0.18983772099018098,
|
|
"num_tokens": 7048074.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.620622062683106,
|
|
"epoch": 3.146551724137931,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004308117983114254,
|
|
"loss": 5.0217,
|
|
"mean_token_accuracy": 0.1914113163948059,
|
|
"num_tokens": 7056887.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.60757007598877,
|
|
"epoch": 3.1504702194357366,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004305946623955754,
|
|
"loss": 4.997,
|
|
"mean_token_accuracy": 0.20222129076719284,
|
|
"num_tokens": 7065462.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.604562282562256,
|
|
"epoch": 3.1543887147335425,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043037724840683516,
|
|
"loss": 5.1107,
|
|
"mean_token_accuracy": 0.18965222388505937,
|
|
"num_tokens": 7074188.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.612287330627441,
|
|
"epoch": 3.158307210031348,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043015955673375876,
|
|
"loss": 5.0605,
|
|
"mean_token_accuracy": 0.19196701645851136,
|
|
"num_tokens": 7082282.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.65072546005249,
|
|
"epoch": 3.1622257053291536,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004299415877653966,
|
|
"loss": 5.1192,
|
|
"mean_token_accuracy": 0.18548710495233536,
|
|
"num_tokens": 7090631.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.6337402820587155,
|
|
"epoch": 3.166144200626959,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004297233418912945,
|
|
"loss": 5.0662,
|
|
"mean_token_accuracy": 0.1869766891002655,
|
|
"num_tokens": 7099546.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.601831007003784,
|
|
"epoch": 3.170062695924765,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004295048195014932,
|
|
"loss": 5.1167,
|
|
"mean_token_accuracy": 0.18336354047060013,
|
|
"num_tokens": 7109423.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.517417573928833,
|
|
"epoch": 3.1739811912225706,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004292860209865277,
|
|
"loss": 4.9744,
|
|
"mean_token_accuracy": 0.19031995832920073,
|
|
"num_tokens": 7117717.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.592552614212036,
|
|
"epoch": 3.177899686520376,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004290669467374263,
|
|
"loss": 5.1914,
|
|
"mean_token_accuracy": 0.17589648067951202,
|
|
"num_tokens": 7126573.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.587223482131958,
|
|
"epoch": 3.1818181818181817,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042884759714571037,
|
|
"loss": 5.0823,
|
|
"mean_token_accuracy": 0.19103572964668275,
|
|
"num_tokens": 7134471.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.5124256134033205,
|
|
"epoch": 3.1857366771159876,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004286279726033932,
|
|
"loss": 5.0167,
|
|
"mean_token_accuracy": 0.19329838156700135,
|
|
"num_tokens": 7143092.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.6850724697113035,
|
|
"epoch": 3.189655172413793,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042840807350297933,
|
|
"loss": 5.0999,
|
|
"mean_token_accuracy": 0.19142334908246994,
|
|
"num_tokens": 7152137.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.617215490341186,
|
|
"epoch": 3.1935736677115987,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042818790023746407,
|
|
"loss": 5.0552,
|
|
"mean_token_accuracy": 0.19031702876091003,
|
|
"num_tokens": 7160037.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.509835529327392,
|
|
"epoch": 3.197492163009404,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042796745320033296,
|
|
"loss": 5.047,
|
|
"mean_token_accuracy": 0.18600157052278518,
|
|
"num_tokens": 7168354.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.595581340789795,
|
|
"epoch": 3.20141065830721,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042774673278556043,
|
|
"loss": 5.1214,
|
|
"mean_token_accuracy": 0.18590071201324462,
|
|
"num_tokens": 7176757.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.62891411781311,
|
|
"epoch": 3.2053291536050157,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004275257393876097,
|
|
"loss": 5.0898,
|
|
"mean_token_accuracy": 0.1872025728225708,
|
|
"num_tokens": 7185489.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.617946767807007,
|
|
"epoch": 3.209247648902821,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004273044734014318,
|
|
"loss": 5.0372,
|
|
"mean_token_accuracy": 0.19098826348781586,
|
|
"num_tokens": 7194431.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.512810707092285,
|
|
"epoch": 3.2131661442006267,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042708293522246486,
|
|
"loss": 5.0125,
|
|
"mean_token_accuracy": 0.19310329854488373,
|
|
"num_tokens": 7203074.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.575091600418091,
|
|
"epoch": 3.2170846394984327,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004268611252466337,
|
|
"loss": 5.0469,
|
|
"mean_token_accuracy": 0.19326422810554506,
|
|
"num_tokens": 7211692.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.525807571411133,
|
|
"epoch": 3.2210031347962382,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004266390438703486,
|
|
"loss": 5.0446,
|
|
"mean_token_accuracy": 0.1934028834104538,
|
|
"num_tokens": 7220285.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.6897385597229,
|
|
"epoch": 3.2249216300940438,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042641669149050493,
|
|
"loss": 5.1999,
|
|
"mean_token_accuracy": 0.17942985594272615,
|
|
"num_tokens": 7230007.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.594956731796264,
|
|
"epoch": 3.2288401253918497,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004261940685044825,
|
|
"loss": 5.0403,
|
|
"mean_token_accuracy": 0.19223703891038896,
|
|
"num_tokens": 7238520.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.531583309173584,
|
|
"epoch": 3.2327586206896552,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042597117531014474,
|
|
"loss": 4.9771,
|
|
"mean_token_accuracy": 0.1953437477350235,
|
|
"num_tokens": 7247228.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.6342696189880375,
|
|
"epoch": 3.2366771159874608,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004257480123058378,
|
|
"loss": 5.2116,
|
|
"mean_token_accuracy": 0.18093717396259307,
|
|
"num_tokens": 7255782.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.690195465087891,
|
|
"epoch": 3.2405956112852663,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042552457989039036,
|
|
"loss": 5.1068,
|
|
"mean_token_accuracy": 0.18073305934667588,
|
|
"num_tokens": 7264946.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.492582893371582,
|
|
"epoch": 3.2445141065830723,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042530087846311213,
|
|
"loss": 4.9286,
|
|
"mean_token_accuracy": 0.20112788528203965,
|
|
"num_tokens": 7273647.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.59495997428894,
|
|
"epoch": 3.248432601880878,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042507690842379396,
|
|
"loss": 5.0643,
|
|
"mean_token_accuracy": 0.18867938071489335,
|
|
"num_tokens": 7283393.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.656988859176636,
|
|
"epoch": 3.2523510971786833,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 5.1515,
|
|
"mean_token_accuracy": 0.19005416333675385,
|
|
"num_tokens": 7292368.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.502977132797241,
|
|
"epoch": 3.256269592476489,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042462816411060025,
|
|
"loss": 5.1008,
|
|
"mean_token_accuracy": 0.18774392306804658,
|
|
"num_tokens": 7301505.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.56800651550293,
|
|
"epoch": 3.260188087774295,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004244033906387035,
|
|
"loss": 5.0586,
|
|
"mean_token_accuracy": 0.19629054218530656,
|
|
"num_tokens": 7310011.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.643670129776001,
|
|
"epoch": 3.2641065830721003,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004241783501587231,
|
|
"loss": 5.0914,
|
|
"mean_token_accuracy": 0.19367300868034362,
|
|
"num_tokens": 7319295.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.591206312179565,
|
|
"epoch": 3.268025078369906,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042395304307284284,
|
|
"loss": 5.1228,
|
|
"mean_token_accuracy": 0.18766565918922423,
|
|
"num_tokens": 7328390.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.647037410736084,
|
|
"epoch": 3.271943573667712,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004237274697837229,
|
|
"loss": 5.1366,
|
|
"mean_token_accuracy": 0.18287569880485535,
|
|
"num_tokens": 7337062.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.568262958526612,
|
|
"epoch": 3.2758620689655173,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004235016306944996,
|
|
"loss": 5.1389,
|
|
"mean_token_accuracy": 0.18775300532579423,
|
|
"num_tokens": 7346364.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.595852327346802,
|
|
"epoch": 3.279780564263323,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004232755262087837,
|
|
"loss": 5.0588,
|
|
"mean_token_accuracy": 0.18506316393613814,
|
|
"num_tokens": 7354896.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.574965143203736,
|
|
"epoch": 3.2836990595611284,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042304915673066083,
|
|
"loss": 5.0979,
|
|
"mean_token_accuracy": 0.19143613576889038,
|
|
"num_tokens": 7363591.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.58932204246521,
|
|
"epoch": 3.287617554858934,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042282252266468985,
|
|
"loss": 5.1112,
|
|
"mean_token_accuracy": 0.18620892167091369,
|
|
"num_tokens": 7372891.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.606708955764771,
|
|
"epoch": 3.29153605015674,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004225956244159025,
|
|
"loss": 5.039,
|
|
"mean_token_accuracy": 0.18780850768089294,
|
|
"num_tokens": 7381945.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.649549102783203,
|
|
"epoch": 3.2954545454545454,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004223684623898029,
|
|
"loss": 5.1844,
|
|
"mean_token_accuracy": 0.18751512169837953,
|
|
"num_tokens": 7391449.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.586066865921021,
|
|
"epoch": 3.299373040752351,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004221410369923662,
|
|
"loss": 5.0433,
|
|
"mean_token_accuracy": 0.19521359503269195,
|
|
"num_tokens": 7399762.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.465223598480224,
|
|
"epoch": 3.303291536050157,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042191334863003873,
|
|
"loss": 4.9447,
|
|
"mean_token_accuracy": 0.1937822088599205,
|
|
"num_tokens": 7408711.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.497321939468383,
|
|
"epoch": 3.3072100313479624,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004216853977097363,
|
|
"loss": 5.036,
|
|
"mean_token_accuracy": 0.19624617397785188,
|
|
"num_tokens": 7416868.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.593992567062378,
|
|
"epoch": 3.311128526645768,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004214571846388442,
|
|
"loss": 5.1351,
|
|
"mean_token_accuracy": 0.1839517429471016,
|
|
"num_tokens": 7425687.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.60465145111084,
|
|
"epoch": 3.3150470219435735,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004212287098252164,
|
|
"loss": 5.037,
|
|
"mean_token_accuracy": 0.19320485591888428,
|
|
"num_tokens": 7433714.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.543931674957276,
|
|
"epoch": 3.3189655172413794,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004209999736771742,
|
|
"loss": 5.1098,
|
|
"mean_token_accuracy": 0.19302091151475906,
|
|
"num_tokens": 7442765.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.572278594970703,
|
|
"epoch": 3.322884012539185,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004207709766035063,
|
|
"loss": 5.0787,
|
|
"mean_token_accuracy": 0.18920503109693526,
|
|
"num_tokens": 7450672.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.625389623641968,
|
|
"epoch": 3.3268025078369905,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004205417190134674,
|
|
"loss": 5.1386,
|
|
"mean_token_accuracy": 0.18321106284856797,
|
|
"num_tokens": 7458758.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.609045791625976,
|
|
"epoch": 3.330721003134796,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004203122013167783,
|
|
"loss": 5.0943,
|
|
"mean_token_accuracy": 0.18476527333259582,
|
|
"num_tokens": 7467527.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.600168943405151,
|
|
"epoch": 3.334639498432602,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042008242392362413,
|
|
"loss": 5.1019,
|
|
"mean_token_accuracy": 0.18938433676958083,
|
|
"num_tokens": 7476659.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.544204044342041,
|
|
"epoch": 3.3385579937304075,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041985238724465433,
|
|
"loss": 5.0192,
|
|
"mean_token_accuracy": 0.1968235895037651,
|
|
"num_tokens": 7485813.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.608414125442505,
|
|
"epoch": 3.342476489028213,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041962209169098193,
|
|
"loss": 5.1218,
|
|
"mean_token_accuracy": 0.18301907777786255,
|
|
"num_tokens": 7494842.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.618138551712036,
|
|
"epoch": 3.346394984326019,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004193915376741823,
|
|
"loss": 5.1412,
|
|
"mean_token_accuracy": 0.18555521219968796,
|
|
"num_tokens": 7503633.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.5488921165466305,
|
|
"epoch": 3.3503134796238245,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004191607256062928,
|
|
"loss": 5.0642,
|
|
"mean_token_accuracy": 0.18925698697566987,
|
|
"num_tokens": 7512759.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.599124765396118,
|
|
"epoch": 3.35423197492163,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004189296558998121,
|
|
"loss": 5.1592,
|
|
"mean_token_accuracy": 0.18589088171720505,
|
|
"num_tokens": 7522284.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.620037937164307,
|
|
"epoch": 3.3581504702194356,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004186983289676992,
|
|
"loss": 5.0736,
|
|
"mean_token_accuracy": 0.1908479243516922,
|
|
"num_tokens": 7531286.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.584924697875977,
|
|
"epoch": 3.3620689655172415,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041846674522337296,
|
|
"loss": 5.0743,
|
|
"mean_token_accuracy": 0.18472516983747483,
|
|
"num_tokens": 7540450.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.51798300743103,
|
|
"epoch": 3.365987460815047,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041823490508071076,
|
|
"loss": 4.9865,
|
|
"mean_token_accuracy": 0.19123904407024384,
|
|
"num_tokens": 7549386.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.6133218765258786,
|
|
"epoch": 3.3699059561128526,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000418002808954049,
|
|
"loss": 5.0974,
|
|
"mean_token_accuracy": 0.1870042935013771,
|
|
"num_tokens": 7558236.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.514813852310181,
|
|
"epoch": 3.373824451410658,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041777045725818057,
|
|
"loss": 4.9672,
|
|
"mean_token_accuracy": 0.198385988175869,
|
|
"num_tokens": 7566984.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.524264907836914,
|
|
"epoch": 3.377742946708464,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000417537850408356,
|
|
"loss": 5.0976,
|
|
"mean_token_accuracy": 0.18298899084329606,
|
|
"num_tokens": 7576018.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.577986526489258,
|
|
"epoch": 3.3816614420062696,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004173049888202814,
|
|
"loss": 5.0815,
|
|
"mean_token_accuracy": 0.18884333372116088,
|
|
"num_tokens": 7585294.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.555334663391113,
|
|
"epoch": 3.385579937304075,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004170718729101179,
|
|
"loss": 5.0578,
|
|
"mean_token_accuracy": 0.19106940180063248,
|
|
"num_tokens": 7594634.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.517987823486328,
|
|
"epoch": 3.389498432601881,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041683850309448187,
|
|
"loss": 5.1007,
|
|
"mean_token_accuracy": 0.18376322835683823,
|
|
"num_tokens": 7603153.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.601560115814209,
|
|
"epoch": 3.3934169278996866,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041660487979044264,
|
|
"loss": 5.067,
|
|
"mean_token_accuracy": 0.1918576881289482,
|
|
"num_tokens": 7612027.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.587055587768555,
|
|
"epoch": 3.397335423197492,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004163710034155231,
|
|
"loss": 5.051,
|
|
"mean_token_accuracy": 0.18780431300401687,
|
|
"num_tokens": 7620725.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.509176588058471,
|
|
"epoch": 3.4012539184952977,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004161368743876982,
|
|
"loss": 5.0336,
|
|
"mean_token_accuracy": 0.1925852969288826,
|
|
"num_tokens": 7628847.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.586984395980835,
|
|
"epoch": 3.405172413793103,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004159024931253945,
|
|
"loss": 5.1265,
|
|
"mean_token_accuracy": 0.18697059601545335,
|
|
"num_tokens": 7637668.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.634477186203003,
|
|
"epoch": 3.409090909090909,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041566786004748943,
|
|
"loss": 5.1532,
|
|
"mean_token_accuracy": 0.18601921051740647,
|
|
"num_tokens": 7646545.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.657234191894531,
|
|
"epoch": 3.4130094043887147,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041543297557331015,
|
|
"loss": 5.154,
|
|
"mean_token_accuracy": 0.17961082905530928,
|
|
"num_tokens": 7654967.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.547342109680176,
|
|
"epoch": 3.41692789968652,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004151978401226335,
|
|
"loss": 5.0607,
|
|
"mean_token_accuracy": 0.1932755321264267,
|
|
"num_tokens": 7663718.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.58386607170105,
|
|
"epoch": 3.420846394984326,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041496245411568435,
|
|
"loss": 5.1043,
|
|
"mean_token_accuracy": 0.1900026321411133,
|
|
"num_tokens": 7672430.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.564857912063599,
|
|
"epoch": 3.4247648902821317,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004147268179731359,
|
|
"loss": 5.0836,
|
|
"mean_token_accuracy": 0.1878344565629959,
|
|
"num_tokens": 7681270.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.602977085113525,
|
|
"epoch": 3.4286833855799372,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041449093211610815,
|
|
"loss": 5.0793,
|
|
"mean_token_accuracy": 0.18902707993984222,
|
|
"num_tokens": 7690124.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.620127964019775,
|
|
"epoch": 3.4326018808777428,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041425479696616734,
|
|
"loss": 5.1132,
|
|
"mean_token_accuracy": 0.18534173369407653,
|
|
"num_tokens": 7699511.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.5457008361816404,
|
|
"epoch": 3.4365203761755487,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 5.0778,
|
|
"mean_token_accuracy": 0.1890060618519783,
|
|
"num_tokens": 7708796.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.616153860092163,
|
|
"epoch": 3.4404388714733543,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00041378178047603845,
|
|
"loss": 5.0653,
|
|
"mean_token_accuracy": 0.19537944197654725,
|
|
"num_tokens": 7717446.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.5605024814605715,
|
|
"epoch": 3.44435736677116,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004135448999812074,
|
|
"loss": 5.1181,
|
|
"mean_token_accuracy": 0.18161804974079132,
|
|
"num_tokens": 7726814.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.562195539474487,
|
|
"epoch": 3.4482758620689653,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004133077718841763,
|
|
"loss": 5.0307,
|
|
"mean_token_accuracy": 0.19746554344892503,
|
|
"num_tokens": 7735895.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.540375423431397,
|
|
"epoch": 3.4521943573667713,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041307039660873113,
|
|
"loss": 5.1018,
|
|
"mean_token_accuracy": 0.19003206342458726,
|
|
"num_tokens": 7744160.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.568145847320556,
|
|
"epoch": 3.456112852664577,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004128327745791002,
|
|
"loss": 5.1675,
|
|
"mean_token_accuracy": 0.18221710473299027,
|
|
"num_tokens": 7752844.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.590904808044433,
|
|
"epoch": 3.4600313479623823,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004125949062199526,
|
|
"loss": 5.1237,
|
|
"mean_token_accuracy": 0.1823289915919304,
|
|
"num_tokens": 7762327.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.550951242446899,
|
|
"epoch": 3.4639498432601883,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041235679195639766,
|
|
"loss": 5.0514,
|
|
"mean_token_accuracy": 0.19206651598215102,
|
|
"num_tokens": 7770813.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.508137178421021,
|
|
"epoch": 3.467868338557994,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041211843221398406,
|
|
"loss": 5.0753,
|
|
"mean_token_accuracy": 0.18570431619882583,
|
|
"num_tokens": 7779357.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.591682004928589,
|
|
"epoch": 3.4717868338557993,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004118798274186994,
|
|
"loss": 5.1153,
|
|
"mean_token_accuracy": 0.18358745872974397,
|
|
"num_tokens": 7787302.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.559949445724487,
|
|
"epoch": 3.475705329153605,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004116409779969691,
|
|
"loss": 5.1076,
|
|
"mean_token_accuracy": 0.18605598509311677,
|
|
"num_tokens": 7795651.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.545140171051026,
|
|
"epoch": 3.479623824451411,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00041140188437565586,
|
|
"loss": 5.0858,
|
|
"mean_token_accuracy": 0.18824351876974105,
|
|
"num_tokens": 7804644.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.654828882217407,
|
|
"epoch": 3.4835423197492164,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041116254698205873,
|
|
"loss": 5.0797,
|
|
"mean_token_accuracy": 0.1937394082546234,
|
|
"num_tokens": 7812479.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.527298498153686,
|
|
"epoch": 3.487460815047022,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041092296624391244,
|
|
"loss": 5.0698,
|
|
"mean_token_accuracy": 0.1886146292090416,
|
|
"num_tokens": 7820678.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.547030210494995,
|
|
"epoch": 3.4913793103448274,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004106831425893865,
|
|
"loss": 5.1045,
|
|
"mean_token_accuracy": 0.19473838657140732,
|
|
"num_tokens": 7829063.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.611020088195801,
|
|
"epoch": 3.4952978056426334,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004104430764470849,
|
|
"loss": 5.1299,
|
|
"mean_token_accuracy": 0.19241247028112413,
|
|
"num_tokens": 7837274.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.582591915130616,
|
|
"epoch": 3.499216300940439,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004102027682460445,
|
|
"loss": 5.0472,
|
|
"mean_token_accuracy": 0.19492462575435637,
|
|
"num_tokens": 7845379.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.588556432723999,
|
|
"epoch": 3.5031347962382444,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004099622184157353,
|
|
"loss": 5.0791,
|
|
"mean_token_accuracy": 0.1893304854631424,
|
|
"num_tokens": 7853760.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.66967453956604,
|
|
"epoch": 3.5070532915360504,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004097214273860586,
|
|
"loss": 5.1521,
|
|
"mean_token_accuracy": 0.18174111247062683,
|
|
"num_tokens": 7862026.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.538205146789551,
|
|
"epoch": 3.510971786833856,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004094803955873471,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.19349520951509475,
|
|
"num_tokens": 7870369.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.5720775604248045,
|
|
"epoch": 3.5148902821316614,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004092391234503638,
|
|
"loss": 5.0965,
|
|
"mean_token_accuracy": 0.19048729985952378,
|
|
"num_tokens": 7878524.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.628887510299682,
|
|
"epoch": 3.518808777429467,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040899761140630094,
|
|
"loss": 5.1588,
|
|
"mean_token_accuracy": 0.17792422771453859,
|
|
"num_tokens": 7887369.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.640236949920654,
|
|
"epoch": 3.5227272727272725,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040875585988677985,
|
|
"loss": 5.1302,
|
|
"mean_token_accuracy": 0.1799825683236122,
|
|
"num_tokens": 7896112.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.52794451713562,
|
|
"epoch": 3.5266457680250785,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004085138693238497,
|
|
"loss": 4.9752,
|
|
"mean_token_accuracy": 0.19562461525201796,
|
|
"num_tokens": 7904580.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.5266457680250785,
|
|
"eval_entropy": 5.370933149212091,
|
|
"eval_loss": 5.715298652648926,
|
|
"eval_mean_token_accuracy": 0.16496776247238235,
|
|
"eval_num_tokens": 7904580.0,
|
|
"eval_runtime": 2.16,
|
|
"eval_samples_per_second": 1908.293,
|
|
"eval_steps_per_second": 238.884,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.6043954372406,
|
|
"epoch": 3.530564263322884,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004082716401499867,
|
|
"loss": 5.1243,
|
|
"mean_token_accuracy": 0.18978817909955978,
|
|
"num_tokens": 7913755.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.566169786453247,
|
|
"epoch": 3.5344827586206895,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040802917279809383,
|
|
"loss": 5.1007,
|
|
"mean_token_accuracy": 0.19534085541963578,
|
|
"num_tokens": 7922385.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.551551151275635,
|
|
"epoch": 3.5384012539184955,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040778646770149953,
|
|
"loss": 5.1368,
|
|
"mean_token_accuracy": 0.1839648514986038,
|
|
"num_tokens": 7931634.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.540170812606812,
|
|
"epoch": 3.542319749216301,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040754352529395716,
|
|
"loss": 5.0316,
|
|
"mean_token_accuracy": 0.18457334488630295,
|
|
"num_tokens": 7939866.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.557271432876587,
|
|
"epoch": 3.5462382445141065,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040730034600964415,
|
|
"loss": 5.1447,
|
|
"mean_token_accuracy": 0.17746772170066832,
|
|
"num_tokens": 7948961.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.4521852970123295,
|
|
"epoch": 3.5501567398119125,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004070569302831613,
|
|
"loss": 4.9911,
|
|
"mean_token_accuracy": 0.18950635939836502,
|
|
"num_tokens": 7958526.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.561177206039429,
|
|
"epoch": 3.554075235109718,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000406813278549532,
|
|
"loss": 5.0983,
|
|
"mean_token_accuracy": 0.18968672007322313,
|
|
"num_tokens": 7966887.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.579365491867065,
|
|
"epoch": 3.5579937304075235,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00040656939124420144,
|
|
"loss": 5.0891,
|
|
"mean_token_accuracy": 0.18481013178825378,
|
|
"num_tokens": 7974929.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.5307363986969,
|
|
"epoch": 3.561912225705329,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004063252688030358,
|
|
"loss": 5.0726,
|
|
"mean_token_accuracy": 0.188212288916111,
|
|
"num_tokens": 7983821.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.5377460479736325,
|
|
"epoch": 3.5658307210031346,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004060809116623213,
|
|
"loss": 5.0637,
|
|
"mean_token_accuracy": 0.18631937652826308,
|
|
"num_tokens": 7992109.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.5388243198394775,
|
|
"epoch": 3.5697492163009406,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000405836320258764,
|
|
"loss": 5.0775,
|
|
"mean_token_accuracy": 0.1965628981590271,
|
|
"num_tokens": 8001098.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.650246000289917,
|
|
"epoch": 3.573667711598746,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004055914950294882,
|
|
"loss": 5.0865,
|
|
"mean_token_accuracy": 0.1838941603899002,
|
|
"num_tokens": 8009559.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.517079544067383,
|
|
"epoch": 3.5775862068965516,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040534643641203645,
|
|
"loss": 5.0175,
|
|
"mean_token_accuracy": 0.1918938398361206,
|
|
"num_tokens": 8018021.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.487963771820068,
|
|
"epoch": 3.5815047021943576,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004051011448443681,
|
|
"loss": 5.0348,
|
|
"mean_token_accuracy": 0.19248413741588594,
|
|
"num_tokens": 8026258.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.591257572174072,
|
|
"epoch": 3.585423197492163,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000404855620764859,
|
|
"loss": 5.0856,
|
|
"mean_token_accuracy": 0.1899063155055046,
|
|
"num_tokens": 8036761.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.60556116104126,
|
|
"epoch": 3.5893416927899686,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004046098646123006,
|
|
"loss": 5.1286,
|
|
"mean_token_accuracy": 0.19197009801864623,
|
|
"num_tokens": 8045115.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.637094402313233,
|
|
"epoch": 3.593260188087774,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040436387682589876,
|
|
"loss": 5.1434,
|
|
"mean_token_accuracy": 0.18741354942321778,
|
|
"num_tokens": 8054334.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.53146014213562,
|
|
"epoch": 3.5971786833855797,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004041176578452737,
|
|
"loss": 5.0863,
|
|
"mean_token_accuracy": 0.18572427630424498,
|
|
"num_tokens": 8063165.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.536058521270752,
|
|
"epoch": 3.6010971786833856,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004038712081104587,
|
|
"loss": 5.0711,
|
|
"mean_token_accuracy": 0.18742595911026,
|
|
"num_tokens": 8072498.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.565552473068237,
|
|
"epoch": 3.605015673981191,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040362452806189927,
|
|
"loss": 5.0894,
|
|
"mean_token_accuracy": 0.18691499829292296,
|
|
"num_tokens": 8081455.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.521443653106689,
|
|
"epoch": 3.6089341692789967,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004033776181404527,
|
|
"loss": 5.1027,
|
|
"mean_token_accuracy": 0.1844627946615219,
|
|
"num_tokens": 8090992.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.587378358840942,
|
|
"epoch": 3.6128526645768027,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040313047878738704,
|
|
"loss": 5.0832,
|
|
"mean_token_accuracy": 0.19045060276985168,
|
|
"num_tokens": 8099976.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.551462268829345,
|
|
"epoch": 3.616771159874608,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004028831104443805,
|
|
"loss": 5.004,
|
|
"mean_token_accuracy": 0.2044840395450592,
|
|
"num_tokens": 8108275.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.549649858474732,
|
|
"epoch": 3.6206896551724137,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 5.1057,
|
|
"mean_token_accuracy": 0.1859031140804291,
|
|
"num_tokens": 8117441.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.509921979904175,
|
|
"epoch": 3.6246081504702197,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00040238768855730214,
|
|
"loss": 5.0691,
|
|
"mean_token_accuracy": 0.18609122633934022,
|
|
"num_tokens": 8125709.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.521451187133789,
|
|
"epoch": 3.628526645768025,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040213963589862963,
|
|
"loss": 5.0777,
|
|
"mean_token_accuracy": 0.18876318633556366,
|
|
"num_tokens": 8134418.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.589888381958008,
|
|
"epoch": 3.6324451410658307,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004018913560208131,
|
|
"loss": 5.1717,
|
|
"mean_token_accuracy": 0.18029381930828095,
|
|
"num_tokens": 8143729.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.528113269805909,
|
|
"epoch": 3.6363636363636362,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004016428493675689,
|
|
"loss": 5.0313,
|
|
"mean_token_accuracy": 0.18823620826005935,
|
|
"num_tokens": 8152613.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.493265676498413,
|
|
"epoch": 3.6402821316614418,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004013941163830187,
|
|
"loss": 5.0134,
|
|
"mean_token_accuracy": 0.1936560496687889,
|
|
"num_tokens": 8160862.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.544780683517456,
|
|
"epoch": 3.6442006269592477,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004011451575116887,
|
|
"loss": 5.1026,
|
|
"mean_token_accuracy": 0.18627175986766814,
|
|
"num_tokens": 8169369.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.522828102111816,
|
|
"epoch": 3.6481191222570533,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004008959731985087,
|
|
"loss": 5.0164,
|
|
"mean_token_accuracy": 0.1896409049630165,
|
|
"num_tokens": 8178152.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.471308088302612,
|
|
"epoch": 3.652037617554859,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040064656388881157,
|
|
"loss": 5.005,
|
|
"mean_token_accuracy": 0.1957460805773735,
|
|
"num_tokens": 8186690.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.480044031143189,
|
|
"epoch": 3.6559561128526648,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004003969300283321,
|
|
"loss": 4.9793,
|
|
"mean_token_accuracy": 0.1986520454287529,
|
|
"num_tokens": 8194976.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.539482736587525,
|
|
"epoch": 3.6598746081504703,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040014707206320653,
|
|
"loss": 5.0781,
|
|
"mean_token_accuracy": 0.1896795004606247,
|
|
"num_tokens": 8202964.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.518435144424439,
|
|
"epoch": 3.663793103448276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039989699043997153,
|
|
"loss": 5.0846,
|
|
"mean_token_accuracy": 0.1837253078818321,
|
|
"num_tokens": 8213084.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.529673624038696,
|
|
"epoch": 3.6677115987460818,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039964668560556356,
|
|
"loss": 5.0068,
|
|
"mean_token_accuracy": 0.19858694523572923,
|
|
"num_tokens": 8221645.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.526202249526977,
|
|
"epoch": 3.6716300940438873,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039939615800731784,
|
|
"loss": 4.9931,
|
|
"mean_token_accuracy": 0.19656671285629274,
|
|
"num_tokens": 8230195.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.505875110626221,
|
|
"epoch": 3.675548589341693,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039914540809296795,
|
|
"loss": 5.0942,
|
|
"mean_token_accuracy": 0.19095501601696013,
|
|
"num_tokens": 8239608.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.60519323348999,
|
|
"epoch": 3.6794670846394983,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003988944363106445,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.19055551737546922,
|
|
"num_tokens": 8247993.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.541627311706543,
|
|
"epoch": 3.683385579937304,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003986432431088749,
|
|
"loss": 5.078,
|
|
"mean_token_accuracy": 0.18809449523687363,
|
|
"num_tokens": 8257279.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.587546634674072,
|
|
"epoch": 3.68730407523511,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000398391828936582,
|
|
"loss": 5.1352,
|
|
"mean_token_accuracy": 0.1858746290206909,
|
|
"num_tokens": 8265166.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.479617357254028,
|
|
"epoch": 3.6912225705329154,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003981401942430838,
|
|
"loss": 4.9385,
|
|
"mean_token_accuracy": 0.19583314806222915,
|
|
"num_tokens": 8273281.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.526025867462158,
|
|
"epoch": 3.695141065830721,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039788833947809217,
|
|
"loss": 5.1265,
|
|
"mean_token_accuracy": 0.1833677113056183,
|
|
"num_tokens": 8282432.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.463887739181518,
|
|
"epoch": 3.699059561128527,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003976362650917125,
|
|
"loss": 4.9491,
|
|
"mean_token_accuracy": 0.19654021114110948,
|
|
"num_tokens": 8290909.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.5147521018981935,
|
|
"epoch": 3.7029780564263324,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039738397153444264,
|
|
"loss": 5.079,
|
|
"mean_token_accuracy": 0.1936231315135956,
|
|
"num_tokens": 8299581.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.557221269607544,
|
|
"epoch": 3.706896551724138,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003971314592571719,
|
|
"loss": 5.0711,
|
|
"mean_token_accuracy": 0.18683652579784393,
|
|
"num_tokens": 8308184.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.537987804412841,
|
|
"epoch": 3.7108150470219434,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003968787287111809,
|
|
"loss": 5.089,
|
|
"mean_token_accuracy": 0.19177159816026687,
|
|
"num_tokens": 8317054.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.5228406429290775,
|
|
"epoch": 3.714733542319749,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039662578034814,
|
|
"loss": 5.0818,
|
|
"mean_token_accuracy": 0.19449300169944764,
|
|
"num_tokens": 8325961.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.6515435695648195,
|
|
"epoch": 3.718652037617555,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039637261462010886,
|
|
"loss": 5.182,
|
|
"mean_token_accuracy": 0.18350995779037477,
|
|
"num_tokens": 8334401.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.6132348537445065,
|
|
"epoch": 3.7225705329153604,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003961192319795358,
|
|
"loss": 5.1452,
|
|
"mean_token_accuracy": 0.1842637836933136,
|
|
"num_tokens": 8343786.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.63161039352417,
|
|
"epoch": 3.726489028213166,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003958656328792565,
|
|
"loss": 5.1102,
|
|
"mean_token_accuracy": 0.18997125029563905,
|
|
"num_tokens": 8352286.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.501084661483764,
|
|
"epoch": 3.730407523510972,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039561181777249396,
|
|
"loss": 5.043,
|
|
"mean_token_accuracy": 0.19152599275112153,
|
|
"num_tokens": 8361274.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.556865310668945,
|
|
"epoch": 3.7343260188087775,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039535778711285676,
|
|
"loss": 5.0877,
|
|
"mean_token_accuracy": 0.1932084947824478,
|
|
"num_tokens": 8369881.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.582962989807129,
|
|
"epoch": 3.738244514106583,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003951035413543388,
|
|
"loss": 5.1656,
|
|
"mean_token_accuracy": 0.1735078513622284,
|
|
"num_tokens": 8378234.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.51520357131958,
|
|
"epoch": 3.742163009404389,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039484908095131874,
|
|
"loss": 5.0725,
|
|
"mean_token_accuracy": 0.19115858376026154,
|
|
"num_tokens": 8387023.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.51248197555542,
|
|
"epoch": 3.7460815047021945,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003945944063585582,
|
|
"loss": 5.0285,
|
|
"mean_token_accuracy": 0.19660211801528932,
|
|
"num_tokens": 8396110.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.51550703048706,
|
|
"epoch": 3.75,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039433951803120225,
|
|
"loss": 4.9949,
|
|
"mean_token_accuracy": 0.20208403319120408,
|
|
"num_tokens": 8405219.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.603872156143188,
|
|
"epoch": 3.7539184952978055,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039408441642477764,
|
|
"loss": 5.2038,
|
|
"mean_token_accuracy": 0.18288621604442595,
|
|
"num_tokens": 8413996.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.507548475265503,
|
|
"epoch": 3.757836990595611,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003938291019951922,
|
|
"loss": 4.959,
|
|
"mean_token_accuracy": 0.20249723345041276,
|
|
"num_tokens": 8421931.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.500359582901001,
|
|
"epoch": 3.761755485893417,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003935735751987344,
|
|
"loss": 5.0481,
|
|
"mean_token_accuracy": 0.1878534436225891,
|
|
"num_tokens": 8430090.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.437503480911255,
|
|
"epoch": 3.7656739811912225,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039331783649207175,
|
|
"loss": 5.0331,
|
|
"mean_token_accuracy": 0.1925066128373146,
|
|
"num_tokens": 8439311.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.542224502563476,
|
|
"epoch": 3.769592476489028,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039306188633225097,
|
|
"loss": 5.1043,
|
|
"mean_token_accuracy": 0.18851862400770186,
|
|
"num_tokens": 8448952.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.526122188568115,
|
|
"epoch": 3.773510971786834,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003928057251766965,
|
|
"loss": 5.0267,
|
|
"mean_token_accuracy": 0.1883058115839958,
|
|
"num_tokens": 8457177.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.498041200637817,
|
|
"epoch": 3.7774294670846396,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039254935348320984,
|
|
"loss": 5.0622,
|
|
"mean_token_accuracy": 0.194136643409729,
|
|
"num_tokens": 8465844.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.600159502029419,
|
|
"epoch": 3.781347962382445,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039229277170996885,
|
|
"loss": 5.1995,
|
|
"mean_token_accuracy": 0.17922210991382598,
|
|
"num_tokens": 8475198.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.602192497253418,
|
|
"epoch": 3.785266457680251,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003920359803155266,
|
|
"loss": 5.0423,
|
|
"mean_token_accuracy": 0.19280895739793777,
|
|
"num_tokens": 8484050.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.504422569274903,
|
|
"epoch": 3.7891849529780566,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039177897975881115,
|
|
"loss": 5.0382,
|
|
"mean_token_accuracy": 0.1929103910923004,
|
|
"num_tokens": 8491825.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.506332063674927,
|
|
"epoch": 3.793103448275862,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003915217704991239,
|
|
"loss": 5.0464,
|
|
"mean_token_accuracy": 0.1932547226548195,
|
|
"num_tokens": 8500651.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.518607473373413,
|
|
"epoch": 3.7970219435736676,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003912643529961397,
|
|
"loss": 5.0482,
|
|
"mean_token_accuracy": 0.19266632497310637,
|
|
"num_tokens": 8508464.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.58462963104248,
|
|
"epoch": 3.800940438871473,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003910067277099053,
|
|
"loss": 5.082,
|
|
"mean_token_accuracy": 0.19516697973012925,
|
|
"num_tokens": 8517676.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.631950950622558,
|
|
"epoch": 3.804858934169279,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.1364,
|
|
"mean_token_accuracy": 0.18472751528024672,
|
|
"num_tokens": 8526926.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.47961106300354,
|
|
"epoch": 3.8087774294670846,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003904908556297293,
|
|
"loss": 5.0078,
|
|
"mean_token_accuracy": 0.1895693615078926,
|
|
"num_tokens": 8535814.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.597270488739014,
|
|
"epoch": 3.81269592476489,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003902326097577345,
|
|
"loss": 5.1215,
|
|
"mean_token_accuracy": 0.18935267478227616,
|
|
"num_tokens": 8544814.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.563870573043824,
|
|
"epoch": 3.816614420062696,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038997415794638206,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.18612706512212754,
|
|
"num_tokens": 8553455.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.490188694000244,
|
|
"epoch": 3.8205329153605017,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003897155006575672,
|
|
"loss": 5.0398,
|
|
"mean_token_accuracy": 0.18479942977428437,
|
|
"num_tokens": 8562048.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.557944202423096,
|
|
"epoch": 3.824451410658307,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038945663835355247,
|
|
"loss": 5.0903,
|
|
"mean_token_accuracy": 0.1898506611585617,
|
|
"num_tokens": 8571855.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.560569620132446,
|
|
"epoch": 3.8283699059561127,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038919757149696665,
|
|
"loss": 5.087,
|
|
"mean_token_accuracy": 0.1825867548584938,
|
|
"num_tokens": 8580425.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.5515789031982425,
|
|
"epoch": 3.8322884012539182,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038893830055080437,
|
|
"loss": 5.096,
|
|
"mean_token_accuracy": 0.1907548114657402,
|
|
"num_tokens": 8589751.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.5596333026885985,
|
|
"epoch": 3.836206896551724,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003886788259784248,
|
|
"loss": 5.1124,
|
|
"mean_token_accuracy": 0.1814299166202545,
|
|
"num_tokens": 8598077.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.600435400009156,
|
|
"epoch": 3.8401253918495297,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038841914824355093,
|
|
"loss": 5.1204,
|
|
"mean_token_accuracy": 0.18435739129781722,
|
|
"num_tokens": 8606691.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.512621736526489,
|
|
"epoch": 3.8440438871473352,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038815926781026914,
|
|
"loss": 5.0652,
|
|
"mean_token_accuracy": 0.1924016445875168,
|
|
"num_tokens": 8615477.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.507816314697266,
|
|
"epoch": 3.847962382445141,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003878991851430279,
|
|
"loss": 5.0804,
|
|
"mean_token_accuracy": 0.19019101858139037,
|
|
"num_tokens": 8623982.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.5093669414520265,
|
|
"epoch": 3.8518808777429467,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003876389007066371,
|
|
"loss": 5.1098,
|
|
"mean_token_accuracy": 0.1928531214594841,
|
|
"num_tokens": 8633017.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.525287246704101,
|
|
"epoch": 3.8557993730407523,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003873784149662672,
|
|
"loss": 5.06,
|
|
"mean_token_accuracy": 0.19075425565242768,
|
|
"num_tokens": 8642299.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.508560276031494,
|
|
"epoch": 3.8597178683385582,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003871177283874484,
|
|
"loss": 4.9898,
|
|
"mean_token_accuracy": 0.1908753827214241,
|
|
"num_tokens": 8651008.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.506883668899536,
|
|
"epoch": 3.8636363636363638,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038685684143606995,
|
|
"loss": 5.0756,
|
|
"mean_token_accuracy": 0.19238980263471603,
|
|
"num_tokens": 8660103.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.433812475204467,
|
|
"epoch": 3.8675548589341693,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003865957545783791,
|
|
"loss": 5.032,
|
|
"mean_token_accuracy": 0.19585823714733125,
|
|
"num_tokens": 8668646.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.531681108474731,
|
|
"epoch": 3.871473354231975,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038633446828098046,
|
|
"loss": 5.1706,
|
|
"mean_token_accuracy": 0.18864066451787947,
|
|
"num_tokens": 8678190.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.667750835418701,
|
|
"epoch": 3.8753918495297803,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000386072983010835,
|
|
"loss": 5.1749,
|
|
"mean_token_accuracy": 0.1785569593310356,
|
|
"num_tokens": 8687336.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.562904119491577,
|
|
"epoch": 3.8793103448275863,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038581129923525914,
|
|
"loss": 5.0756,
|
|
"mean_token_accuracy": 0.18508994579315186,
|
|
"num_tokens": 8695939.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.516730833053589,
|
|
"epoch": 3.883228840125392,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038554941742192445,
|
|
"loss": 5.1387,
|
|
"mean_token_accuracy": 0.18860141634941102,
|
|
"num_tokens": 8704828.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.575774192810059,
|
|
"epoch": 3.8871473354231973,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003852873380388561,
|
|
"loss": 5.0532,
|
|
"mean_token_accuracy": 0.19733287543058395,
|
|
"num_tokens": 8714427.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.585229969024658,
|
|
"epoch": 3.8910658307210033,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003850250615544323,
|
|
"loss": 5.1496,
|
|
"mean_token_accuracy": 0.19043119251728058,
|
|
"num_tokens": 8723737.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.467834329605102,
|
|
"epoch": 3.894984326018809,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038476258843738386,
|
|
"loss": 5.0935,
|
|
"mean_token_accuracy": 0.18972273021936417,
|
|
"num_tokens": 8733149.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.478960704803467,
|
|
"epoch": 3.8989028213166144,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038449991915679273,
|
|
"loss": 5.0414,
|
|
"mean_token_accuracy": 0.1944599613547325,
|
|
"num_tokens": 8742000.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.567214775085449,
|
|
"epoch": 3.9028213166144203,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003842370541820915,
|
|
"loss": 5.1178,
|
|
"mean_token_accuracy": 0.187737275660038,
|
|
"num_tokens": 8751069.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.53871603012085,
|
|
"epoch": 3.906739811912226,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038397399398306243,
|
|
"loss": 5.0393,
|
|
"mean_token_accuracy": 0.18819471895694734,
|
|
"num_tokens": 8760092.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.520770692825318,
|
|
"epoch": 3.9106583072100314,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038371073902983684,
|
|
"loss": 5.0458,
|
|
"mean_token_accuracy": 0.19228698164224625,
|
|
"num_tokens": 8768039.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.542436027526856,
|
|
"epoch": 3.914576802507837,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000383447289792894,
|
|
"loss": 5.1,
|
|
"mean_token_accuracy": 0.18285368680953978,
|
|
"num_tokens": 8776630.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.57312502861023,
|
|
"epoch": 3.9184952978056424,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038318364674306036,
|
|
"loss": 5.0533,
|
|
"mean_token_accuracy": 0.18965477794408797,
|
|
"num_tokens": 8785272.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 3.9184952978056424,
|
|
"eval_entropy": 5.36408500153889,
|
|
"eval_loss": 5.6676201820373535,
|
|
"eval_mean_token_accuracy": 0.16819456405937672,
|
|
"eval_num_tokens": 8785272.0,
|
|
"eval_runtime": 2.1603,
|
|
"eval_samples_per_second": 1908.045,
|
|
"eval_steps_per_second": 238.853,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.5379270076751705,
|
|
"epoch": 3.9224137931034484,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038291981035150883,
|
|
"loss": 5.0667,
|
|
"mean_token_accuracy": 0.19032427221536635,
|
|
"num_tokens": 8794197.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.554198980331421,
|
|
"epoch": 3.926332288401254,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003826557810897579,
|
|
"loss": 5.1025,
|
|
"mean_token_accuracy": 0.18662202209234238,
|
|
"num_tokens": 8803195.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.586854553222656,
|
|
"epoch": 3.9302507836990594,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003823915594296705,
|
|
"loss": 5.0533,
|
|
"mean_token_accuracy": 0.19136017560958862,
|
|
"num_tokens": 8811772.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.422827529907226,
|
|
"epoch": 3.9341692789968654,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003821271458434538,
|
|
"loss": 5.0162,
|
|
"mean_token_accuracy": 0.19654913246631622,
|
|
"num_tokens": 8820482.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.453502511978149,
|
|
"epoch": 3.938087774294671,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038186254080365756,
|
|
"loss": 5.0879,
|
|
"mean_token_accuracy": 0.1923722356557846,
|
|
"num_tokens": 8828761.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.4765932083129885,
|
|
"epoch": 3.9420062695924765,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000381597744783174,
|
|
"loss": 4.9846,
|
|
"mean_token_accuracy": 0.19031898230314254,
|
|
"num_tokens": 8837165.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.5273174285888675,
|
|
"epoch": 3.945924764890282,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038133275825523645,
|
|
"loss": 5.0518,
|
|
"mean_token_accuracy": 0.19188373535871506,
|
|
"num_tokens": 8845472.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.5340431213378904,
|
|
"epoch": 3.9498432601880875,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038106758169341887,
|
|
"loss": 5.0729,
|
|
"mean_token_accuracy": 0.19060606360435486,
|
|
"num_tokens": 8855246.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.572251415252685,
|
|
"epoch": 3.9537617554858935,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003808022155716348,
|
|
"loss": 5.1248,
|
|
"mean_token_accuracy": 0.19278124570846558,
|
|
"num_tokens": 8863667.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.504932403564453,
|
|
"epoch": 3.957680250783699,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003805366603641364,
|
|
"loss": 4.9808,
|
|
"mean_token_accuracy": 0.19673547744750977,
|
|
"num_tokens": 8872306.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.431030416488648,
|
|
"epoch": 3.9615987460815045,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038027091654551406,
|
|
"loss": 4.9971,
|
|
"mean_token_accuracy": 0.194322569668293,
|
|
"num_tokens": 8882037.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.441543006896973,
|
|
"epoch": 3.9655172413793105,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038000498459069487,
|
|
"loss": 5.0524,
|
|
"mean_token_accuracy": 0.19480356127023696,
|
|
"num_tokens": 8890685.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.472086000442505,
|
|
"epoch": 3.969435736677116,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003797388649749426,
|
|
"loss": 5.033,
|
|
"mean_token_accuracy": 0.19273014813661576,
|
|
"num_tokens": 8899927.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.529286813735962,
|
|
"epoch": 3.9733542319749215,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003794725581738559,
|
|
"loss": 5.0777,
|
|
"mean_token_accuracy": 0.18776633590459824,
|
|
"num_tokens": 8908571.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.5456408023834225,
|
|
"epoch": 3.9772727272727275,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037920606466336834,
|
|
"loss": 4.996,
|
|
"mean_token_accuracy": 0.1924750864505768,
|
|
"num_tokens": 8917246.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.547796773910522,
|
|
"epoch": 3.981191222570533,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000378939384919747,
|
|
"loss": 5.086,
|
|
"mean_token_accuracy": 0.18750833719968796,
|
|
"num_tokens": 8926116.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.571321773529053,
|
|
"epoch": 3.9851097178683386,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003786725194195918,
|
|
"loss": 5.1288,
|
|
"mean_token_accuracy": 0.1865018755197525,
|
|
"num_tokens": 8934914.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.557705688476562,
|
|
"epoch": 3.989028213166144,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 5.036,
|
|
"mean_token_accuracy": 0.18699363619089127,
|
|
"num_tokens": 8943321.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.53307318687439,
|
|
"epoch": 3.9929467084639496,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037813823305773883,
|
|
"loss": 5.0613,
|
|
"mean_token_accuracy": 0.18817638754844665,
|
|
"num_tokens": 8951848.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.581093978881836,
|
|
"epoch": 3.9968652037617556,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003778708131508974,
|
|
"loss": 5.088,
|
|
"mean_token_accuracy": 0.18577350527048112,
|
|
"num_tokens": 8961291.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.608343887329101,
|
|
"epoch": 4.0007836990595615,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037760320939723307,
|
|
"loss": 5.1058,
|
|
"mean_token_accuracy": 0.18915147334337234,
|
|
"num_tokens": 8971116.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.513419580459595,
|
|
"epoch": 4.004702194357367,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037733542227499727,
|
|
"loss": 4.7598,
|
|
"mean_token_accuracy": 0.2157080963253975,
|
|
"num_tokens": 8979362.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.567613983154297,
|
|
"epoch": 4.008620689655173,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00037706745226276893,
|
|
"loss": 4.9156,
|
|
"mean_token_accuracy": 0.2058505207300186,
|
|
"num_tokens": 8987728.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.514210605621338,
|
|
"epoch": 4.012539184952978,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003767992998394539,
|
|
"loss": 4.7646,
|
|
"mean_token_accuracy": 0.2089159071445465,
|
|
"num_tokens": 8995817.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.529784107208252,
|
|
"epoch": 4.016457680250784,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003765309654842842,
|
|
"loss": 4.8496,
|
|
"mean_token_accuracy": 0.20495108366012574,
|
|
"num_tokens": 9005683.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.416915321350098,
|
|
"epoch": 4.02037617554859,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003762624496768166,
|
|
"loss": 4.7346,
|
|
"mean_token_accuracy": 0.21893513798713685,
|
|
"num_tokens": 9015408.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.570098495483398,
|
|
"epoch": 4.024294670846395,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003759937528969325,
|
|
"loss": 4.8942,
|
|
"mean_token_accuracy": 0.20396323949098588,
|
|
"num_tokens": 9024381.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.513882780075074,
|
|
"epoch": 4.028213166144201,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037572487562483666,
|
|
"loss": 4.7788,
|
|
"mean_token_accuracy": 0.2112806186079979,
|
|
"num_tokens": 9033092.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.476330232620239,
|
|
"epoch": 4.032131661442007,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037545581834105623,
|
|
"loss": 4.8989,
|
|
"mean_token_accuracy": 0.21034932583570481,
|
|
"num_tokens": 9042251.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.463940143585205,
|
|
"epoch": 4.036050156739812,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037518658152644035,
|
|
"loss": 4.783,
|
|
"mean_token_accuracy": 0.20679629743099212,
|
|
"num_tokens": 9050628.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.518752098083496,
|
|
"epoch": 4.039968652037618,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003749171656621588,
|
|
"loss": 4.8132,
|
|
"mean_token_accuracy": 0.218623548746109,
|
|
"num_tokens": 9060080.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.524572420120239,
|
|
"epoch": 4.043887147335423,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003746475712297014,
|
|
"loss": 4.8218,
|
|
"mean_token_accuracy": 0.21361033618450165,
|
|
"num_tokens": 9069456.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.4999042510986325,
|
|
"epoch": 4.047805642633229,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003743777987108771,
|
|
"loss": 4.8346,
|
|
"mean_token_accuracy": 0.2145877406001091,
|
|
"num_tokens": 9078790.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.397429084777832,
|
|
"epoch": 4.051724137931035,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003741078485878132,
|
|
"loss": 4.8192,
|
|
"mean_token_accuracy": 0.20727519541978837,
|
|
"num_tokens": 9086888.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.511934757232666,
|
|
"epoch": 4.05564263322884,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003738377213429542,
|
|
"loss": 4.8562,
|
|
"mean_token_accuracy": 0.20303749293088913,
|
|
"num_tokens": 9095139.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.5287628173828125,
|
|
"epoch": 4.059561128526646,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003735674174590614,
|
|
"loss": 4.7848,
|
|
"mean_token_accuracy": 0.21545929163694383,
|
|
"num_tokens": 9103464.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.474196100234986,
|
|
"epoch": 4.063479623824452,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037329693741921166,
|
|
"loss": 4.7986,
|
|
"mean_token_accuracy": 0.21141539961099626,
|
|
"num_tokens": 9111290.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.365725326538086,
|
|
"epoch": 4.067398119122257,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003730262817067967,
|
|
"loss": 4.6594,
|
|
"mean_token_accuracy": 0.2169172540307045,
|
|
"num_tokens": 9120224.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.4814482688903805,
|
|
"epoch": 4.071316614420063,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003727554508055221,
|
|
"loss": 4.8174,
|
|
"mean_token_accuracy": 0.20918414145708084,
|
|
"num_tokens": 9128524.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.431653261184692,
|
|
"epoch": 4.075235109717869,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003724844451994066,
|
|
"loss": 4.7045,
|
|
"mean_token_accuracy": 0.2165905550122261,
|
|
"num_tokens": 9137026.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.467705774307251,
|
|
"epoch": 4.079153605015674,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037221326537278113,
|
|
"loss": 4.7933,
|
|
"mean_token_accuracy": 0.20697344541549684,
|
|
"num_tokens": 9146312.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.457153081893921,
|
|
"epoch": 4.08307210031348,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037194191181028806,
|
|
"loss": 4.7977,
|
|
"mean_token_accuracy": 0.21021723598241807,
|
|
"num_tokens": 9155226.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.536397266387939,
|
|
"epoch": 4.086990595611285,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037167038499688024,
|
|
"loss": 4.9414,
|
|
"mean_token_accuracy": 0.20643623620271684,
|
|
"num_tokens": 9164090.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.502128076553345,
|
|
"epoch": 4.090909090909091,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037139868541782,
|
|
"loss": 4.7712,
|
|
"mean_token_accuracy": 0.22177315205335618,
|
|
"num_tokens": 9173002.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.530263471603393,
|
|
"epoch": 4.094827586206897,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003711268135586787,
|
|
"loss": 4.8225,
|
|
"mean_token_accuracy": 0.20743722319602967,
|
|
"num_tokens": 9182020.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.435852670669556,
|
|
"epoch": 4.098746081504702,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037085476990533523,
|
|
"loss": 4.8376,
|
|
"mean_token_accuracy": 0.20410535633563995,
|
|
"num_tokens": 9190848.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.458233642578125,
|
|
"epoch": 4.102664576802508,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003705825549439759,
|
|
"loss": 4.7529,
|
|
"mean_token_accuracy": 0.21708489954471588,
|
|
"num_tokens": 9198808.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.558369159698486,
|
|
"epoch": 4.106583072100314,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037031016916109295,
|
|
"loss": 4.8876,
|
|
"mean_token_accuracy": 0.21121736317873002,
|
|
"num_tokens": 9207923.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.491108655929565,
|
|
"epoch": 4.110501567398119,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037003761304348374,
|
|
"loss": 4.8354,
|
|
"mean_token_accuracy": 0.20643606632947922,
|
|
"num_tokens": 9216759.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.41330509185791,
|
|
"epoch": 4.114420062695925,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00036976488707825056,
|
|
"loss": 4.8033,
|
|
"mean_token_accuracy": 0.20072809010744094,
|
|
"num_tokens": 9225951.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.428546476364136,
|
|
"epoch": 4.118338557993731,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00036949199175279876,
|
|
"loss": 4.8399,
|
|
"mean_token_accuracy": 0.2062118336558342,
|
|
"num_tokens": 9234413.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.58500828742981,
|
|
"epoch": 4.122257053291536,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003692189275548364,
|
|
"loss": 4.9663,
|
|
"mean_token_accuracy": 0.2000826597213745,
|
|
"num_tokens": 9244002.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.512565422058105,
|
|
"epoch": 4.126175548589342,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003689456949723737,
|
|
"loss": 4.7984,
|
|
"mean_token_accuracy": 0.21031637489795685,
|
|
"num_tokens": 9252689.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.5140220642089846,
|
|
"epoch": 4.130094043887147,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003686722944937215,
|
|
"loss": 4.7895,
|
|
"mean_token_accuracy": 0.2111169531941414,
|
|
"num_tokens": 9261118.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.427572202682495,
|
|
"epoch": 4.134012539184953,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036839872660749084,
|
|
"loss": 4.7926,
|
|
"mean_token_accuracy": 0.209438294172287,
|
|
"num_tokens": 9269205.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.451401996612549,
|
|
"epoch": 4.137931034482759,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003681249918025918,
|
|
"loss": 4.8163,
|
|
"mean_token_accuracy": 0.20893827676773072,
|
|
"num_tokens": 9277770.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.43188328742981,
|
|
"epoch": 4.141849529780564,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036785109056823297,
|
|
"loss": 4.8455,
|
|
"mean_token_accuracy": 0.21019160747528076,
|
|
"num_tokens": 9286589.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.502201747894287,
|
|
"epoch": 4.14576802507837,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036757702339392,
|
|
"loss": 4.824,
|
|
"mean_token_accuracy": 0.2045150563120842,
|
|
"num_tokens": 9295264.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.443982219696045,
|
|
"epoch": 4.149686520376176,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036730279076945574,
|
|
"loss": 4.8275,
|
|
"mean_token_accuracy": 0.207076396048069,
|
|
"num_tokens": 9303525.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.422685241699218,
|
|
"epoch": 4.153605015673981,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003670283931849382,
|
|
"loss": 4.8673,
|
|
"mean_token_accuracy": 0.20219362080097197,
|
|
"num_tokens": 9312839.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.469949674606323,
|
|
"epoch": 4.157523510971787,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00036675383113076023,
|
|
"loss": 4.836,
|
|
"mean_token_accuracy": 0.20911472886800767,
|
|
"num_tokens": 9321462.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.463937425613404,
|
|
"epoch": 4.161442006269592,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00036647910509760896,
|
|
"loss": 4.8441,
|
|
"mean_token_accuracy": 0.20798775404691697,
|
|
"num_tokens": 9329315.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.544539976119995,
|
|
"epoch": 4.165360501567398,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036620421557646414,
|
|
"loss": 4.83,
|
|
"mean_token_accuracy": 0.20499703288078308,
|
|
"num_tokens": 9336997.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.406382608413696,
|
|
"epoch": 4.169278996865204,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000365929163058598,
|
|
"loss": 4.7866,
|
|
"mean_token_accuracy": 0.21024256348609924,
|
|
"num_tokens": 9345364.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.465565156936646,
|
|
"epoch": 4.173197492163009,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 4.7865,
|
|
"mean_token_accuracy": 0.2123578131198883,
|
|
"num_tokens": 9353961.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.546124076843261,
|
|
"epoch": 4.177115987460815,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003653785709992462,
|
|
"loss": 4.9226,
|
|
"mean_token_accuracy": 0.20314039289951324,
|
|
"num_tokens": 9362901.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.4580528259277346,
|
|
"epoch": 4.181034482758621,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036510303244175775,
|
|
"loss": 4.8469,
|
|
"mean_token_accuracy": 0.20778143107891084,
|
|
"num_tokens": 9371800.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.446552658081055,
|
|
"epoch": 4.184952978056426,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036482733285554073,
|
|
"loss": 4.8367,
|
|
"mean_token_accuracy": 0.21264727860689164,
|
|
"num_tokens": 9380033.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.605759048461914,
|
|
"epoch": 4.188871473354232,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036455147273331507,
|
|
"loss": 4.9035,
|
|
"mean_token_accuracy": 0.20848263502120973,
|
|
"num_tokens": 9389264.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.449625873565674,
|
|
"epoch": 4.192789968652038,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003642754525680876,
|
|
"loss": 4.8215,
|
|
"mean_token_accuracy": 0.20863172113895417,
|
|
"num_tokens": 9398112.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.493907642364502,
|
|
"epoch": 4.196708463949843,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036399927285315104,
|
|
"loss": 4.9436,
|
|
"mean_token_accuracy": 0.19718344658613204,
|
|
"num_tokens": 9407400.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.469917583465576,
|
|
"epoch": 4.200626959247649,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003637229340820834,
|
|
"loss": 4.8272,
|
|
"mean_token_accuracy": 0.2022486373782158,
|
|
"num_tokens": 9416351.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.526383924484253,
|
|
"epoch": 4.204545454545454,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003634464367487467,
|
|
"loss": 4.8686,
|
|
"mean_token_accuracy": 0.2038404807448387,
|
|
"num_tokens": 9424577.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.4818004131317135,
|
|
"epoch": 4.20846394984326,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003631697813472867,
|
|
"loss": 4.8533,
|
|
"mean_token_accuracy": 0.21101049482822418,
|
|
"num_tokens": 9434242.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.4067178726196286,
|
|
"epoch": 4.212382445141066,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00036289296837213136,
|
|
"loss": 4.8009,
|
|
"mean_token_accuracy": 0.21353423595428467,
|
|
"num_tokens": 9443247.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.4146871089935305,
|
|
"epoch": 4.216300940438871,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00036261599831799034,
|
|
"loss": 4.7748,
|
|
"mean_token_accuracy": 0.21413675248622893,
|
|
"num_tokens": 9451998.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.513681221008301,
|
|
"epoch": 4.220219435736677,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003623388716798541,
|
|
"loss": 4.7494,
|
|
"mean_token_accuracy": 0.21292776465415955,
|
|
"num_tokens": 9459993.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.42663893699646,
|
|
"epoch": 4.224137931034483,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036206158895299277,
|
|
"loss": 4.7755,
|
|
"mean_token_accuracy": 0.21563078463077545,
|
|
"num_tokens": 9468036.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.364408397674561,
|
|
"epoch": 4.228056426332288,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036178415063295574,
|
|
"loss": 4.88,
|
|
"mean_token_accuracy": 0.21001806408166884,
|
|
"num_tokens": 9477031.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.478297281265259,
|
|
"epoch": 4.231974921630094,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00036150655721557,
|
|
"loss": 4.781,
|
|
"mean_token_accuracy": 0.21143459677696227,
|
|
"num_tokens": 9485687.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.463390684127807,
|
|
"epoch": 4.2358934169279,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003612288091969402,
|
|
"loss": 4.8198,
|
|
"mean_token_accuracy": 0.21180083155632018,
|
|
"num_tokens": 9494613.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.470304346084594,
|
|
"epoch": 4.239811912225705,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036095090707344697,
|
|
"loss": 4.8267,
|
|
"mean_token_accuracy": 0.20474908053874968,
|
|
"num_tokens": 9503522.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.446960973739624,
|
|
"epoch": 4.243730407523511,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036067285134174654,
|
|
"loss": 4.8277,
|
|
"mean_token_accuracy": 0.20671806037425994,
|
|
"num_tokens": 9512293.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.537747001647949,
|
|
"epoch": 4.247648902821316,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003603946424987696,
|
|
"loss": 4.945,
|
|
"mean_token_accuracy": 0.20333030074834824,
|
|
"num_tokens": 9520999.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.490629386901856,
|
|
"epoch": 4.251567398119122,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036011628104172045,
|
|
"loss": 4.8653,
|
|
"mean_token_accuracy": 0.20180012732744218,
|
|
"num_tokens": 9529798.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.528127384185791,
|
|
"epoch": 4.255485893416928,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035983776746807613,
|
|
"loss": 4.8167,
|
|
"mean_token_accuracy": 0.20935892760753633,
|
|
"num_tokens": 9538668.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.519613361358642,
|
|
"epoch": 4.259404388714733,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003595591022755855,
|
|
"loss": 4.8743,
|
|
"mean_token_accuracy": 0.20259860903024673,
|
|
"num_tokens": 9546973.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.373533773422241,
|
|
"epoch": 4.263322884012539,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003592802859622684,
|
|
"loss": 4.7848,
|
|
"mean_token_accuracy": 0.20720868557691574,
|
|
"num_tokens": 9555947.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.46702241897583,
|
|
"epoch": 4.267241379310345,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000359001319026415,
|
|
"loss": 4.8724,
|
|
"mean_token_accuracy": 0.2071203202009201,
|
|
"num_tokens": 9564730.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.420458793640137,
|
|
"epoch": 4.27115987460815,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035872220196658426,
|
|
"loss": 4.8348,
|
|
"mean_token_accuracy": 0.20729437321424485,
|
|
"num_tokens": 9573470.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.503951358795166,
|
|
"epoch": 4.275078369905956,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035844293528160384,
|
|
"loss": 4.8599,
|
|
"mean_token_accuracy": 0.20201792418956757,
|
|
"num_tokens": 9581607.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.462487077713012,
|
|
"epoch": 4.278996865203762,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003581635194705682,
|
|
"loss": 4.8212,
|
|
"mean_token_accuracy": 0.20852640718221666,
|
|
"num_tokens": 9589801.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.479101085662842,
|
|
"epoch": 4.282915360501567,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00035788395503283914,
|
|
"loss": 4.8795,
|
|
"mean_token_accuracy": 0.19981007725000383,
|
|
"num_tokens": 9599983.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.507000684738159,
|
|
"epoch": 4.286833855799373,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00035760424246804334,
|
|
"loss": 4.8021,
|
|
"mean_token_accuracy": 0.2108507975935936,
|
|
"num_tokens": 9608681.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.438380241394043,
|
|
"epoch": 4.290752351097178,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003573243822760726,
|
|
"loss": 4.8043,
|
|
"mean_token_accuracy": 0.21195460259914398,
|
|
"num_tokens": 9616461.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.48713321685791,
|
|
"epoch": 4.294670846394984,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003570443749570825,
|
|
"loss": 4.9002,
|
|
"mean_token_accuracy": 0.20393263101577758,
|
|
"num_tokens": 9625480.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.497563886642456,
|
|
"epoch": 4.29858934169279,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003567642210114914,
|
|
"loss": 4.8558,
|
|
"mean_token_accuracy": 0.20644507110118865,
|
|
"num_tokens": 9634172.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.408561325073242,
|
|
"epoch": 4.302507836990595,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035648392093998,
|
|
"loss": 4.8363,
|
|
"mean_token_accuracy": 0.2095470666885376,
|
|
"num_tokens": 9643196.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.414874124526977,
|
|
"epoch": 4.306426332288401,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003562034752434899,
|
|
"loss": 4.7616,
|
|
"mean_token_accuracy": 0.21216128319501876,
|
|
"num_tokens": 9652440.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.455464696884155,
|
|
"epoch": 4.310344827586207,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003559228844232229,
|
|
"loss": 4.8354,
|
|
"mean_token_accuracy": 0.20523983091115952,
|
|
"num_tokens": 9661248.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.310344827586207,
|
|
"eval_entropy": 5.2803950697876685,
|
|
"eval_loss": 5.6886115074157715,
|
|
"eval_mean_token_accuracy": 0.16800745053924332,
|
|
"eval_num_tokens": 9661248.0,
|
|
"eval_runtime": 2.1544,
|
|
"eval_samples_per_second": 1913.311,
|
|
"eval_steps_per_second": 239.512,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.497241830825805,
|
|
"epoch": 4.314263322884012,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003556421489806406,
|
|
"loss": 4.878,
|
|
"mean_token_accuracy": 0.20082256942987442,
|
|
"num_tokens": 9670544.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.482183742523193,
|
|
"epoch": 4.318181818181818,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003553612694174624,
|
|
"loss": 4.8399,
|
|
"mean_token_accuracy": 0.20866239368915557,
|
|
"num_tokens": 9679331.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.573589849472046,
|
|
"epoch": 4.322100313479623,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003550802462356659,
|
|
"loss": 4.9233,
|
|
"mean_token_accuracy": 0.19969965070486068,
|
|
"num_tokens": 9688211.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.3721638202667235,
|
|
"epoch": 4.326018808777429,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035479907993748484,
|
|
"loss": 4.7668,
|
|
"mean_token_accuracy": 0.2158796101808548,
|
|
"num_tokens": 9696231.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.443095684051514,
|
|
"epoch": 4.329937304075235,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035451777102540915,
|
|
"loss": 4.8554,
|
|
"mean_token_accuracy": 0.19912099242210388,
|
|
"num_tokens": 9705577.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.5596246242523195,
|
|
"epoch": 4.33385579937304,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003542363200021835,
|
|
"loss": 4.8786,
|
|
"mean_token_accuracy": 0.20782244205474854,
|
|
"num_tokens": 9715003.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.436708593368531,
|
|
"epoch": 4.337774294670846,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003539547273708064,
|
|
"loss": 4.768,
|
|
"mean_token_accuracy": 0.2091422736644745,
|
|
"num_tokens": 9723513.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.371388626098633,
|
|
"epoch": 4.341692789968652,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00035367299363452966,
|
|
"loss": 4.8407,
|
|
"mean_token_accuracy": 0.20161120295524598,
|
|
"num_tokens": 9731800.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.418384265899658,
|
|
"epoch": 4.3456112852664575,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00035339111929685694,
|
|
"loss": 4.8171,
|
|
"mean_token_accuracy": 0.20139016807079316,
|
|
"num_tokens": 9740356.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.510761499404907,
|
|
"epoch": 4.349529780564263,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035310910486154365,
|
|
"loss": 4.8544,
|
|
"mean_token_accuracy": 0.19833854585886002,
|
|
"num_tokens": 9749529.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.502518987655639,
|
|
"epoch": 4.353448275862069,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003528269508325951,
|
|
"loss": 4.841,
|
|
"mean_token_accuracy": 0.210841603577137,
|
|
"num_tokens": 9758750.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.439648675918579,
|
|
"epoch": 4.3573667711598745,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 4.9125,
|
|
"mean_token_accuracy": 0.20635788440704345,
|
|
"num_tokens": 9766992.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.522937488555908,
|
|
"epoch": 4.36128526645768,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035226222601106105,
|
|
"loss": 4.9121,
|
|
"mean_token_accuracy": 0.20442684590816498,
|
|
"num_tokens": 9777022.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.441367530822754,
|
|
"epoch": 4.3652037617554855,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003519796562277303,
|
|
"loss": 4.6924,
|
|
"mean_token_accuracy": 0.21872189044952392,
|
|
"num_tokens": 9785863.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.417543983459472,
|
|
"epoch": 4.3691222570532915,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003516969488692723,
|
|
"loss": 4.8366,
|
|
"mean_token_accuracy": 0.20567006915807723,
|
|
"num_tokens": 9794357.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.358813190460205,
|
|
"epoch": 4.3730407523510975,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003514141044409308,
|
|
"loss": 4.8164,
|
|
"mean_token_accuracy": 0.2044922113418579,
|
|
"num_tokens": 9803128.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.45006046295166,
|
|
"epoch": 4.3769592476489025,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003511311234481947,
|
|
"loss": 4.9065,
|
|
"mean_token_accuracy": 0.1976408913731575,
|
|
"num_tokens": 9812046.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.423652172088623,
|
|
"epoch": 4.3808777429467085,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035084800639679695,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.21023883670568466,
|
|
"num_tokens": 9820935.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.492334699630737,
|
|
"epoch": 4.3847962382445145,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035056475379271356,
|
|
"loss": 4.9154,
|
|
"mean_token_accuracy": 0.20375131964683532,
|
|
"num_tokens": 9829585.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.570540571212769,
|
|
"epoch": 4.38871473354232,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003502813661421629,
|
|
"loss": 4.97,
|
|
"mean_token_accuracy": 0.1994202360510826,
|
|
"num_tokens": 9839302.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.481574201583863,
|
|
"epoch": 4.3926332288401255,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003499978439516047,
|
|
"loss": 4.8769,
|
|
"mean_token_accuracy": 0.2035578429698944,
|
|
"num_tokens": 9848330.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.411958694458008,
|
|
"epoch": 4.396551724137931,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034971418772773895,
|
|
"loss": 4.8313,
|
|
"mean_token_accuracy": 0.211450931429863,
|
|
"num_tokens": 9856928.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.478770923614502,
|
|
"epoch": 4.400470219435737,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003494303979775054,
|
|
"loss": 4.8723,
|
|
"mean_token_accuracy": 0.20543282330036164,
|
|
"num_tokens": 9865464.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.479744958877563,
|
|
"epoch": 4.4043887147335425,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034914647520808235,
|
|
"loss": 4.8508,
|
|
"mean_token_accuracy": 0.21272414177656174,
|
|
"num_tokens": 9874345.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.3961564064025875,
|
|
"epoch": 4.408307210031348,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003488624199268859,
|
|
"loss": 4.8006,
|
|
"mean_token_accuracy": 0.2147693619132042,
|
|
"num_tokens": 9883135.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.487041664123535,
|
|
"epoch": 4.412225705329154,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003485782326415687,
|
|
"loss": 4.8903,
|
|
"mean_token_accuracy": 0.20854106545448303,
|
|
"num_tokens": 9892229.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.480099201202393,
|
|
"epoch": 4.41614420062696,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003482939138600197,
|
|
"loss": 4.8646,
|
|
"mean_token_accuracy": 0.20610063672065734,
|
|
"num_tokens": 9901044.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.450689029693604,
|
|
"epoch": 4.420062695924765,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003480094640903626,
|
|
"loss": 4.9062,
|
|
"mean_token_accuracy": 0.19999598264694213,
|
|
"num_tokens": 9909818.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.407378673553467,
|
|
"epoch": 4.423981191222571,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034772488384095517,
|
|
"loss": 4.7691,
|
|
"mean_token_accuracy": 0.21420664936304093,
|
|
"num_tokens": 9919295.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.46484751701355,
|
|
"epoch": 4.427899686520377,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00034744017362038854,
|
|
"loss": 4.8577,
|
|
"mean_token_accuracy": 0.2084245875477791,
|
|
"num_tokens": 9927539.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.467625713348388,
|
|
"epoch": 4.431818181818182,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034715533393748604,
|
|
"loss": 4.9087,
|
|
"mean_token_accuracy": 0.20310912728309632,
|
|
"num_tokens": 9936540.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.431950855255127,
|
|
"epoch": 4.435736677115988,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034687036530130244,
|
|
"loss": 4.7688,
|
|
"mean_token_accuracy": 0.21449691653251649,
|
|
"num_tokens": 9945065.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.425051593780518,
|
|
"epoch": 4.439655172413793,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003465852682211227,
|
|
"loss": 4.805,
|
|
"mean_token_accuracy": 0.2089216575026512,
|
|
"num_tokens": 9954155.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.421067142486573,
|
|
"epoch": 4.443573667711599,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003463000432064617,
|
|
"loss": 4.8786,
|
|
"mean_token_accuracy": 0.20336292535066605,
|
|
"num_tokens": 9963269.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.517705726623535,
|
|
"epoch": 4.447492163009405,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034601469076706276,
|
|
"loss": 4.9084,
|
|
"mean_token_accuracy": 0.20399987250566481,
|
|
"num_tokens": 9972688.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.560425186157227,
|
|
"epoch": 4.45141065830721,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003457292114128968,
|
|
"loss": 4.8632,
|
|
"mean_token_accuracy": 0.20390735566616058,
|
|
"num_tokens": 9981283.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.443235588073731,
|
|
"epoch": 4.455329153605016,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000345443605654162,
|
|
"loss": 4.8818,
|
|
"mean_token_accuracy": 0.20368001461029053,
|
|
"num_tokens": 9989870.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.4746335506439205,
|
|
"epoch": 4.459247648902822,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00034515787400128195,
|
|
"loss": 4.8913,
|
|
"mean_token_accuracy": 0.2063031017780304,
|
|
"num_tokens": 9998136.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.45536961555481,
|
|
"epoch": 4.463166144200627,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003448720169649054,
|
|
"loss": 4.8263,
|
|
"mean_token_accuracy": 0.207182377576828,
|
|
"num_tokens": 10006730.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.447148609161377,
|
|
"epoch": 4.467084639498433,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003445860350559055,
|
|
"loss": 4.8426,
|
|
"mean_token_accuracy": 0.21463246196508406,
|
|
"num_tokens": 10015517.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.422385931015015,
|
|
"epoch": 4.471003134796238,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000344299928785378,
|
|
"loss": 4.8445,
|
|
"mean_token_accuracy": 0.20539455860853195,
|
|
"num_tokens": 10024552.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.445158815383911,
|
|
"epoch": 4.474921630094044,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00034401369866464133,
|
|
"loss": 4.8401,
|
|
"mean_token_accuracy": 0.20810834616422652,
|
|
"num_tokens": 10033523.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.440689992904663,
|
|
"epoch": 4.47884012539185,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034372734520523505,
|
|
"loss": 4.7348,
|
|
"mean_token_accuracy": 0.21276017874479294,
|
|
"num_tokens": 10041988.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.377323579788208,
|
|
"epoch": 4.482758620689655,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003434408689189193,
|
|
"loss": 4.8238,
|
|
"mean_token_accuracy": 0.21314817070960998,
|
|
"num_tokens": 10050543.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.4567423343658445,
|
|
"epoch": 4.486677115987461,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003431542703176737,
|
|
"loss": 4.8498,
|
|
"mean_token_accuracy": 0.2034946858882904,
|
|
"num_tokens": 10059300.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.459370422363281,
|
|
"epoch": 4.490595611285267,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003428675499136963,
|
|
"loss": 4.8655,
|
|
"mean_token_accuracy": 0.2078411966562271,
|
|
"num_tokens": 10068535.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.536391830444336,
|
|
"epoch": 4.494514106583072,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000342580708219403,
|
|
"loss": 4.9418,
|
|
"mean_token_accuracy": 0.20440726578235627,
|
|
"num_tokens": 10077731.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.481202125549316,
|
|
"epoch": 4.498432601880878,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003422937457474264,
|
|
"loss": 4.8363,
|
|
"mean_token_accuracy": 0.20314988791942595,
|
|
"num_tokens": 10087146.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.456269550323486,
|
|
"epoch": 4.502351097178684,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034200666301061495,
|
|
"loss": 4.8409,
|
|
"mean_token_accuracy": 0.20937649309635162,
|
|
"num_tokens": 10096659.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.484100866317749,
|
|
"epoch": 4.506269592476489,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034171946052203217,
|
|
"loss": 4.9371,
|
|
"mean_token_accuracy": 0.19988409280776978,
|
|
"num_tokens": 10105403.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.5373283386230465,
|
|
"epoch": 4.510188087774295,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034143213879495526,
|
|
"loss": 4.8555,
|
|
"mean_token_accuracy": 0.2082296371459961,
|
|
"num_tokens": 10115410.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.460855197906494,
|
|
"epoch": 4.514106583072101,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000341144698342875,
|
|
"loss": 4.8676,
|
|
"mean_token_accuracy": 0.20223168581724166,
|
|
"num_tokens": 10123800.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.506167268753051,
|
|
"epoch": 4.518025078369906,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003408571396794939,
|
|
"loss": 4.9081,
|
|
"mean_token_accuracy": 0.19684920012950896,
|
|
"num_tokens": 10132775.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.472491884231568,
|
|
"epoch": 4.521943573667712,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003405694633187259,
|
|
"loss": 4.7976,
|
|
"mean_token_accuracy": 0.21152450442314147,
|
|
"num_tokens": 10141811.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.3745176792144775,
|
|
"epoch": 4.525862068965517,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034028166977469545,
|
|
"loss": 4.8105,
|
|
"mean_token_accuracy": 0.21029423624277116,
|
|
"num_tokens": 10150823.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.464593601226807,
|
|
"epoch": 4.529780564263323,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033999375956173607,
|
|
"loss": 4.837,
|
|
"mean_token_accuracy": 0.2068099483847618,
|
|
"num_tokens": 10158983.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.518088531494141,
|
|
"epoch": 4.533699059561129,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003397057331943902,
|
|
"loss": 4.8307,
|
|
"mean_token_accuracy": 0.20385924875736236,
|
|
"num_tokens": 10167596.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.478815126419067,
|
|
"epoch": 4.537617554858934,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003394175911874076,
|
|
"loss": 4.9147,
|
|
"mean_token_accuracy": 0.20867562144994736,
|
|
"num_tokens": 10176216.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.42698655128479,
|
|
"epoch": 4.54153605015674,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 4.8079,
|
|
"mean_token_accuracy": 0.20513638854026794,
|
|
"num_tokens": 10185685.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.426970911026001,
|
|
"epoch": 4.545454545454545,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003388409623145637,
|
|
"loss": 4.81,
|
|
"mean_token_accuracy": 0.20208559185266495,
|
|
"num_tokens": 10193679.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.42051420211792,
|
|
"epoch": 4.549373040752351,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00033855247647923177,
|
|
"loss": 4.8601,
|
|
"mean_token_accuracy": 0.20875582844018936,
|
|
"num_tokens": 10201154.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.411418533325195,
|
|
"epoch": 4.553291536050157,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003382638770653198,
|
|
"loss": 4.7843,
|
|
"mean_token_accuracy": 0.21591236889362336,
|
|
"num_tokens": 10209180.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.500545501708984,
|
|
"epoch": 4.557210031347962,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003379751645886017,
|
|
"loss": 4.8839,
|
|
"mean_token_accuracy": 0.19719190150499344,
|
|
"num_tokens": 10217430.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.4424763202667235,
|
|
"epoch": 4.561128526645768,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003376863395650537,
|
|
"loss": 4.9366,
|
|
"mean_token_accuracy": 0.20359685271978378,
|
|
"num_tokens": 10226453.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.568612003326416,
|
|
"epoch": 4.565047021943574,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033739740251085263,
|
|
"loss": 4.9501,
|
|
"mean_token_accuracy": 0.2091359555721283,
|
|
"num_tokens": 10235005.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.561180877685547,
|
|
"epoch": 4.568965517241379,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033710835394237603,
|
|
"loss": 4.8854,
|
|
"mean_token_accuracy": 0.20939993113279343,
|
|
"num_tokens": 10243800.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.455348062515259,
|
|
"epoch": 4.572884012539185,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033681919437620066,
|
|
"loss": 4.8394,
|
|
"mean_token_accuracy": 0.21136530339717866,
|
|
"num_tokens": 10252370.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.461749601364136,
|
|
"epoch": 4.576802507836991,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003365299243291014,
|
|
"loss": 4.914,
|
|
"mean_token_accuracy": 0.2008674204349518,
|
|
"num_tokens": 10260777.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.441798496246338,
|
|
"epoch": 4.580721003134796,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033624054431805067,
|
|
"loss": 4.9211,
|
|
"mean_token_accuracy": 0.19799282401800156,
|
|
"num_tokens": 10270571.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.5008080959320065,
|
|
"epoch": 4.584639498432602,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033595105486021763,
|
|
"loss": 4.8918,
|
|
"mean_token_accuracy": 0.20493955612182618,
|
|
"num_tokens": 10279284.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.422674036026001,
|
|
"epoch": 4.588557993730408,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033566145647296675,
|
|
"loss": 4.783,
|
|
"mean_token_accuracy": 0.2057187795639038,
|
|
"num_tokens": 10288610.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.424646329879761,
|
|
"epoch": 4.592476489028213,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033537174967385744,
|
|
"loss": 4.7904,
|
|
"mean_token_accuracy": 0.2089690774679184,
|
|
"num_tokens": 10296603.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.418612718582153,
|
|
"epoch": 4.596394984326019,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033508193498064254,
|
|
"loss": 4.7907,
|
|
"mean_token_accuracy": 0.21070562154054642,
|
|
"num_tokens": 10305199.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.392703628540039,
|
|
"epoch": 4.600313479623824,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033479201291126807,
|
|
"loss": 4.746,
|
|
"mean_token_accuracy": 0.21014370173215866,
|
|
"num_tokens": 10313967.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.472643661499023,
|
|
"epoch": 4.60423197492163,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003345019839838717,
|
|
"loss": 4.8411,
|
|
"mean_token_accuracy": 0.21318153738975526,
|
|
"num_tokens": 10322532.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.393186712265015,
|
|
"epoch": 4.608150470219436,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003342118487167822,
|
|
"loss": 4.7206,
|
|
"mean_token_accuracy": 0.2130552500486374,
|
|
"num_tokens": 10330631.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.3743401050567625,
|
|
"epoch": 4.612068965517241,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033392160762851813,
|
|
"loss": 4.7493,
|
|
"mean_token_accuracy": 0.2140085682272911,
|
|
"num_tokens": 10339229.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.499026966094971,
|
|
"epoch": 4.615987460815047,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003336312612377877,
|
|
"loss": 4.864,
|
|
"mean_token_accuracy": 0.20080927163362502,
|
|
"num_tokens": 10348882.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.581766128540039,
|
|
"epoch": 4.619905956112853,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033334081006348683,
|
|
"loss": 4.9917,
|
|
"mean_token_accuracy": 0.19342346638441085,
|
|
"num_tokens": 10358154.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.502346467971802,
|
|
"epoch": 4.623824451410658,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003330502546246988,
|
|
"loss": 4.8993,
|
|
"mean_token_accuracy": 0.20678748637437822,
|
|
"num_tokens": 10367100.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.367815542221069,
|
|
"epoch": 4.627742946708464,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033275959544069335,
|
|
"loss": 4.6984,
|
|
"mean_token_accuracy": 0.22344945818185807,
|
|
"num_tokens": 10375711.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.4175632953643795,
|
|
"epoch": 4.631661442006269,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033246883303092564,
|
|
"loss": 4.8302,
|
|
"mean_token_accuracy": 0.21189723312854766,
|
|
"num_tokens": 10384513.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.521723031997681,
|
|
"epoch": 4.635579937304075,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033217796791503503,
|
|
"loss": 4.906,
|
|
"mean_token_accuracy": 0.20050406455993652,
|
|
"num_tokens": 10393564.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.512802600860596,
|
|
"epoch": 4.639498432601881,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003318870006128449,
|
|
"loss": 4.9073,
|
|
"mean_token_accuracy": 0.20143700689077376,
|
|
"num_tokens": 10402042.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.434813833236694,
|
|
"epoch": 4.643416927899686,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003315959316443608,
|
|
"loss": 4.7797,
|
|
"mean_token_accuracy": 0.21071168780326843,
|
|
"num_tokens": 10410847.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.470376062393188,
|
|
"epoch": 4.647335423197492,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003313047615297703,
|
|
"loss": 4.8837,
|
|
"mean_token_accuracy": 0.19958594888448716,
|
|
"num_tokens": 10420366.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.427135610580445,
|
|
"epoch": 4.651253918495298,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033101349078944165,
|
|
"loss": 4.8048,
|
|
"mean_token_accuracy": 0.206406269967556,
|
|
"num_tokens": 10428870.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.41593074798584,
|
|
"epoch": 4.655172413793103,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003307221199439227,
|
|
"loss": 4.8695,
|
|
"mean_token_accuracy": 0.1988285094499588,
|
|
"num_tokens": 10437714.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.362565422058106,
|
|
"epoch": 4.659090909090909,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033043064951394045,
|
|
"loss": 4.7705,
|
|
"mean_token_accuracy": 0.20917622447013856,
|
|
"num_tokens": 10447198.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.441958904266357,
|
|
"epoch": 4.663009404388715,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00033013908002039994,
|
|
"loss": 4.8343,
|
|
"mean_token_accuracy": 0.21132338643074036,
|
|
"num_tokens": 10455790.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.496818542480469,
|
|
"epoch": 4.66692789968652,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032984741198438305,
|
|
"loss": 4.9288,
|
|
"mean_token_accuracy": 0.20463453084230424,
|
|
"num_tokens": 10464094.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.466181945800781,
|
|
"epoch": 4.670846394984326,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003295556459271478,
|
|
"loss": 4.8923,
|
|
"mean_token_accuracy": 0.2061282441020012,
|
|
"num_tokens": 10473201.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.469547748565674,
|
|
"epoch": 4.674764890282132,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003292637823701276,
|
|
"loss": 4.8895,
|
|
"mean_token_accuracy": 0.19792680442333221,
|
|
"num_tokens": 10482617.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.453397941589356,
|
|
"epoch": 4.678683385579937,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003289718218349298,
|
|
"loss": 4.7713,
|
|
"mean_token_accuracy": 0.2151390552520752,
|
|
"num_tokens": 10491702.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.55360860824585,
|
|
"epoch": 4.682601880877743,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003286797648433351,
|
|
"loss": 4.9801,
|
|
"mean_token_accuracy": 0.19853509664535524,
|
|
"num_tokens": 10501189.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.4602171897888185,
|
|
"epoch": 4.686520376175548,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000328387611917297,
|
|
"loss": 4.8457,
|
|
"mean_token_accuracy": 0.20577973127365112,
|
|
"num_tokens": 10510272.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.525967741012574,
|
|
"epoch": 4.690438871473354,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003280953635789401,
|
|
"loss": 4.9247,
|
|
"mean_token_accuracy": 0.2049766004085541,
|
|
"num_tokens": 10518782.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.422622442245483,
|
|
"epoch": 4.69435736677116,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003278030203505594,
|
|
"loss": 4.8516,
|
|
"mean_token_accuracy": 0.20228924453258515,
|
|
"num_tokens": 10528262.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.42066216468811,
|
|
"epoch": 4.698275862068965,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003275105827546197,
|
|
"loss": 4.8413,
|
|
"mean_token_accuracy": 0.2067961126565933,
|
|
"num_tokens": 10536399.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.503605079650879,
|
|
"epoch": 4.702194357366771,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032721805131375446,
|
|
"loss": 4.9012,
|
|
"mean_token_accuracy": 0.20219298601150512,
|
|
"num_tokens": 10544842.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 4.702194357366771,
|
|
"eval_entropy": 5.317808843398279,
|
|
"eval_loss": 5.6548027992248535,
|
|
"eval_mean_token_accuracy": 0.17102461573920508,
|
|
"eval_num_tokens": 10544842.0,
|
|
"eval_runtime": 2.3545,
|
|
"eval_samples_per_second": 1750.681,
|
|
"eval_steps_per_second": 219.154,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.479676723480225,
|
|
"epoch": 4.706112852664576,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003269254265507647,
|
|
"loss": 4.8084,
|
|
"mean_token_accuracy": 0.21306666731834412,
|
|
"num_tokens": 10553630.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.41502194404602,
|
|
"epoch": 4.710031347962382,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00032663270898861845,
|
|
"loss": 4.8308,
|
|
"mean_token_accuracy": 0.20325940251350402,
|
|
"num_tokens": 10561714.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.455433702468872,
|
|
"epoch": 4.713949843260188,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032633989915044944,
|
|
"loss": 4.903,
|
|
"mean_token_accuracy": 0.1989065244793892,
|
|
"num_tokens": 10571075.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.49522852897644,
|
|
"epoch": 4.717868338557993,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00032604699755955615,
|
|
"loss": 4.8082,
|
|
"mean_token_accuracy": 0.21177192628383637,
|
|
"num_tokens": 10579698.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.416851902008057,
|
|
"epoch": 4.721786833855799,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032575400473940135,
|
|
"loss": 4.8343,
|
|
"mean_token_accuracy": 0.20871866941452027,
|
|
"num_tokens": 10588060.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.382996702194214,
|
|
"epoch": 4.725705329153605,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 4.861,
|
|
"mean_token_accuracy": 0.2051233768463135,
|
|
"num_tokens": 10596482.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.430132627487183,
|
|
"epoch": 4.72962382445141,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003251677475059721,
|
|
"loss": 4.8781,
|
|
"mean_token_accuracy": 0.20395821183919907,
|
|
"num_tokens": 10605630.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.511685752868653,
|
|
"epoch": 4.733542319749216,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032487448414043435,
|
|
"loss": 4.8617,
|
|
"mean_token_accuracy": 0.21270882487297058,
|
|
"num_tokens": 10614772.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.513677501678467,
|
|
"epoch": 4.737460815047022,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032458113164110665,
|
|
"loss": 4.8622,
|
|
"mean_token_accuracy": 0.20942794233560563,
|
|
"num_tokens": 10623583.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.444927120208741,
|
|
"epoch": 4.741379310344827,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003242876905322575,
|
|
"loss": 4.8196,
|
|
"mean_token_accuracy": 0.22010601162910462,
|
|
"num_tokens": 10632136.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.432428073883057,
|
|
"epoch": 4.745297805642633,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003239941613383138,
|
|
"loss": 4.9313,
|
|
"mean_token_accuracy": 0.19969863444566727,
|
|
"num_tokens": 10641306.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.485778427124023,
|
|
"epoch": 4.749216300940439,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032370054458385973,
|
|
"loss": 4.8598,
|
|
"mean_token_accuracy": 0.20574829280376433,
|
|
"num_tokens": 10650866.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.534658622741699,
|
|
"epoch": 4.753134796238244,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000323406840793636,
|
|
"loss": 4.8668,
|
|
"mean_token_accuracy": 0.21293672323226928,
|
|
"num_tokens": 10659252.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.434871101379395,
|
|
"epoch": 4.75705329153605,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000323113050492539,
|
|
"loss": 4.7786,
|
|
"mean_token_accuracy": 0.20822954922914505,
|
|
"num_tokens": 10667987.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.459078121185303,
|
|
"epoch": 4.7609717868338555,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032281917420561967,
|
|
"loss": 4.9536,
|
|
"mean_token_accuracy": 0.19923066794872285,
|
|
"num_tokens": 10676524.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.437264013290405,
|
|
"epoch": 4.764890282131661,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003225252124580825,
|
|
"loss": 4.8743,
|
|
"mean_token_accuracy": 0.20358722954988479,
|
|
"num_tokens": 10685940.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.47061858177185,
|
|
"epoch": 4.768808777429467,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032223116577528475,
|
|
"loss": 4.8865,
|
|
"mean_token_accuracy": 0.2006385162472725,
|
|
"num_tokens": 10694522.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.497819519042968,
|
|
"epoch": 4.7727272727272725,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003219370346827356,
|
|
"loss": 4.8146,
|
|
"mean_token_accuracy": 0.20626613795757293,
|
|
"num_tokens": 10703868.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.409205961227417,
|
|
"epoch": 4.7766457680250785,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000321642819706095,
|
|
"loss": 4.8089,
|
|
"mean_token_accuracy": 0.20851438343524933,
|
|
"num_tokens": 10712044.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.511595153808594,
|
|
"epoch": 4.7805642633228835,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003213485213711729,
|
|
"loss": 5.0163,
|
|
"mean_token_accuracy": 0.19482806026935579,
|
|
"num_tokens": 10720806.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.420252132415771,
|
|
"epoch": 4.7844827586206895,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000321054140203928,
|
|
"loss": 4.745,
|
|
"mean_token_accuracy": 0.21585031300783158,
|
|
"num_tokens": 10729238.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.399838256835937,
|
|
"epoch": 4.7884012539184955,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003207596767304672,
|
|
"loss": 4.808,
|
|
"mean_token_accuracy": 0.20325492322444916,
|
|
"num_tokens": 10738175.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.518722581863403,
|
|
"epoch": 4.7923197492163006,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032046513147704456,
|
|
"loss": 4.8865,
|
|
"mean_token_accuracy": 0.20176589488983154,
|
|
"num_tokens": 10747475.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.458754348754883,
|
|
"epoch": 4.7962382445141065,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032017050497006,
|
|
"loss": 4.8345,
|
|
"mean_token_accuracy": 0.2083256721496582,
|
|
"num_tokens": 10756739.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.383963775634766,
|
|
"epoch": 4.8001567398119125,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00031987579773605913,
|
|
"loss": 4.8385,
|
|
"mean_token_accuracy": 0.20915457457304001,
|
|
"num_tokens": 10765141.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.481559896469117,
|
|
"epoch": 4.804075235109718,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003195810103017314,
|
|
"loss": 4.9407,
|
|
"mean_token_accuracy": 0.19920160323381425,
|
|
"num_tokens": 10774759.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.509219789505005,
|
|
"epoch": 4.8079937304075235,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003192861431939098,
|
|
"loss": 4.9016,
|
|
"mean_token_accuracy": 0.20372161865234376,
|
|
"num_tokens": 10783717.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.453897523880005,
|
|
"epoch": 4.8119122257053295,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003189911969395697,
|
|
"loss": 4.8946,
|
|
"mean_token_accuracy": 0.2069030225276947,
|
|
"num_tokens": 10792797.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.411454200744629,
|
|
"epoch": 4.815830721003135,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00031869617206582777,
|
|
"loss": 4.767,
|
|
"mean_token_accuracy": 0.21036030650138854,
|
|
"num_tokens": 10802249.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.4576003074646,
|
|
"epoch": 4.8197492163009406,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003184010690999413,
|
|
"loss": 4.8468,
|
|
"mean_token_accuracy": 0.20611144751310348,
|
|
"num_tokens": 10810722.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.524273157119751,
|
|
"epoch": 4.8236677115987465,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003181058885693073,
|
|
"loss": 4.9423,
|
|
"mean_token_accuracy": 0.19975735247135162,
|
|
"num_tokens": 10819115.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.487400150299072,
|
|
"epoch": 4.827586206896552,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003178106310014612,
|
|
"loss": 4.9194,
|
|
"mean_token_accuracy": 0.20226120203733444,
|
|
"num_tokens": 10827419.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.383101224899292,
|
|
"epoch": 4.831504702194358,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003175152969240759,
|
|
"loss": 4.793,
|
|
"mean_token_accuracy": 0.21193346977233887,
|
|
"num_tokens": 10836610.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.426295948028565,
|
|
"epoch": 4.835423197492163,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00031721988686496156,
|
|
"loss": 4.881,
|
|
"mean_token_accuracy": 0.20687132179737092,
|
|
"num_tokens": 10845127.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.474702501296997,
|
|
"epoch": 4.839341692789969,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003169244013520638,
|
|
"loss": 4.8891,
|
|
"mean_token_accuracy": 0.20814439207315444,
|
|
"num_tokens": 10853458.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.519427967071533,
|
|
"epoch": 4.843260188087775,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000316628840913463,
|
|
"loss": 4.9756,
|
|
"mean_token_accuracy": 0.19604101330041884,
|
|
"num_tokens": 10862134.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.489586687088012,
|
|
"epoch": 4.84717868338558,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003163332060773738,
|
|
"loss": 4.9987,
|
|
"mean_token_accuracy": 0.19708484411239624,
|
|
"num_tokens": 10872127.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.486190462112427,
|
|
"epoch": 4.851097178683386,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003160374973721434,
|
|
"loss": 4.88,
|
|
"mean_token_accuracy": 0.2007260411977768,
|
|
"num_tokens": 10880617.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.510869312286377,
|
|
"epoch": 4.855015673981192,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003157417153262513,
|
|
"loss": 4.8783,
|
|
"mean_token_accuracy": 0.20762752294540404,
|
|
"num_tokens": 10888769.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.470195150375366,
|
|
"epoch": 4.858934169278997,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031544586046830796,
|
|
"loss": 4.8348,
|
|
"mean_token_accuracy": 0.2071237787604332,
|
|
"num_tokens": 10897638.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.415426635742188,
|
|
"epoch": 4.862852664576803,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000315149933327054,
|
|
"loss": 4.7999,
|
|
"mean_token_accuracy": 0.20946355313062667,
|
|
"num_tokens": 10906224.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.352281856536865,
|
|
"epoch": 4.866771159874608,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00031485393443135914,
|
|
"loss": 4.7908,
|
|
"mean_token_accuracy": 0.21291908472776414,
|
|
"num_tokens": 10915390.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.48149471282959,
|
|
"epoch": 4.870689655172414,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003145578643102214,
|
|
"loss": 4.8649,
|
|
"mean_token_accuracy": 0.20878373831510544,
|
|
"num_tokens": 10923719.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.429383850097656,
|
|
"epoch": 4.87460815047022,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003142617234927662,
|
|
"loss": 4.8268,
|
|
"mean_token_accuracy": 0.2096204310655594,
|
|
"num_tokens": 10932610.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.4389684200286865,
|
|
"epoch": 4.878526645768025,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00031396551250824513,
|
|
"loss": 4.8227,
|
|
"mean_token_accuracy": 0.21061211079359055,
|
|
"num_tokens": 10941315.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.322816991806031,
|
|
"epoch": 4.882445141065831,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003136692318860352,
|
|
"loss": 4.7816,
|
|
"mean_token_accuracy": 0.21058593392372132,
|
|
"num_tokens": 10949781.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.520660781860352,
|
|
"epoch": 4.886363636363637,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003133728821556381,
|
|
"loss": 4.9815,
|
|
"mean_token_accuracy": 0.20111907571554183,
|
|
"num_tokens": 10958207.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.465389823913574,
|
|
"epoch": 4.890282131661442,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00031307646384667854,
|
|
"loss": 4.866,
|
|
"mean_token_accuracy": 0.1984990805387497,
|
|
"num_tokens": 10967162.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.487727355957031,
|
|
"epoch": 4.894200626959248,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00031277997748890433,
|
|
"loss": 4.9387,
|
|
"mean_token_accuracy": 0.20355436354875564,
|
|
"num_tokens": 10975517.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.455415725708008,
|
|
"epoch": 4.898119122257054,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003124834236121847,
|
|
"loss": 4.8262,
|
|
"mean_token_accuracy": 0.2067200556397438,
|
|
"num_tokens": 10984668.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.392430830001831,
|
|
"epoch": 4.902037617554859,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031218680274650934,
|
|
"loss": 4.8232,
|
|
"mean_token_accuracy": 0.19985046982765198,
|
|
"num_tokens": 10993475.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.377801609039307,
|
|
"epoch": 4.905956112852665,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00031189011542198794,
|
|
"loss": 4.78,
|
|
"mean_token_accuracy": 0.219709874689579,
|
|
"num_tokens": 11002419.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.381848382949829,
|
|
"epoch": 4.90987460815047,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 4.848,
|
|
"mean_token_accuracy": 0.21038996130228044,
|
|
"num_tokens": 11011271.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.517763090133667,
|
|
"epoch": 4.913793103448276,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031129654351743816,
|
|
"loss": 4.8832,
|
|
"mean_token_accuracy": 0.20463715642690658,
|
|
"num_tokens": 11020461.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.496543455123901,
|
|
"epoch": 4.917711598746082,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031099965999821906,
|
|
"loss": 4.8086,
|
|
"mean_token_accuracy": 0.21447691768407823,
|
|
"num_tokens": 11029251.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.456275081634521,
|
|
"epoch": 4.921630094043887,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003107027121417704,
|
|
"loss": 4.8158,
|
|
"mean_token_accuracy": 0.21184113770723342,
|
|
"num_tokens": 11038149.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.512465190887451,
|
|
"epoch": 4.925548589341693,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00031040570047878634,
|
|
"loss": 4.893,
|
|
"mean_token_accuracy": 0.20563915967941285,
|
|
"num_tokens": 11046262.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.492912101745605,
|
|
"epoch": 4.929467084639499,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031010862554007473,
|
|
"loss": 4.9631,
|
|
"mean_token_accuracy": 0.1953013837337494,
|
|
"num_tokens": 11055755.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.413470458984375,
|
|
"epoch": 4.933385579937304,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003098114878565567,
|
|
"loss": 4.8453,
|
|
"mean_token_accuracy": 0.1995655506849289,
|
|
"num_tokens": 11064534.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.443273067474365,
|
|
"epoch": 4.93730407523511,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030951428795926536,
|
|
"loss": 4.8923,
|
|
"mean_token_accuracy": 0.207352314889431,
|
|
"num_tokens": 11073768.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.451230001449585,
|
|
"epoch": 4.941222570532915,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003092170263793452,
|
|
"loss": 4.8624,
|
|
"mean_token_accuracy": 0.2114374428987503,
|
|
"num_tokens": 11083683.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.518076658248901,
|
|
"epoch": 4.945141065830721,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030891970364805096,
|
|
"loss": 4.9146,
|
|
"mean_token_accuracy": 0.20530039370059966,
|
|
"num_tokens": 11092368.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.392894506454468,
|
|
"epoch": 4.949059561128527,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003086223202967464,
|
|
"loss": 4.8377,
|
|
"mean_token_accuracy": 0.20929081737995148,
|
|
"num_tokens": 11100544.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.438666820526123,
|
|
"epoch": 4.952978056426332,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000308324876856904,
|
|
"loss": 4.877,
|
|
"mean_token_accuracy": 0.20165518969297408,
|
|
"num_tokens": 11108962.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.500327444076538,
|
|
"epoch": 4.956896551724138,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030802737386010314,
|
|
"loss": 4.93,
|
|
"mean_token_accuracy": 0.20328806787729264,
|
|
"num_tokens": 11117011.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.45417046546936,
|
|
"epoch": 4.960815047021944,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003077298118380301,
|
|
"loss": 4.8749,
|
|
"mean_token_accuracy": 0.20784866958856582,
|
|
"num_tokens": 11126009.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.44194803237915,
|
|
"epoch": 4.964733542319749,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003074321913224765,
|
|
"loss": 4.773,
|
|
"mean_token_accuracy": 0.2105044424533844,
|
|
"num_tokens": 11134404.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.490171718597412,
|
|
"epoch": 4.968652037617555,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003071345128453382,
|
|
"loss": 4.8934,
|
|
"mean_token_accuracy": 0.20293166637420654,
|
|
"num_tokens": 11142245.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.390559101104737,
|
|
"epoch": 4.972570532915361,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003068367769386153,
|
|
"loss": 4.878,
|
|
"mean_token_accuracy": 0.20472471714019774,
|
|
"num_tokens": 11150881.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.3934276580810545,
|
|
"epoch": 4.976489028213166,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00030653898413440973,
|
|
"loss": 4.8195,
|
|
"mean_token_accuracy": 0.20788175016641616,
|
|
"num_tokens": 11159421.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.455231428146362,
|
|
"epoch": 4.980407523510972,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003062411349649258,
|
|
"loss": 4.8845,
|
|
"mean_token_accuracy": 0.20203738659620285,
|
|
"num_tokens": 11168633.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.4118890285491945,
|
|
"epoch": 4.984326018808778,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00030594322996246816,
|
|
"loss": 4.7373,
|
|
"mean_token_accuracy": 0.2162599429488182,
|
|
"num_tokens": 11177435.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.500855731964111,
|
|
"epoch": 4.988244514106583,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030564526965944127,
|
|
"loss": 4.9416,
|
|
"mean_token_accuracy": 0.2018051862716675,
|
|
"num_tokens": 11186195.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.434712886810303,
|
|
"epoch": 4.992163009404389,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003053472545883488,
|
|
"loss": 4.8035,
|
|
"mean_token_accuracy": 0.21392928957939147,
|
|
"num_tokens": 11194087.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.473199653625488,
|
|
"epoch": 4.996081504702194,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00030504918528179164,
|
|
"loss": 4.8851,
|
|
"mean_token_accuracy": 0.20922296047210692,
|
|
"num_tokens": 11202677.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.48469820022583,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00030475106227246824,
|
|
"loss": 4.9119,
|
|
"mean_token_accuracy": 0.19940580129623414,
|
|
"num_tokens": 11211560.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.45655837059021,
|
|
"epoch": 5.003918495297806,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003044528860931726,
|
|
"loss": 4.6726,
|
|
"mean_token_accuracy": 0.23001022040843963,
|
|
"num_tokens": 11220186.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.493988704681397,
|
|
"epoch": 5.007836990595611,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003041546572767939,
|
|
"loss": 4.6815,
|
|
"mean_token_accuracy": 0.21826550364494324,
|
|
"num_tokens": 11229148.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.429565334320069,
|
|
"epoch": 5.011755485893417,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003038563763563156,
|
|
"loss": 4.6013,
|
|
"mean_token_accuracy": 0.22954623997211457,
|
|
"num_tokens": 11237585.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.360487079620361,
|
|
"epoch": 5.015673981191223,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00030355804386481374,
|
|
"loss": 4.5718,
|
|
"mean_token_accuracy": 0.22283912748098372,
|
|
"num_tokens": 11246203.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.430585193634033,
|
|
"epoch": 5.019592476489028,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00030325966033545716,
|
|
"loss": 4.6929,
|
|
"mean_token_accuracy": 0.22394753843545914,
|
|
"num_tokens": 11254999.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.441541624069214,
|
|
"epoch": 5.023510971786834,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00030296122630150534,
|
|
"loss": 4.6442,
|
|
"mean_token_accuracy": 0.22267541587352752,
|
|
"num_tokens": 11264110.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.457357692718506,
|
|
"epoch": 5.027429467084639,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00030266274229630844,
|
|
"loss": 4.587,
|
|
"mean_token_accuracy": 0.2324410229921341,
|
|
"num_tokens": 11273452.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.329038953781128,
|
|
"epoch": 5.031347962382445,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00030236420885330586,
|
|
"loss": 4.5887,
|
|
"mean_token_accuracy": 0.22599131762981414,
|
|
"num_tokens": 11281926.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.416195631027222,
|
|
"epoch": 5.035266457680251,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003020656265060251,
|
|
"loss": 4.645,
|
|
"mean_token_accuracy": 0.2318156436085701,
|
|
"num_tokens": 11290630.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.449637460708618,
|
|
"epoch": 5.039184952978056,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003017669957880813,
|
|
"loss": 4.6671,
|
|
"mean_token_accuracy": 0.22400776594877242,
|
|
"num_tokens": 11299126.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.474372673034668,
|
|
"epoch": 5.043103448275862,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000301468317233176,
|
|
"loss": 4.7405,
|
|
"mean_token_accuracy": 0.21109896302223205,
|
|
"num_tokens": 11308331.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.435002279281616,
|
|
"epoch": 5.047021943573668,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00030116959137509614,
|
|
"loss": 4.6353,
|
|
"mean_token_accuracy": 0.21799202114343644,
|
|
"num_tokens": 11316245.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.483400154113769,
|
|
"epoch": 5.050940438871473,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030087081874771325,
|
|
"loss": 4.6508,
|
|
"mean_token_accuracy": 0.2227489620447159,
|
|
"num_tokens": 11324994.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.361777305603027,
|
|
"epoch": 5.054858934169279,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00030057199988498245,
|
|
"loss": 4.5982,
|
|
"mean_token_accuracy": 0.22669370472431183,
|
|
"num_tokens": 11333255.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.348836469650268,
|
|
"epoch": 5.058777429467085,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003002731353209416,
|
|
"loss": 4.6228,
|
|
"mean_token_accuracy": 0.22205099910497667,
|
|
"num_tokens": 11341857.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.4486305713653564,
|
|
"epoch": 5.06269592476489,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002999742255897102,
|
|
"loss": 4.7353,
|
|
"mean_token_accuracy": 0.21484713554382323,
|
|
"num_tokens": 11350933.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.422053241729737,
|
|
"epoch": 5.066614420062696,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00029967527122548826,
|
|
"loss": 4.6447,
|
|
"mean_token_accuracy": 0.2195791408419609,
|
|
"num_tokens": 11360346.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.4442760944366455,
|
|
"epoch": 5.070532915360501,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002993762727625556,
|
|
"loss": 4.6734,
|
|
"mean_token_accuracy": 0.2204880729317665,
|
|
"num_tokens": 11369401.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.454881048202514,
|
|
"epoch": 5.074451410658307,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002990772307352712,
|
|
"loss": 4.6423,
|
|
"mean_token_accuracy": 0.22213075309991837,
|
|
"num_tokens": 11377882.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.435122394561768,
|
|
"epoch": 5.078369905956113,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002987781456780715,
|
|
"loss": 4.6546,
|
|
"mean_token_accuracy": 0.22697107791900634,
|
|
"num_tokens": 11386934.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.414770841598511,
|
|
"epoch": 5.082288401253918,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029847901812547,
|
|
"loss": 4.6807,
|
|
"mean_token_accuracy": 0.21619892865419388,
|
|
"num_tokens": 11395929.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.408761215209961,
|
|
"epoch": 5.086206896551724,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00029817984861205626,
|
|
"loss": 4.6248,
|
|
"mean_token_accuracy": 0.22671225517988206,
|
|
"num_tokens": 11405664.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.4376952171325685,
|
|
"epoch": 5.09012539184953,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002978806376724945,
|
|
"loss": 4.739,
|
|
"mean_token_accuracy": 0.21877314448356627,
|
|
"num_tokens": 11414667.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.446389818191529,
|
|
"epoch": 5.094043887147335,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 4.6889,
|
|
"mean_token_accuracy": 0.2178817421197891,
|
|
"num_tokens": 11423907.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.094043887147335,
|
|
"eval_entropy": 5.243673949278603,
|
|
"eval_loss": 5.661767482757568,
|
|
"eval_mean_token_accuracy": 0.17161547378167624,
|
|
"eval_num_tokens": 11423907.0,
|
|
"eval_runtime": 2.1533,
|
|
"eval_samples_per_second": 1914.254,
|
|
"eval_steps_per_second": 239.63,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.455598592758179,
|
|
"epoch": 5.097962382445141,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00029728209365395433,
|
|
"loss": 4.734,
|
|
"mean_token_accuracy": 0.2258662074804306,
|
|
"num_tokens": 11433824.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.446443319320679,
|
|
"epoch": 5.101880877742946,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00029698276164467105,
|
|
"loss": 4.5994,
|
|
"mean_token_accuracy": 0.22565912455320358,
|
|
"num_tokens": 11442371.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.450401782989502,
|
|
"epoch": 5.105799373040752,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002966833903486284,
|
|
"loss": 4.6531,
|
|
"mean_token_accuracy": 0.222481369972229,
|
|
"num_tokens": 11451657.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.355702686309814,
|
|
"epoch": 5.109717868338558,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029638398030085134,
|
|
"loss": 4.5778,
|
|
"mean_token_accuracy": 0.22199634462594986,
|
|
"num_tokens": 11460064.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.44189715385437,
|
|
"epoch": 5.113636363636363,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029608453203643434,
|
|
"loss": 4.6829,
|
|
"mean_token_accuracy": 0.2193632736802101,
|
|
"num_tokens": 11469184.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.491392803192139,
|
|
"epoch": 5.117554858934169,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002957850460905398,
|
|
"loss": 4.6131,
|
|
"mean_token_accuracy": 0.23096911907196044,
|
|
"num_tokens": 11478069.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.458559989929199,
|
|
"epoch": 5.121473354231975,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00029548552299839774,
|
|
"loss": 4.6405,
|
|
"mean_token_accuracy": 0.226235269010067,
|
|
"num_tokens": 11486572.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.39644045829773,
|
|
"epoch": 5.12539184952978,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002951859632953046,
|
|
"loss": 4.6908,
|
|
"mean_token_accuracy": 0.22003690004348755,
|
|
"num_tokens": 11495857.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.391125202178955,
|
|
"epoch": 5.129310344827586,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00029488636751662196,
|
|
"loss": 4.6736,
|
|
"mean_token_accuracy": 0.21946213096380235,
|
|
"num_tokens": 11504875.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.42102017402649,
|
|
"epoch": 5.133228840125392,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002945867361977762,
|
|
"loss": 4.6698,
|
|
"mean_token_accuracy": 0.21734658777713775,
|
|
"num_tokens": 11514509.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.424039506912232,
|
|
"epoch": 5.137147335423197,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00029428706987425705,
|
|
"loss": 4.7083,
|
|
"mean_token_accuracy": 0.21342183500528336,
|
|
"num_tokens": 11523046.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.4253593444824215,
|
|
"epoch": 5.141065830721003,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002939873690816167,
|
|
"loss": 4.6114,
|
|
"mean_token_accuracy": 0.22346448004245759,
|
|
"num_tokens": 11531270.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.415647029876709,
|
|
"epoch": 5.144984326018808,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002936876343554692,
|
|
"loss": 4.6701,
|
|
"mean_token_accuracy": 0.2224391683936119,
|
|
"num_tokens": 11539722.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.40508713722229,
|
|
"epoch": 5.148902821316614,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000293387866231489,
|
|
"loss": 4.6948,
|
|
"mean_token_accuracy": 0.2252419576048851,
|
|
"num_tokens": 11549614.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.363164710998535,
|
|
"epoch": 5.15282131661442,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00029308806524541047,
|
|
"loss": 4.552,
|
|
"mean_token_accuracy": 0.22525391280651091,
|
|
"num_tokens": 11557674.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.406286144256592,
|
|
"epoch": 5.156739811912225,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00029278823193302643,
|
|
"loss": 4.607,
|
|
"mean_token_accuracy": 0.22571078389883042,
|
|
"num_tokens": 11566691.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.444485473632812,
|
|
"epoch": 5.160658307210031,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00029248836683018774,
|
|
"loss": 4.6143,
|
|
"mean_token_accuracy": 0.22688743621110916,
|
|
"num_tokens": 11574824.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.343643999099731,
|
|
"epoch": 5.164576802507837,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029218847047280197,
|
|
"loss": 4.6154,
|
|
"mean_token_accuracy": 0.2261706367135048,
|
|
"num_tokens": 11582988.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.411760711669922,
|
|
"epoch": 5.168495297805642,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002918885433968323,
|
|
"loss": 4.6874,
|
|
"mean_token_accuracy": 0.22110223323106765,
|
|
"num_tokens": 11591553.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.356959295272827,
|
|
"epoch": 5.172413793103448,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002915885861382974,
|
|
"loss": 4.6898,
|
|
"mean_token_accuracy": 0.21576202660799026,
|
|
"num_tokens": 11600347.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.44577989578247,
|
|
"epoch": 5.1763322884012535,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029128859923326935,
|
|
"loss": 4.69,
|
|
"mean_token_accuracy": 0.2231975257396698,
|
|
"num_tokens": 11609217.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.4808571338653564,
|
|
"epoch": 5.1802507836990594,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00029098858321787336,
|
|
"loss": 4.6793,
|
|
"mean_token_accuracy": 0.22759685963392257,
|
|
"num_tokens": 11617355.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.421450185775757,
|
|
"epoch": 5.184169278996865,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029068853862828677,
|
|
"loss": 4.6621,
|
|
"mean_token_accuracy": 0.222191621363163,
|
|
"num_tokens": 11626152.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.364876556396484,
|
|
"epoch": 5.1880877742946705,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002903884660007378,
|
|
"loss": 4.6243,
|
|
"mean_token_accuracy": 0.219116447865963,
|
|
"num_tokens": 11634868.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.450121021270752,
|
|
"epoch": 5.1920062695924765,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00029008836587150505,
|
|
"loss": 4.7281,
|
|
"mean_token_accuracy": 0.21937878131866456,
|
|
"num_tokens": 11644199.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.519433879852295,
|
|
"epoch": 5.195924764890282,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002897882387769159,
|
|
"loss": 4.7659,
|
|
"mean_token_accuracy": 0.21460155248641968,
|
|
"num_tokens": 11653227.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.349029302597046,
|
|
"epoch": 5.1998432601880875,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002894880852533464,
|
|
"loss": 4.5654,
|
|
"mean_token_accuracy": 0.23619267642498015,
|
|
"num_tokens": 11661628.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.374560403823852,
|
|
"epoch": 5.2037617554858935,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00028918790583721936,
|
|
"loss": 4.6324,
|
|
"mean_token_accuracy": 0.22359517812728882,
|
|
"num_tokens": 11670616.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.382354593276977,
|
|
"epoch": 5.2076802507836994,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002888877010650041,
|
|
"loss": 4.6277,
|
|
"mean_token_accuracy": 0.23017132431268691,
|
|
"num_tokens": 11679069.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.428422689437866,
|
|
"epoch": 5.2115987460815045,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00028858747147321527,
|
|
"loss": 4.6404,
|
|
"mean_token_accuracy": 0.22740848660469054,
|
|
"num_tokens": 11688184.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.42650957107544,
|
|
"epoch": 5.2155172413793105,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002882872175984118,
|
|
"loss": 4.6839,
|
|
"mean_token_accuracy": 0.22440064698457718,
|
|
"num_tokens": 11696389.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.420708894729614,
|
|
"epoch": 5.219435736677116,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002879869399771962,
|
|
"loss": 4.7058,
|
|
"mean_token_accuracy": 0.21862089782953262,
|
|
"num_tokens": 11704881.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.435741281509399,
|
|
"epoch": 5.2233542319749215,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00028768663914621295,
|
|
"loss": 4.7475,
|
|
"mean_token_accuracy": 0.2187763810157776,
|
|
"num_tokens": 11713566.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.42583475112915,
|
|
"epoch": 5.2272727272727275,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002873863156421486,
|
|
"loss": 4.7003,
|
|
"mean_token_accuracy": 0.22291228771209717,
|
|
"num_tokens": 11722427.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.450865268707275,
|
|
"epoch": 5.231191222570533,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00028708597000172984,
|
|
"loss": 4.6441,
|
|
"mean_token_accuracy": 0.2247304990887642,
|
|
"num_tokens": 11731010.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.37913408279419,
|
|
"epoch": 5.235109717868339,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002867856027617229,
|
|
"loss": 4.7028,
|
|
"mean_token_accuracy": 0.22253734469413758,
|
|
"num_tokens": 11739569.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.393967866897583,
|
|
"epoch": 5.2390282131661445,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00028648521445893304,
|
|
"loss": 4.6453,
|
|
"mean_token_accuracy": 0.2206690341234207,
|
|
"num_tokens": 11747687.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.422339963912964,
|
|
"epoch": 5.24294670846395,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002861848056302026,
|
|
"loss": 4.7105,
|
|
"mean_token_accuracy": 0.22338415533304215,
|
|
"num_tokens": 11756713.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.4141088962554935,
|
|
"epoch": 5.246865203761756,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028588437681241106,
|
|
"loss": 4.6257,
|
|
"mean_token_accuracy": 0.2249574065208435,
|
|
"num_tokens": 11765625.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.449790954589844,
|
|
"epoch": 5.250783699059561,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028558392854247333,
|
|
"loss": 4.7221,
|
|
"mean_token_accuracy": 0.21989889591932296,
|
|
"num_tokens": 11774594.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.390649938583374,
|
|
"epoch": 5.254702194357367,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002852834613573391,
|
|
"loss": 4.6667,
|
|
"mean_token_accuracy": 0.2323150709271431,
|
|
"num_tokens": 11782983.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.351734399795532,
|
|
"epoch": 5.258620689655173,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002849829757939922,
|
|
"loss": 4.6438,
|
|
"mean_token_accuracy": 0.21888629645109176,
|
|
"num_tokens": 11791940.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.418111753463745,
|
|
"epoch": 5.262539184952978,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002846824723894487,
|
|
"loss": 4.6562,
|
|
"mean_token_accuracy": 0.2276826947927475,
|
|
"num_tokens": 11800357.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.4512347221374515,
|
|
"epoch": 5.266457680250784,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002843819516807572,
|
|
"loss": 4.7226,
|
|
"mean_token_accuracy": 0.2177785024046898,
|
|
"num_tokens": 11809005.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.465693235397339,
|
|
"epoch": 5.27037617554859,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00028408141420499685,
|
|
"loss": 4.6796,
|
|
"mean_token_accuracy": 0.21739601492881774,
|
|
"num_tokens": 11818021.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.42656569480896,
|
|
"epoch": 5.274294670846395,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002837808604992768,
|
|
"loss": 4.7002,
|
|
"mean_token_accuracy": 0.2164924845099449,
|
|
"num_tokens": 11826791.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.354179716110229,
|
|
"epoch": 5.278213166144201,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 4.5892,
|
|
"mean_token_accuracy": 0.23143772333860396,
|
|
"num_tokens": 11835110.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.382524824142456,
|
|
"epoch": 5.282131661442007,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002831797065465386,
|
|
"loss": 4.7329,
|
|
"mean_token_accuracy": 0.2153598517179489,
|
|
"num_tokens": 11842658.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.346162796020508,
|
|
"epoch": 5.286050156739812,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028287910737388,
|
|
"loss": 4.6862,
|
|
"mean_token_accuracy": 0.22583182156085968,
|
|
"num_tokens": 11851028.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.422614669799804,
|
|
"epoch": 5.289968652037618,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000282578494119979,
|
|
"loss": 4.6511,
|
|
"mean_token_accuracy": 0.2277819663286209,
|
|
"num_tokens": 11859471.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.38887734413147,
|
|
"epoch": 5.293887147335424,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00028227786732208014,
|
|
"loss": 4.5943,
|
|
"mean_token_accuracy": 0.22692828476428986,
|
|
"num_tokens": 11868625.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.385702466964721,
|
|
"epoch": 5.297805642633229,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00028197722751745246,
|
|
"loss": 4.6375,
|
|
"mean_token_accuracy": 0.22572360783815384,
|
|
"num_tokens": 11877744.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.383715009689331,
|
|
"epoch": 5.301724137931035,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002816765752433879,
|
|
"loss": 4.6413,
|
|
"mean_token_accuracy": 0.23066455125808716,
|
|
"num_tokens": 11886445.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.2896452903747555,
|
|
"epoch": 5.30564263322884,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00028137591103720075,
|
|
"loss": 4.6389,
|
|
"mean_token_accuracy": 0.22339109033346177,
|
|
"num_tokens": 11896040.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.361033153533936,
|
|
"epoch": 5.309561128526646,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002810752354362267,
|
|
"loss": 4.664,
|
|
"mean_token_accuracy": 0.214099483191967,
|
|
"num_tokens": 11904996.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.455094623565674,
|
|
"epoch": 5.313479623824452,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000280774548977822,
|
|
"loss": 4.6746,
|
|
"mean_token_accuracy": 0.22767408639192582,
|
|
"num_tokens": 11913357.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.398895359039306,
|
|
"epoch": 5.317398119122257,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028047385219936196,
|
|
"loss": 4.7149,
|
|
"mean_token_accuracy": 0.22330257743597032,
|
|
"num_tokens": 11922196.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.321312808990479,
|
|
"epoch": 5.321316614420063,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028017314563824044,
|
|
"loss": 4.581,
|
|
"mean_token_accuracy": 0.23016263097524642,
|
|
"num_tokens": 11930682.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.359769201278686,
|
|
"epoch": 5.325235109717869,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00027987242983186896,
|
|
"loss": 4.6791,
|
|
"mean_token_accuracy": 0.2148265242576599,
|
|
"num_tokens": 11939470.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.419424772262573,
|
|
"epoch": 5.329153605015674,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00027957170531767525,
|
|
"loss": 4.6737,
|
|
"mean_token_accuracy": 0.22054995596408844,
|
|
"num_tokens": 11948183.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.4517310619354244,
|
|
"epoch": 5.33307210031348,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027927097263310294,
|
|
"loss": 4.6869,
|
|
"mean_token_accuracy": 0.21976241022348403,
|
|
"num_tokens": 11957027.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.372526168823242,
|
|
"epoch": 5.336990595611285,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002789702323156099,
|
|
"loss": 4.6458,
|
|
"mean_token_accuracy": 0.22326040714979173,
|
|
"num_tokens": 11965871.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.453324174880981,
|
|
"epoch": 5.340909090909091,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027866948490266815,
|
|
"loss": 4.7595,
|
|
"mean_token_accuracy": 0.21311501413583755,
|
|
"num_tokens": 11974384.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.427382707595825,
|
|
"epoch": 5.344827586206897,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002783687309317618,
|
|
"loss": 4.7065,
|
|
"mean_token_accuracy": 0.2173201248049736,
|
|
"num_tokens": 11983892.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.505741310119629,
|
|
"epoch": 5.348746081504702,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002780679709403871,
|
|
"loss": 4.722,
|
|
"mean_token_accuracy": 0.21889595985412597,
|
|
"num_tokens": 11992990.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.356076192855835,
|
|
"epoch": 5.352664576802508,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00027776720546605086,
|
|
"loss": 4.5323,
|
|
"mean_token_accuracy": 0.23686861544847487,
|
|
"num_tokens": 12001679.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.3767256259918215,
|
|
"epoch": 5.356583072100314,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002774664350462697,
|
|
"loss": 4.7019,
|
|
"mean_token_accuracy": 0.22255587428808213,
|
|
"num_tokens": 12011300.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.326117753982544,
|
|
"epoch": 5.360501567398119,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027716566021856933,
|
|
"loss": 4.668,
|
|
"mean_token_accuracy": 0.22509484142065048,
|
|
"num_tokens": 12019776.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.360721921920776,
|
|
"epoch": 5.364420062695925,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000276864881520483,
|
|
"loss": 4.6986,
|
|
"mean_token_accuracy": 0.21908129304647445,
|
|
"num_tokens": 12028838.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.465584230422974,
|
|
"epoch": 5.368338557993731,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002765640994895509,
|
|
"loss": 4.6552,
|
|
"mean_token_accuracy": 0.22322648912668228,
|
|
"num_tokens": 12036903.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.427506589889527,
|
|
"epoch": 5.372257053291536,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027626331466331955,
|
|
"loss": 4.6482,
|
|
"mean_token_accuracy": 0.21786865144968032,
|
|
"num_tokens": 12045838.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.472641372680664,
|
|
"epoch": 5.376175548589342,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027596252757933995,
|
|
"loss": 4.7954,
|
|
"mean_token_accuracy": 0.21363594383001328,
|
|
"num_tokens": 12055791.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.374497556686402,
|
|
"epoch": 5.380094043887147,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002756617387751674,
|
|
"loss": 4.5922,
|
|
"mean_token_accuracy": 0.23031418174505233,
|
|
"num_tokens": 12064842.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.37699556350708,
|
|
"epoch": 5.384012539184953,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002753609487883606,
|
|
"loss": 4.6618,
|
|
"mean_token_accuracy": 0.2239861860871315,
|
|
"num_tokens": 12073487.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.320009183883667,
|
|
"epoch": 5.387931034482759,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027506015815647965,
|
|
"loss": 4.6233,
|
|
"mean_token_accuracy": 0.2217847689986229,
|
|
"num_tokens": 12082000.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.395949220657348,
|
|
"epoch": 5.391849529780564,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027475936741708636,
|
|
"loss": 4.6798,
|
|
"mean_token_accuracy": 0.22249826043844223,
|
|
"num_tokens": 12090347.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.4145745754241945,
|
|
"epoch": 5.39576802507837,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002744585771077425,
|
|
"loss": 4.6405,
|
|
"mean_token_accuracy": 0.22382888048887253,
|
|
"num_tokens": 12098340.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.3864531993865965,
|
|
"epoch": 5.399686520376176,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027415778776600913,
|
|
"loss": 4.683,
|
|
"mean_token_accuracy": 0.22752432525157928,
|
|
"num_tokens": 12106771.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.42605996131897,
|
|
"epoch": 5.403605015673981,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002738569999294456,
|
|
"loss": 4.6885,
|
|
"mean_token_accuracy": 0.21987025886774064,
|
|
"num_tokens": 12115436.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.381915760040283,
|
|
"epoch": 5.407523510971787,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002735562141356085,
|
|
"loss": 4.5856,
|
|
"mean_token_accuracy": 0.22474498599767684,
|
|
"num_tokens": 12123939.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.499166631698609,
|
|
"epoch": 5.411442006269592,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002732554309220509,
|
|
"loss": 4.81,
|
|
"mean_token_accuracy": 0.219595006108284,
|
|
"num_tokens": 12132803.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.3465766429901125,
|
|
"epoch": 5.415360501567398,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002729546508263211,
|
|
"loss": 4.7077,
|
|
"mean_token_accuracy": 0.22283690124750138,
|
|
"num_tokens": 12141875.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.426230716705322,
|
|
"epoch": 5.419278996865204,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002726538743859618,
|
|
"loss": 4.6721,
|
|
"mean_token_accuracy": 0.22482959926128387,
|
|
"num_tokens": 12150685.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.39882116317749,
|
|
"epoch": 5.423197492163009,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002723531021385095,
|
|
"loss": 4.6351,
|
|
"mean_token_accuracy": 0.2248759776353836,
|
|
"num_tokens": 12159070.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.3739077091217045,
|
|
"epoch": 5.427115987460815,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002720523346214928,
|
|
"loss": 4.6549,
|
|
"mean_token_accuracy": 0.22238496243953704,
|
|
"num_tokens": 12167676.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.382834339141846,
|
|
"epoch": 5.431034482758621,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00027175157237243204,
|
|
"loss": 4.6715,
|
|
"mean_token_accuracy": 0.21960637122392654,
|
|
"num_tokens": 12176222.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.427330017089844,
|
|
"epoch": 5.434952978056426,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002714508159288382,
|
|
"loss": 4.6603,
|
|
"mean_token_accuracy": 0.22749463021755217,
|
|
"num_tokens": 12185171.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.465666437149048,
|
|
"epoch": 5.438871473354232,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002711500658282118,
|
|
"loss": 4.7223,
|
|
"mean_token_accuracy": 0.22708866894245147,
|
|
"num_tokens": 12193875.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.3069953441619875,
|
|
"epoch": 5.442789968652038,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00027084932260804193,
|
|
"loss": 4.6323,
|
|
"mean_token_accuracy": 0.22376984357833862,
|
|
"num_tokens": 12202681.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.374058628082276,
|
|
"epoch": 5.446708463949843,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002705485868058056,
|
|
"loss": 4.6431,
|
|
"mean_token_accuracy": 0.23054510056972505,
|
|
"num_tokens": 12210854.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.4512653827667235,
|
|
"epoch": 5.450626959247649,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00027024785895896644,
|
|
"loss": 4.684,
|
|
"mean_token_accuracy": 0.22167194932699202,
|
|
"num_tokens": 12219643.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.410632228851318,
|
|
"epoch": 5.454545454545454,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026994713960497383,
|
|
"loss": 4.6962,
|
|
"mean_token_accuracy": 0.21880388110876084,
|
|
"num_tokens": 12228767.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.398604440689087,
|
|
"epoch": 5.45846394984326,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026964642928126197,
|
|
"loss": 4.7406,
|
|
"mean_token_accuracy": 0.22047108560800552,
|
|
"num_tokens": 12237468.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.334792757034302,
|
|
"epoch": 5.462382445141066,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 4.7092,
|
|
"mean_token_accuracy": 0.22184640020132065,
|
|
"num_tokens": 12246040.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.434741115570068,
|
|
"epoch": 5.466300940438871,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026904503787433614,
|
|
"loss": 4.7926,
|
|
"mean_token_accuracy": 0.2124181091785431,
|
|
"num_tokens": 12254646.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.4346527576446535,
|
|
"epoch": 5.470219435736677,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000268744357865906,
|
|
"loss": 4.6713,
|
|
"mean_token_accuracy": 0.2237667113542557,
|
|
"num_tokens": 12263306.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.389755058288574,
|
|
"epoch": 5.474137931034483,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026844368903732263,
|
|
"loss": 4.6814,
|
|
"mean_token_accuracy": 0.22320443987846375,
|
|
"num_tokens": 12272498.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.313151454925537,
|
|
"epoch": 5.478056426332288,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026814303192593015,
|
|
"loss": 4.5579,
|
|
"mean_token_accuracy": 0.23305868804454805,
|
|
"num_tokens": 12281592.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.3931866645812985,
|
|
"epoch": 5.481974921630094,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026784238706905136,
|
|
"loss": 4.6879,
|
|
"mean_token_accuracy": 0.22350150495767593,
|
|
"num_tokens": 12290112.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.44029598236084,
|
|
"epoch": 5.485893416927899,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002675417550039874,
|
|
"loss": 4.7423,
|
|
"mean_token_accuracy": 0.21732740998268127,
|
|
"num_tokens": 12299868.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 5.485893416927899,
|
|
"eval_entropy": 5.217394616252692,
|
|
"eval_loss": 5.651699542999268,
|
|
"eval_mean_token_accuracy": 0.17197310757036358,
|
|
"eval_num_tokens": 12299868.0,
|
|
"eval_runtime": 2.1644,
|
|
"eval_samples_per_second": 1904.437,
|
|
"eval_steps_per_second": 238.401,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.410634565353393,
|
|
"epoch": 5.489811912225705,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026724113626801655,
|
|
"loss": 4.657,
|
|
"mean_token_accuracy": 0.22261969298124312,
|
|
"num_tokens": 12307903.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.487033557891846,
|
|
"epoch": 5.493730407523511,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002669405313983932,
|
|
"loss": 4.7747,
|
|
"mean_token_accuracy": 0.21729949712753296,
|
|
"num_tokens": 12317287.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.520665216445923,
|
|
"epoch": 5.497648902821316,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000266639940932347,
|
|
"loss": 4.7668,
|
|
"mean_token_accuracy": 0.22022165060043336,
|
|
"num_tokens": 12325770.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.4067230224609375,
|
|
"epoch": 5.501567398119122,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002663393654070819,
|
|
"loss": 4.6905,
|
|
"mean_token_accuracy": 0.2154438391327858,
|
|
"num_tokens": 12334800.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.443048667907715,
|
|
"epoch": 5.505485893416928,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002660388053597752,
|
|
"loss": 4.7056,
|
|
"mean_token_accuracy": 0.21614101082086562,
|
|
"num_tokens": 12343362.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.414756965637207,
|
|
"epoch": 5.509404388714733,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026573826132757623,
|
|
"loss": 4.7105,
|
|
"mean_token_accuracy": 0.21949589848518372,
|
|
"num_tokens": 12352277.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.3577831268310545,
|
|
"epoch": 5.513322884012539,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026543773384760593,
|
|
"loss": 4.5689,
|
|
"mean_token_accuracy": 0.23594664186239242,
|
|
"num_tokens": 12361244.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.402964448928833,
|
|
"epoch": 5.517241379310345,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002651372234569557,
|
|
"loss": 4.657,
|
|
"mean_token_accuracy": 0.21992406249046326,
|
|
"num_tokens": 12370228.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.395831632614136,
|
|
"epoch": 5.52115987460815,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002648367306926864,
|
|
"loss": 4.6667,
|
|
"mean_token_accuracy": 0.22161114513874053,
|
|
"num_tokens": 12379392.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.368933296203613,
|
|
"epoch": 5.525078369905956,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002645362560918271,
|
|
"loss": 4.7129,
|
|
"mean_token_accuracy": 0.22444327771663666,
|
|
"num_tokens": 12387334.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.360473871231079,
|
|
"epoch": 5.528996865203762,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000264235800191375,
|
|
"loss": 4.6238,
|
|
"mean_token_accuracy": 0.22640235126018524,
|
|
"num_tokens": 12396875.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.449250411987305,
|
|
"epoch": 5.532915360501567,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002639353635282932,
|
|
"loss": 4.654,
|
|
"mean_token_accuracy": 0.23056591302156448,
|
|
"num_tokens": 12405123.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.505821561813354,
|
|
"epoch": 5.536833855799373,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026363494663951076,
|
|
"loss": 4.7654,
|
|
"mean_token_accuracy": 0.21673670560121536,
|
|
"num_tokens": 12413208.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.413863658905029,
|
|
"epoch": 5.540752351097178,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026333455006192146,
|
|
"loss": 4.7627,
|
|
"mean_token_accuracy": 0.21454296559095382,
|
|
"num_tokens": 12421704.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.416645145416259,
|
|
"epoch": 5.544670846394984,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026303417433238267,
|
|
"loss": 4.7147,
|
|
"mean_token_accuracy": 0.21767688989639283,
|
|
"num_tokens": 12430610.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.397741031646729,
|
|
"epoch": 5.54858934169279,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002627338199877146,
|
|
"loss": 4.6126,
|
|
"mean_token_accuracy": 0.2279224395751953,
|
|
"num_tokens": 12439867.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.400185012817383,
|
|
"epoch": 5.552507836990595,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002624334875646989,
|
|
"loss": 4.6835,
|
|
"mean_token_accuracy": 0.22191415280103682,
|
|
"num_tokens": 12449540.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.381630373001099,
|
|
"epoch": 5.556426332288401,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026213317760007864,
|
|
"loss": 4.7072,
|
|
"mean_token_accuracy": 0.22684354335069656,
|
|
"num_tokens": 12458138.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.456050395965576,
|
|
"epoch": 5.560344827586206,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002618328906305562,
|
|
"loss": 4.7942,
|
|
"mean_token_accuracy": 0.21496521383523942,
|
|
"num_tokens": 12467921.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.473070192337036,
|
|
"epoch": 5.564263322884012,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002615326271927931,
|
|
"loss": 4.6706,
|
|
"mean_token_accuracy": 0.21914683580398558,
|
|
"num_tokens": 12476873.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.46126766204834,
|
|
"epoch": 5.568181818181818,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026123238782340893,
|
|
"loss": 4.6774,
|
|
"mean_token_accuracy": 0.2224356397986412,
|
|
"num_tokens": 12485924.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.41128945350647,
|
|
"epoch": 5.572100313479623,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026093217305898003,
|
|
"loss": 4.6304,
|
|
"mean_token_accuracy": 0.22486422210931778,
|
|
"num_tokens": 12494533.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.3274911403656,
|
|
"epoch": 5.576018808777429,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026063198343603887,
|
|
"loss": 4.5774,
|
|
"mean_token_accuracy": 0.2299424320459366,
|
|
"num_tokens": 12503137.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.349985837936401,
|
|
"epoch": 5.579937304075235,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00026033181949107286,
|
|
"loss": 4.696,
|
|
"mean_token_accuracy": 0.21918326765298843,
|
|
"num_tokens": 12511701.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.33917989730835,
|
|
"epoch": 5.58385579937304,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002600316817605239,
|
|
"loss": 4.6917,
|
|
"mean_token_accuracy": 0.23009887486696243,
|
|
"num_tokens": 12520097.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.402477073669433,
|
|
"epoch": 5.587774294670846,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002597315707807866,
|
|
"loss": 4.6838,
|
|
"mean_token_accuracy": 0.22503326833248138,
|
|
"num_tokens": 12528526.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.433033180236817,
|
|
"epoch": 5.591692789968652,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025943148708820784,
|
|
"loss": 4.701,
|
|
"mean_token_accuracy": 0.22399130314588547,
|
|
"num_tokens": 12537888.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.400707530975342,
|
|
"epoch": 5.5956112852664575,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002591314312190861,
|
|
"loss": 4.6919,
|
|
"mean_token_accuracy": 0.21885019838809966,
|
|
"num_tokens": 12546396.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.3567023277282715,
|
|
"epoch": 5.599529780564263,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025883140370966975,
|
|
"loss": 4.6357,
|
|
"mean_token_accuracy": 0.22659466415643692,
|
|
"num_tokens": 12554985.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.41511549949646,
|
|
"epoch": 5.603448275862069,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002585314050961565,
|
|
"loss": 4.6383,
|
|
"mean_token_accuracy": 0.2370347335934639,
|
|
"num_tokens": 12564379.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.385776805877685,
|
|
"epoch": 5.6073667711598745,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002582314359146927,
|
|
"loss": 4.6981,
|
|
"mean_token_accuracy": 0.21699567139148712,
|
|
"num_tokens": 12572655.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.393927812576294,
|
|
"epoch": 5.61128526645768,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025793149670137175,
|
|
"loss": 4.7069,
|
|
"mean_token_accuracy": 0.2205183833837509,
|
|
"num_tokens": 12581778.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.329346323013306,
|
|
"epoch": 5.6152037617554855,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002576315879922337,
|
|
"loss": 4.6012,
|
|
"mean_token_accuracy": 0.22954040467739106,
|
|
"num_tokens": 12590324.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.3544281959533695,
|
|
"epoch": 5.6191222570532915,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00025733171032326404,
|
|
"loss": 4.7324,
|
|
"mean_token_accuracy": 0.2104465678334236,
|
|
"num_tokens": 12599761.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.441565942764282,
|
|
"epoch": 5.6230407523510975,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002570318642303929,
|
|
"loss": 4.7393,
|
|
"mean_token_accuracy": 0.21390973031520844,
|
|
"num_tokens": 12608853.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.429282712936401,
|
|
"epoch": 5.6269592476489025,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025673205024949377,
|
|
"loss": 4.6358,
|
|
"mean_token_accuracy": 0.22299316227436067,
|
|
"num_tokens": 12617434.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.4020544528961185,
|
|
"epoch": 5.6308777429467085,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002564322689163828,
|
|
"loss": 4.6985,
|
|
"mean_token_accuracy": 0.2222519800066948,
|
|
"num_tokens": 12626021.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.375642156600952,
|
|
"epoch": 5.6347962382445145,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025613252076681786,
|
|
"loss": 4.6691,
|
|
"mean_token_accuracy": 0.22293932884931564,
|
|
"num_tokens": 12634952.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.3604011058807375,
|
|
"epoch": 5.63871473354232,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002558328063364976,
|
|
"loss": 4.665,
|
|
"mean_token_accuracy": 0.22402241379022597,
|
|
"num_tokens": 12644094.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.389332485198975,
|
|
"epoch": 5.6426332288401255,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002555331261610603,
|
|
"loss": 4.7019,
|
|
"mean_token_accuracy": 0.21415142267942427,
|
|
"num_tokens": 12652687.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.397766065597534,
|
|
"epoch": 5.646551724137931,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 4.657,
|
|
"mean_token_accuracy": 0.2260015904903412,
|
|
"num_tokens": 12661437.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.4552319049835205,
|
|
"epoch": 5.650470219435737,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00025493387071708024,
|
|
"loss": 4.7142,
|
|
"mean_token_accuracy": 0.22402200549840928,
|
|
"num_tokens": 12669893.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.40191912651062,
|
|
"epoch": 5.6543887147335425,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025463429651950426,
|
|
"loss": 4.7304,
|
|
"mean_token_accuracy": 0.22092269510030746,
|
|
"num_tokens": 12679349.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.4283873558044435,
|
|
"epoch": 5.658307210031348,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002543347587187425,
|
|
"loss": 4.6927,
|
|
"mean_token_accuracy": 0.21239266991615297,
|
|
"num_tokens": 12687163.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.447069454193115,
|
|
"epoch": 5.662225705329154,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00025403525785011767,
|
|
"loss": 4.7002,
|
|
"mean_token_accuracy": 0.21786643713712692,
|
|
"num_tokens": 12695646.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.444200325012207,
|
|
"epoch": 5.66614420062696,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025373579444888634,
|
|
"loss": 4.6514,
|
|
"mean_token_accuracy": 0.23234357088804244,
|
|
"num_tokens": 12705144.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.411109924316406,
|
|
"epoch": 5.670062695924765,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000253436369050238,
|
|
"loss": 4.6709,
|
|
"mean_token_accuracy": 0.21923717856407166,
|
|
"num_tokens": 12713881.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.372307157516479,
|
|
"epoch": 5.673981191222571,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025313698218929474,
|
|
"loss": 4.6303,
|
|
"mean_token_accuracy": 0.22847383916378022,
|
|
"num_tokens": 12722452.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.368383502960205,
|
|
"epoch": 5.677899686520377,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00025283763440110923,
|
|
"loss": 4.642,
|
|
"mean_token_accuracy": 0.227105550467968,
|
|
"num_tokens": 12731105.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.356529235839844,
|
|
"epoch": 5.681818181818182,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002525383262206645,
|
|
"loss": 4.6475,
|
|
"mean_token_accuracy": 0.22905827015638353,
|
|
"num_tokens": 12739096.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.396253395080566,
|
|
"epoch": 5.685736677115988,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002522390581828726,
|
|
"loss": 4.6641,
|
|
"mean_token_accuracy": 0.2209637776017189,
|
|
"num_tokens": 12747910.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.3391711711883545,
|
|
"epoch": 5.689655172413794,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025193983082257444,
|
|
"loss": 4.5775,
|
|
"mean_token_accuracy": 0.23328263610601424,
|
|
"num_tokens": 12757387.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.355838966369629,
|
|
"epoch": 5.693573667711599,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002516406446745376,
|
|
"loss": 4.6894,
|
|
"mean_token_accuracy": 0.22074238806962967,
|
|
"num_tokens": 12766678.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.353663444519043,
|
|
"epoch": 5.697492163009405,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002513415002734562,
|
|
"loss": 4.6444,
|
|
"mean_token_accuracy": 0.2262887701392174,
|
|
"num_tokens": 12775499.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.425950336456299,
|
|
"epoch": 5.70141065830721,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00025104239815394995,
|
|
"loss": 4.7277,
|
|
"mean_token_accuracy": 0.22083000391721724,
|
|
"num_tokens": 12784173.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.436673641204834,
|
|
"epoch": 5.705329153605016,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002507433388505627,
|
|
"loss": 4.752,
|
|
"mean_token_accuracy": 0.2192749172449112,
|
|
"num_tokens": 12792991.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.3657444477081295,
|
|
"epoch": 5.709247648902822,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025044432289776197,
|
|
"loss": 4.6797,
|
|
"mean_token_accuracy": 0.21862591356039046,
|
|
"num_tokens": 12801868.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.451618957519531,
|
|
"epoch": 5.713166144200627,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00025014535082993765,
|
|
"loss": 4.7108,
|
|
"mean_token_accuracy": 0.2286446288228035,
|
|
"num_tokens": 12810467.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.391029405593872,
|
|
"epoch": 5.717084639498433,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002498464231814015,
|
|
"loss": 4.6891,
|
|
"mean_token_accuracy": 0.22305250614881517,
|
|
"num_tokens": 12818874.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.391650819778443,
|
|
"epoch": 5.721003134796238,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00024954754048638555,
|
|
"loss": 4.618,
|
|
"mean_token_accuracy": 0.22212349325418473,
|
|
"num_tokens": 12828029.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.359018135070801,
|
|
"epoch": 5.724921630094044,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024924870327904165,
|
|
"loss": 4.6474,
|
|
"mean_token_accuracy": 0.2227567046880722,
|
|
"num_tokens": 12836185.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.501624011993409,
|
|
"epoch": 5.72884012539185,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002489499120934403,
|
|
"loss": 4.8057,
|
|
"mean_token_accuracy": 0.21358209997415542,
|
|
"num_tokens": 12845810.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.3717132091522215,
|
|
"epoch": 5.732758620689655,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024865116746357004,
|
|
"loss": 4.5681,
|
|
"mean_token_accuracy": 0.23473684638738632,
|
|
"num_tokens": 12854567.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.3523331642150875,
|
|
"epoch": 5.736677115987461,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002483524699233356,
|
|
"loss": 4.6512,
|
|
"mean_token_accuracy": 0.22755155116319656,
|
|
"num_tokens": 12863172.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.376792049407959,
|
|
"epoch": 5.740595611285267,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024805382000655824,
|
|
"loss": 4.7459,
|
|
"mean_token_accuracy": 0.2230014607310295,
|
|
"num_tokens": 12873335.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.473364305496216,
|
|
"epoch": 5.744514106583072,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002477552182469736,
|
|
"loss": 4.7236,
|
|
"mean_token_accuracy": 0.22451044470071793,
|
|
"num_tokens": 12882695.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.373348569869995,
|
|
"epoch": 5.748432601880878,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002474566651782315,
|
|
"loss": 4.6962,
|
|
"mean_token_accuracy": 0.22079328894615174,
|
|
"num_tokens": 12890739.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.3422284603118895,
|
|
"epoch": 5.752351097178684,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002471581613338946,
|
|
"loss": 4.6661,
|
|
"mean_token_accuracy": 0.22051780223846434,
|
|
"num_tokens": 12899503.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.419259548187256,
|
|
"epoch": 5.756269592476489,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002468597072474378,
|
|
"loss": 4.7137,
|
|
"mean_token_accuracy": 0.22393731474876405,
|
|
"num_tokens": 12908405.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.427824592590332,
|
|
"epoch": 5.760188087774295,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024656130345224685,
|
|
"loss": 4.7826,
|
|
"mean_token_accuracy": 0.21181162148714067,
|
|
"num_tokens": 12917529.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.453652238845825,
|
|
"epoch": 5.764106583072101,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024626295048161775,
|
|
"loss": 4.726,
|
|
"mean_token_accuracy": 0.2200993537902832,
|
|
"num_tokens": 12926183.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.5012062072753904,
|
|
"epoch": 5.768025078369906,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002459646488687557,
|
|
"loss": 4.7823,
|
|
"mean_token_accuracy": 0.2160002425312996,
|
|
"num_tokens": 12934598.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.38240761756897,
|
|
"epoch": 5.771943573667712,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002456663991467739,
|
|
"loss": 4.6773,
|
|
"mean_token_accuracy": 0.2164653569459915,
|
|
"num_tokens": 12943101.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.435108089447022,
|
|
"epoch": 5.775862068965517,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024536820184869313,
|
|
"loss": 4.7311,
|
|
"mean_token_accuracy": 0.22121361643075943,
|
|
"num_tokens": 12951717.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.4583597660064695,
|
|
"epoch": 5.779780564263323,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002450700575074401,
|
|
"loss": 4.8032,
|
|
"mean_token_accuracy": 0.214124196767807,
|
|
"num_tokens": 12960761.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.419252252578735,
|
|
"epoch": 5.783699059561129,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024477196665584735,
|
|
"loss": 4.615,
|
|
"mean_token_accuracy": 0.23208827674388885,
|
|
"num_tokens": 12969216.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.440701341629028,
|
|
"epoch": 5.787617554858934,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002444739298266513,
|
|
"loss": 4.6736,
|
|
"mean_token_accuracy": 0.2207224801182747,
|
|
"num_tokens": 12977260.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.36804428100586,
|
|
"epoch": 5.79153605015674,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024417594755249227,
|
|
"loss": 4.6626,
|
|
"mean_token_accuracy": 0.22637103199958802,
|
|
"num_tokens": 12985383.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.411852312088013,
|
|
"epoch": 5.795454545454545,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00024387802036591295,
|
|
"loss": 4.6911,
|
|
"mean_token_accuracy": 0.22195020765066148,
|
|
"num_tokens": 12994675.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.425024747848511,
|
|
"epoch": 5.799373040752351,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002435801487993573,
|
|
"loss": 4.6563,
|
|
"mean_token_accuracy": 0.22789936661720275,
|
|
"num_tokens": 13002979.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.391240644454956,
|
|
"epoch": 5.803291536050157,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00024328233338517024,
|
|
"loss": 4.6615,
|
|
"mean_token_accuracy": 0.22018144577741622,
|
|
"num_tokens": 13011343.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.406405878067017,
|
|
"epoch": 5.807210031347962,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024298457465559626,
|
|
"loss": 4.7829,
|
|
"mean_token_accuracy": 0.2119566962122917,
|
|
"num_tokens": 13019684.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.389628076553345,
|
|
"epoch": 5.811128526645768,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024268687314277844,
|
|
"loss": 4.6648,
|
|
"mean_token_accuracy": 0.22112244218587876,
|
|
"num_tokens": 13028358.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.337295007705689,
|
|
"epoch": 5.815047021943574,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024238922937875758,
|
|
"loss": 4.6317,
|
|
"mean_token_accuracy": 0.22462811768054963,
|
|
"num_tokens": 13037092.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.469033670425415,
|
|
"epoch": 5.818965517241379,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002420916438954714,
|
|
"loss": 4.7861,
|
|
"mean_token_accuracy": 0.21779586225748063,
|
|
"num_tokens": 13045624.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.425481081008911,
|
|
"epoch": 5.822884012539185,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002417941172247535,
|
|
"loss": 4.7062,
|
|
"mean_token_accuracy": 0.21687985062599183,
|
|
"num_tokens": 13054384.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.366958904266357,
|
|
"epoch": 5.826802507836991,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00024149664989833215,
|
|
"loss": 4.7348,
|
|
"mean_token_accuracy": 0.21953156292438508,
|
|
"num_tokens": 13063786.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.450530672073365,
|
|
"epoch": 5.830721003134796,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 4.7337,
|
|
"mean_token_accuracy": 0.2095175191760063,
|
|
"num_tokens": 13073446.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.379944896697998,
|
|
"epoch": 5.834639498432602,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024090189540476153,
|
|
"loss": 4.675,
|
|
"mean_token_accuracy": 0.21573282778263092,
|
|
"num_tokens": 13081780.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.448997449874878,
|
|
"epoch": 5.838557993730408,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024060460930053513,
|
|
"loss": 4.6896,
|
|
"mean_token_accuracy": 0.22558096051216125,
|
|
"num_tokens": 13089995.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.3700416564941404,
|
|
"epoch": 5.842476489028213,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00024030738466644875,
|
|
"loss": 4.6698,
|
|
"mean_token_accuracy": 0.2182238832116127,
|
|
"num_tokens": 13098767.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.370834255218506,
|
|
"epoch": 5.846394984326019,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002400102220336911,
|
|
"loss": 4.7357,
|
|
"mean_token_accuracy": 0.21591129899024963,
|
|
"num_tokens": 13107744.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.369661903381347,
|
|
"epoch": 5.850313479623824,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002397131219333401,
|
|
"loss": 4.6615,
|
|
"mean_token_accuracy": 0.2217957481741905,
|
|
"num_tokens": 13116400.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.344120836257934,
|
|
"epoch": 5.85423197492163,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002394160848963617,
|
|
"loss": 4.6484,
|
|
"mean_token_accuracy": 0.23005529642105102,
|
|
"num_tokens": 13124973.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.459923505783081,
|
|
"epoch": 5.858150470219436,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023911911145360916,
|
|
"loss": 4.7078,
|
|
"mean_token_accuracy": 0.2182393938302994,
|
|
"num_tokens": 13133061.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.468188095092773,
|
|
"epoch": 5.862068965517241,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023882220213582235,
|
|
"loss": 4.7327,
|
|
"mean_token_accuracy": 0.21527129858732225,
|
|
"num_tokens": 13141322.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.425698089599609,
|
|
"epoch": 5.865987460815047,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002385253574736262,
|
|
"loss": 4.7033,
|
|
"mean_token_accuracy": 0.2205759823322296,
|
|
"num_tokens": 13150553.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.484853982925415,
|
|
"epoch": 5.869905956112853,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023822857799753034,
|
|
"loss": 4.8022,
|
|
"mean_token_accuracy": 0.2143026754260063,
|
|
"num_tokens": 13160308.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.46251277923584,
|
|
"epoch": 5.873824451410658,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023793186423792784,
|
|
"loss": 4.7583,
|
|
"mean_token_accuracy": 0.22172479331493378,
|
|
"num_tokens": 13168219.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.458767890930176,
|
|
"epoch": 5.877742946708464,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023763521672509432,
|
|
"loss": 4.7393,
|
|
"mean_token_accuracy": 0.2196473628282547,
|
|
"num_tokens": 13176980.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 5.877742946708464,
|
|
"eval_entropy": 5.253590213235959,
|
|
"eval_loss": 5.632047653198242,
|
|
"eval_mean_token_accuracy": 0.17322684388405593,
|
|
"eval_num_tokens": 13176980.0,
|
|
"eval_runtime": 2.1614,
|
|
"eval_samples_per_second": 1907.106,
|
|
"eval_steps_per_second": 238.735,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.494829559326172,
|
|
"epoch": 5.881661442006269,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023733863598918688,
|
|
"loss": 4.793,
|
|
"mean_token_accuracy": 0.20787972807884217,
|
|
"num_tokens": 13186267.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.466404724121094,
|
|
"epoch": 5.885579937304075,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002370421225602433,
|
|
"loss": 4.7745,
|
|
"mean_token_accuracy": 0.2167092517018318,
|
|
"num_tokens": 13195457.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.464521646499634,
|
|
"epoch": 5.889498432601881,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023674567696818145,
|
|
"loss": 4.7506,
|
|
"mean_token_accuracy": 0.21707697212696075,
|
|
"num_tokens": 13203707.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.422394514083862,
|
|
"epoch": 5.893416927899686,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002364492997427975,
|
|
"loss": 4.6709,
|
|
"mean_token_accuracy": 0.21764766722917556,
|
|
"num_tokens": 13213013.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.383342695236206,
|
|
"epoch": 5.897335423197492,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00023615299141376545,
|
|
"loss": 4.6894,
|
|
"mean_token_accuracy": 0.22216297388076783,
|
|
"num_tokens": 13222356.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.369948482513427,
|
|
"epoch": 5.901253918495298,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023585675251063665,
|
|
"loss": 4.7151,
|
|
"mean_token_accuracy": 0.22025256007909774,
|
|
"num_tokens": 13231791.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.391248369216919,
|
|
"epoch": 5.905172413793103,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002355605835628379,
|
|
"loss": 4.6321,
|
|
"mean_token_accuracy": 0.22624724060297013,
|
|
"num_tokens": 13240445.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.418439054489136,
|
|
"epoch": 5.909090909090909,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023526448509967097,
|
|
"loss": 4.7169,
|
|
"mean_token_accuracy": 0.21420717388391494,
|
|
"num_tokens": 13249048.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.403291940689087,
|
|
"epoch": 5.913009404388715,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023496845765031215,
|
|
"loss": 4.6163,
|
|
"mean_token_accuracy": 0.22270614206790923,
|
|
"num_tokens": 13258206.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.441283273696899,
|
|
"epoch": 5.91692789968652,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023467250174381021,
|
|
"loss": 4.7024,
|
|
"mean_token_accuracy": 0.2197059065103531,
|
|
"num_tokens": 13266756.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.417847537994385,
|
|
"epoch": 5.920846394984326,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00023437661790908648,
|
|
"loss": 4.763,
|
|
"mean_token_accuracy": 0.2140905663371086,
|
|
"num_tokens": 13276033.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.446958684921265,
|
|
"epoch": 5.924764890282132,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023408080667493332,
|
|
"loss": 4.698,
|
|
"mean_token_accuracy": 0.22076408416032792,
|
|
"num_tokens": 13285151.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.378478765487671,
|
|
"epoch": 5.928683385579937,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023378506857001335,
|
|
"loss": 4.6296,
|
|
"mean_token_accuracy": 0.2223990485072136,
|
|
"num_tokens": 13293412.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.428835344314575,
|
|
"epoch": 5.932601880877743,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023348940412285858,
|
|
"loss": 4.7401,
|
|
"mean_token_accuracy": 0.21040181517601014,
|
|
"num_tokens": 13301982.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.471237421035767,
|
|
"epoch": 5.936520376175548,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00023319381386186923,
|
|
"loss": 4.718,
|
|
"mean_token_accuracy": 0.22403409481048583,
|
|
"num_tokens": 13309766.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.369287776947021,
|
|
"epoch": 5.940438871473354,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000232898298315313,
|
|
"loss": 4.6915,
|
|
"mean_token_accuracy": 0.22452086210250854,
|
|
"num_tokens": 13317960.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.465484285354615,
|
|
"epoch": 5.94435736677116,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002326028580113243,
|
|
"loss": 4.751,
|
|
"mean_token_accuracy": 0.21063393950462342,
|
|
"num_tokens": 13326909.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.508973503112793,
|
|
"epoch": 5.948275862068965,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023230749347790265,
|
|
"loss": 4.7689,
|
|
"mean_token_accuracy": 0.21749730557203292,
|
|
"num_tokens": 13335861.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.436627674102783,
|
|
"epoch": 5.952194357366771,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023201220524291232,
|
|
"loss": 4.7392,
|
|
"mean_token_accuracy": 0.21399267613887787,
|
|
"num_tokens": 13345253.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.428041934967041,
|
|
"epoch": 5.956112852664576,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023171699383408152,
|
|
"loss": 4.721,
|
|
"mean_token_accuracy": 0.21830467581748964,
|
|
"num_tokens": 13354353.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.433291816711426,
|
|
"epoch": 5.960031347962382,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023142185977900065,
|
|
"loss": 4.7799,
|
|
"mean_token_accuracy": 0.21575697064399718,
|
|
"num_tokens": 13363030.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.436038446426392,
|
|
"epoch": 5.963949843260188,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00023112680360512218,
|
|
"loss": 4.7024,
|
|
"mean_token_accuracy": 0.2279262736439705,
|
|
"num_tokens": 13372797.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.43800721168518,
|
|
"epoch": 5.967868338557993,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023083182583975942,
|
|
"loss": 4.6413,
|
|
"mean_token_accuracy": 0.21934011876583098,
|
|
"num_tokens": 13381425.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.322489213943482,
|
|
"epoch": 5.971786833855799,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002305369270100855,
|
|
"loss": 4.6353,
|
|
"mean_token_accuracy": 0.2309732362627983,
|
|
"num_tokens": 13390246.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.361888885498047,
|
|
"epoch": 5.975705329153605,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002302421076431322,
|
|
"loss": 4.6667,
|
|
"mean_token_accuracy": 0.22746681272983552,
|
|
"num_tokens": 13398851.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.429968309402466,
|
|
"epoch": 5.97962382445141,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00022994736826578976,
|
|
"loss": 4.7347,
|
|
"mean_token_accuracy": 0.21088544875383378,
|
|
"num_tokens": 13407870.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.4338891983032225,
|
|
"epoch": 5.983542319749216,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002296527094048052,
|
|
"loss": 4.7374,
|
|
"mean_token_accuracy": 0.21535477489233018,
|
|
"num_tokens": 13416311.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.448184251785278,
|
|
"epoch": 5.987460815047022,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002293581315867817,
|
|
"loss": 4.7999,
|
|
"mean_token_accuracy": 0.21410685479640962,
|
|
"num_tokens": 13425669.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.3719429016113285,
|
|
"epoch": 5.991379310344827,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002290636353381774,
|
|
"loss": 4.6702,
|
|
"mean_token_accuracy": 0.22326229512691498,
|
|
"num_tokens": 13434448.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.411982250213623,
|
|
"epoch": 5.995297805642633,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022876922118530508,
|
|
"loss": 4.7262,
|
|
"mean_token_accuracy": 0.21898142546415328,
|
|
"num_tokens": 13443712.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.3412220001220705,
|
|
"epoch": 5.999216300940439,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002284748896543305,
|
|
"loss": 4.5299,
|
|
"mean_token_accuracy": 0.23179074227809907,
|
|
"num_tokens": 13452511.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.37401967048645,
|
|
"epoch": 6.003134796238244,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022818064127127163,
|
|
"loss": 4.5172,
|
|
"mean_token_accuracy": 0.23661162257194518,
|
|
"num_tokens": 13460990.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.403070163726807,
|
|
"epoch": 6.00705329153605,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022788647656199834,
|
|
"loss": 4.6441,
|
|
"mean_token_accuracy": 0.23491066992282866,
|
|
"num_tokens": 13470293.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.445391273498535,
|
|
"epoch": 6.0109717868338555,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022759239605223052,
|
|
"loss": 4.5652,
|
|
"mean_token_accuracy": 0.2319922313094139,
|
|
"num_tokens": 13479508.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.376840686798095,
|
|
"epoch": 6.014890282131661,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 4.4694,
|
|
"mean_token_accuracy": 0.23580356687307358,
|
|
"num_tokens": 13488163.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.493274974822998,
|
|
"epoch": 6.018808777429467,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022700448973333808,
|
|
"loss": 4.6347,
|
|
"mean_token_accuracy": 0.23090777397155762,
|
|
"num_tokens": 13497404.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.427306461334228,
|
|
"epoch": 6.0227272727272725,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022671066497489744,
|
|
"loss": 4.5563,
|
|
"mean_token_accuracy": 0.2354293391108513,
|
|
"num_tokens": 13506142.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.383467340469361,
|
|
"epoch": 6.0266457680250785,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002264169265173282,
|
|
"loss": 4.4828,
|
|
"mean_token_accuracy": 0.24172000885009765,
|
|
"num_tokens": 13514525.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.36836347579956,
|
|
"epoch": 6.030564263322884,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022612327488558866,
|
|
"loss": 4.473,
|
|
"mean_token_accuracy": 0.24309679567813874,
|
|
"num_tokens": 13523732.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.322946691513062,
|
|
"epoch": 6.0344827586206895,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022582971060448187,
|
|
"loss": 4.4923,
|
|
"mean_token_accuracy": 0.24305695444345474,
|
|
"num_tokens": 13532904.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.3451848983764645,
|
|
"epoch": 6.0384012539184955,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022553623419865488,
|
|
"loss": 4.5394,
|
|
"mean_token_accuracy": 0.2324364185333252,
|
|
"num_tokens": 13542174.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.387595129013062,
|
|
"epoch": 6.0423197492163006,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022524284619259745,
|
|
"loss": 4.5523,
|
|
"mean_token_accuracy": 0.2323182389140129,
|
|
"num_tokens": 13550591.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.381853008270264,
|
|
"epoch": 6.0462382445141065,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002249495471106417,
|
|
"loss": 4.5246,
|
|
"mean_token_accuracy": 0.23654732555150987,
|
|
"num_tokens": 13559088.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.316314220428467,
|
|
"epoch": 6.0501567398119125,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022465633747696052,
|
|
"loss": 4.4342,
|
|
"mean_token_accuracy": 0.24096310883760452,
|
|
"num_tokens": 13568511.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.41100583076477,
|
|
"epoch": 6.054075235109718,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022436321781556712,
|
|
"loss": 4.6276,
|
|
"mean_token_accuracy": 0.22436762750148773,
|
|
"num_tokens": 13577750.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.384617471694947,
|
|
"epoch": 6.0579937304075235,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002240701886503138,
|
|
"loss": 4.5467,
|
|
"mean_token_accuracy": 0.23236613422632219,
|
|
"num_tokens": 13587138.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.3994673728942875,
|
|
"epoch": 6.0619122257053295,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022377725050489128,
|
|
"loss": 4.5651,
|
|
"mean_token_accuracy": 0.23556134849786758,
|
|
"num_tokens": 13595459.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.452786588668824,
|
|
"epoch": 6.065830721003135,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022348440390282753,
|
|
"loss": 4.6371,
|
|
"mean_token_accuracy": 0.2331371158361435,
|
|
"num_tokens": 13605227.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.440693426132202,
|
|
"epoch": 6.0697492163009406,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00022319164936748676,
|
|
"loss": 4.6165,
|
|
"mean_token_accuracy": 0.23309418708086013,
|
|
"num_tokens": 13614139.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.417187738418579,
|
|
"epoch": 6.0736677115987465,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022289898742206904,
|
|
"loss": 4.5282,
|
|
"mean_token_accuracy": 0.23577687591314317,
|
|
"num_tokens": 13623420.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.444424819946289,
|
|
"epoch": 6.077586206896552,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002226064185896085,
|
|
"loss": 4.5893,
|
|
"mean_token_accuracy": 0.23402267545461655,
|
|
"num_tokens": 13632655.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.416470575332641,
|
|
"epoch": 6.081504702194358,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022231394339297322,
|
|
"loss": 4.5054,
|
|
"mean_token_accuracy": 0.2384484440088272,
|
|
"num_tokens": 13641782.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.370397233963013,
|
|
"epoch": 6.085423197492163,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00022202156235486358,
|
|
"loss": 4.5017,
|
|
"mean_token_accuracy": 0.23616237044334412,
|
|
"num_tokens": 13650023.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.372197437286377,
|
|
"epoch": 6.089341692789969,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022172927599781224,
|
|
"loss": 4.4979,
|
|
"mean_token_accuracy": 0.23759032040834427,
|
|
"num_tokens": 13659025.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.4308192253112795,
|
|
"epoch": 6.093260188087775,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022143708484418206,
|
|
"loss": 4.5645,
|
|
"mean_token_accuracy": 0.22680082321166992,
|
|
"num_tokens": 13668133.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.41222915649414,
|
|
"epoch": 6.09717868338558,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022114498941616607,
|
|
"loss": 4.5525,
|
|
"mean_token_accuracy": 0.231083844602108,
|
|
"num_tokens": 13677661.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.375866842269898,
|
|
"epoch": 6.101097178683386,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022085299023578614,
|
|
"loss": 4.543,
|
|
"mean_token_accuracy": 0.2378760024905205,
|
|
"num_tokens": 13686117.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.348766756057739,
|
|
"epoch": 6.105015673981192,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022056108782489225,
|
|
"loss": 4.4733,
|
|
"mean_token_accuracy": 0.24007949382066726,
|
|
"num_tokens": 13694792.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.378859806060791,
|
|
"epoch": 6.108934169278997,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022026928270516122,
|
|
"loss": 4.58,
|
|
"mean_token_accuracy": 0.22720586210489274,
|
|
"num_tokens": 13704038.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.409124231338501,
|
|
"epoch": 6.112852664576803,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021997757539809622,
|
|
"loss": 4.5563,
|
|
"mean_token_accuracy": 0.23301581740379335,
|
|
"num_tokens": 13713266.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.358797597885132,
|
|
"epoch": 6.116771159874608,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021968596642502546,
|
|
"loss": 4.492,
|
|
"mean_token_accuracy": 0.24536516815423964,
|
|
"num_tokens": 13721847.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.329495143890381,
|
|
"epoch": 6.120689655172414,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021939445630710138,
|
|
"loss": 4.5148,
|
|
"mean_token_accuracy": 0.23661040216684343,
|
|
"num_tokens": 13730925.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.398707389831543,
|
|
"epoch": 6.12460815047022,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021910304556529986,
|
|
"loss": 4.6446,
|
|
"mean_token_accuracy": 0.2285678654909134,
|
|
"num_tokens": 13739814.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.445919561386108,
|
|
"epoch": 6.128526645768025,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021881173472041926,
|
|
"loss": 4.504,
|
|
"mean_token_accuracy": 0.2447410359978676,
|
|
"num_tokens": 13748859.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.362501192092895,
|
|
"epoch": 6.132445141065831,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021852052429307924,
|
|
"loss": 4.5855,
|
|
"mean_token_accuracy": 0.23866025358438492,
|
|
"num_tokens": 13757249.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.390389060974121,
|
|
"epoch": 6.136363636363637,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00021822941480371995,
|
|
"loss": 4.5568,
|
|
"mean_token_accuracy": 0.2341816246509552,
|
|
"num_tokens": 13765838.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.382616329193115,
|
|
"epoch": 6.140282131661442,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00021793840677260135,
|
|
"loss": 4.5062,
|
|
"mean_token_accuracy": 0.24213809818029403,
|
|
"num_tokens": 13775465.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.492204332351685,
|
|
"epoch": 6.144200626959248,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00021764750071980195,
|
|
"loss": 4.6306,
|
|
"mean_token_accuracy": 0.2255866676568985,
|
|
"num_tokens": 13785373.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.404697608947754,
|
|
"epoch": 6.148119122257054,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00021735669716521806,
|
|
"loss": 4.5563,
|
|
"mean_token_accuracy": 0.24643851220607757,
|
|
"num_tokens": 13794812.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.366805648803711,
|
|
"epoch": 6.152037617554859,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021706599662856265,
|
|
"loss": 4.5311,
|
|
"mean_token_accuracy": 0.23548984229564668,
|
|
"num_tokens": 13803681.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.343835973739624,
|
|
"epoch": 6.155956112852665,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021677539962936494,
|
|
"loss": 4.6266,
|
|
"mean_token_accuracy": 0.2243844285607338,
|
|
"num_tokens": 13812772.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.399751949310303,
|
|
"epoch": 6.15987460815047,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021648490668696875,
|
|
"loss": 4.5399,
|
|
"mean_token_accuracy": 0.23415076583623887,
|
|
"num_tokens": 13821280.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.364692306518554,
|
|
"epoch": 6.163793103448276,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00021619451832053199,
|
|
"loss": 4.5202,
|
|
"mean_token_accuracy": 0.23737911880016327,
|
|
"num_tokens": 13829717.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.425049066543579,
|
|
"epoch": 6.167711598746082,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000215904235049026,
|
|
"loss": 4.6231,
|
|
"mean_token_accuracy": 0.22701039016246796,
|
|
"num_tokens": 13838723.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.363614463806153,
|
|
"epoch": 6.171630094043887,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002156140573912338,
|
|
"loss": 4.6019,
|
|
"mean_token_accuracy": 0.2319874495267868,
|
|
"num_tokens": 13847081.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.403508996963501,
|
|
"epoch": 6.175548589341693,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00021532398586574997,
|
|
"loss": 4.5185,
|
|
"mean_token_accuracy": 0.24495033621788026,
|
|
"num_tokens": 13855142.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.466889953613281,
|
|
"epoch": 6.179467084639499,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021503402099097959,
|
|
"loss": 4.6267,
|
|
"mean_token_accuracy": 0.23401828110218048,
|
|
"num_tokens": 13863586.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.406942129135132,
|
|
"epoch": 6.183385579937304,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00021474416328513664,
|
|
"loss": 4.6192,
|
|
"mean_token_accuracy": 0.23092295974493027,
|
|
"num_tokens": 13872286.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.372610282897949,
|
|
"epoch": 6.18730407523511,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002144544132662438,
|
|
"loss": 4.5515,
|
|
"mean_token_accuracy": 0.2354155510663986,
|
|
"num_tokens": 13880866.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.364731693267823,
|
|
"epoch": 6.191222570532915,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002141647714521316,
|
|
"loss": 4.501,
|
|
"mean_token_accuracy": 0.23398676961660386,
|
|
"num_tokens": 13889015.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.430646705627441,
|
|
"epoch": 6.195141065830721,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00021387523836043676,
|
|
"loss": 4.601,
|
|
"mean_token_accuracy": 0.23500339686870575,
|
|
"num_tokens": 13898457.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.406093788146973,
|
|
"epoch": 6.199059561128527,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 4.5562,
|
|
"mean_token_accuracy": 0.24149138182401658,
|
|
"num_tokens": 13906722.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.368709087371826,
|
|
"epoch": 6.202978056426332,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002132965004138742,
|
|
"loss": 4.4878,
|
|
"mean_token_accuracy": 0.24469409734010697,
|
|
"num_tokens": 13914714.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.380543613433838,
|
|
"epoch": 6.206896551724138,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002130072965933052,
|
|
"loss": 4.5668,
|
|
"mean_token_accuracy": 0.2339995175600052,
|
|
"num_tokens": 13923378.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.329337930679321,
|
|
"epoch": 6.210815047021944,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021271820356374876,
|
|
"loss": 4.5607,
|
|
"mean_token_accuracy": 0.23087091445922853,
|
|
"num_tokens": 13931947.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.361795234680176,
|
|
"epoch": 6.214733542319749,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021242922184186108,
|
|
"loss": 4.473,
|
|
"mean_token_accuracy": 0.23698604702949524,
|
|
"num_tokens": 13940949.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.434546422958374,
|
|
"epoch": 6.218652037617555,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002121403519440993,
|
|
"loss": 4.5948,
|
|
"mean_token_accuracy": 0.23060452789068223,
|
|
"num_tokens": 13949944.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.385509443283081,
|
|
"epoch": 6.222570532915361,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021185159438672085,
|
|
"loss": 4.5299,
|
|
"mean_token_accuracy": 0.23406966626644135,
|
|
"num_tokens": 13958938.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.3564540386199955,
|
|
"epoch": 6.226489028213166,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021156294968578232,
|
|
"loss": 4.4855,
|
|
"mean_token_accuracy": 0.24379798769950867,
|
|
"num_tokens": 13967875.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.32689733505249,
|
|
"epoch": 6.230407523510972,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00021127441835713838,
|
|
"loss": 4.477,
|
|
"mean_token_accuracy": 0.23883217126131057,
|
|
"num_tokens": 13975972.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.331301212310791,
|
|
"epoch": 6.234326018808777,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002109860009164416,
|
|
"loss": 4.4649,
|
|
"mean_token_accuracy": 0.2391668066382408,
|
|
"num_tokens": 13985081.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.3476957321167,
|
|
"epoch": 6.238244514106583,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002106976978791405,
|
|
"loss": 4.5921,
|
|
"mean_token_accuracy": 0.23263794481754302,
|
|
"num_tokens": 13992426.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.336037302017212,
|
|
"epoch": 6.242163009404389,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021040950976047935,
|
|
"loss": 4.5098,
|
|
"mean_token_accuracy": 0.23985662311315536,
|
|
"num_tokens": 14001328.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.33992862701416,
|
|
"epoch": 6.246081504702194,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021012143707549726,
|
|
"loss": 4.4762,
|
|
"mean_token_accuracy": 0.23653718084096909,
|
|
"num_tokens": 14010340.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.435575866699219,
|
|
"epoch": 6.25,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020983348033902662,
|
|
"loss": 4.6711,
|
|
"mean_token_accuracy": 0.22348899990320206,
|
|
"num_tokens": 14019213.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.388225078582764,
|
|
"epoch": 6.253918495297806,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020954564006569294,
|
|
"loss": 4.5963,
|
|
"mean_token_accuracy": 0.22562501281499864,
|
|
"num_tokens": 14028501.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.311678123474121,
|
|
"epoch": 6.257836990595611,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020925791676991318,
|
|
"loss": 4.5063,
|
|
"mean_token_accuracy": 0.23685155361890792,
|
|
"num_tokens": 14037969.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.431684827804565,
|
|
"epoch": 6.261755485893417,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00020897031096589586,
|
|
"loss": 4.5335,
|
|
"mean_token_accuracy": 0.2500784173607826,
|
|
"num_tokens": 14046987.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.395283651351929,
|
|
"epoch": 6.265673981191223,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020868282316763902,
|
|
"loss": 4.5698,
|
|
"mean_token_accuracy": 0.23082169145345688,
|
|
"num_tokens": 14055894.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.289697074890137,
|
|
"epoch": 6.269592476489028,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00020839545388893,
|
|
"loss": 4.5259,
|
|
"mean_token_accuracy": 0.2345459431409836,
|
|
"num_tokens": 14065119.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.269592476489028,
|
|
"eval_entropy": 5.217381423757982,
|
|
"eval_loss": 5.654690742492676,
|
|
"eval_mean_token_accuracy": 0.17240633263952973,
|
|
"eval_num_tokens": 14065119.0,
|
|
"eval_runtime": 2.162,
|
|
"eval_samples_per_second": 1906.547,
|
|
"eval_steps_per_second": 238.665,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.391924428939819,
|
|
"epoch": 6.273510971786834,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020810820364334438,
|
|
"loss": 4.5111,
|
|
"mean_token_accuracy": 0.23921810537576677,
|
|
"num_tokens": 14073320.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.373762083053589,
|
|
"epoch": 6.277429467084639,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020782107294424484,
|
|
"loss": 4.5479,
|
|
"mean_token_accuracy": 0.23705143183469773,
|
|
"num_tokens": 14081730.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.295891904830933,
|
|
"epoch": 6.281347962382445,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002075340623047804,
|
|
"loss": 4.5153,
|
|
"mean_token_accuracy": 0.2405627653002739,
|
|
"num_tokens": 14090738.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.410293817520142,
|
|
"epoch": 6.285266457680251,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00020724717223788574,
|
|
"loss": 4.6081,
|
|
"mean_token_accuracy": 0.23222208619117737,
|
|
"num_tokens": 14099412.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.345663690567017,
|
|
"epoch": 6.289184952978056,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020696040325627991,
|
|
"loss": 4.58,
|
|
"mean_token_accuracy": 0.23623015284538268,
|
|
"num_tokens": 14108334.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.364134359359741,
|
|
"epoch": 6.293103448275862,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020667375587246563,
|
|
"loss": 4.6019,
|
|
"mean_token_accuracy": 0.2357444182038307,
|
|
"num_tokens": 14116853.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.430433940887451,
|
|
"epoch": 6.297021943573668,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020638723059872806,
|
|
"loss": 4.6714,
|
|
"mean_token_accuracy": 0.225820392370224,
|
|
"num_tokens": 14126735.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.410908174514771,
|
|
"epoch": 6.300940438871473,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00020610082794713454,
|
|
"loss": 4.6278,
|
|
"mean_token_accuracy": 0.23538079857826233,
|
|
"num_tokens": 14136749.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.407780408859253,
|
|
"epoch": 6.304858934169279,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002058145484295329,
|
|
"loss": 4.5978,
|
|
"mean_token_accuracy": 0.2309465855360031,
|
|
"num_tokens": 14145204.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.478087759017944,
|
|
"epoch": 6.308777429467085,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00020552839255755107,
|
|
"loss": 4.6552,
|
|
"mean_token_accuracy": 0.22803185135126114,
|
|
"num_tokens": 14154347.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.3730981826782225,
|
|
"epoch": 6.31269592476489,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020524236084259613,
|
|
"loss": 4.5767,
|
|
"mean_token_accuracy": 0.23220402002334595,
|
|
"num_tokens": 14163039.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.334766864776611,
|
|
"epoch": 6.316614420062696,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00020495645379585303,
|
|
"loss": 4.4971,
|
|
"mean_token_accuracy": 0.2393469050526619,
|
|
"num_tokens": 14172228.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.404000234603882,
|
|
"epoch": 6.320532915360501,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020467067192828404,
|
|
"loss": 4.6216,
|
|
"mean_token_accuracy": 0.2343720316886902,
|
|
"num_tokens": 14181514.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.4160051345825195,
|
|
"epoch": 6.324451410658307,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020438501575062752,
|
|
"loss": 4.632,
|
|
"mean_token_accuracy": 0.22792236059904097,
|
|
"num_tokens": 14190654.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.322032499313354,
|
|
"epoch": 6.328369905956113,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020409948577339765,
|
|
"loss": 4.5064,
|
|
"mean_token_accuracy": 0.23621250241994857,
|
|
"num_tokens": 14199177.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.375668907165528,
|
|
"epoch": 6.332288401253918,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020381408250688267,
|
|
"loss": 4.5355,
|
|
"mean_token_accuracy": 0.24090880453586577,
|
|
"num_tokens": 14207643.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.420732402801514,
|
|
"epoch": 6.336206896551724,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002035288064611444,
|
|
"loss": 4.6459,
|
|
"mean_token_accuracy": 0.22801629602909088,
|
|
"num_tokens": 14216884.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.411921548843384,
|
|
"epoch": 6.34012539184953,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002032436581460177,
|
|
"loss": 4.5606,
|
|
"mean_token_accuracy": 0.22774063646793366,
|
|
"num_tokens": 14225324.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.3847418308258055,
|
|
"epoch": 6.344043887147335,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020295863807110853,
|
|
"loss": 4.5061,
|
|
"mean_token_accuracy": 0.23978188037872314,
|
|
"num_tokens": 14233732.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.5077409744262695,
|
|
"epoch": 6.347962382445141,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020267374674579408,
|
|
"loss": 4.6378,
|
|
"mean_token_accuracy": 0.22778765708208085,
|
|
"num_tokens": 14242080.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.359587383270264,
|
|
"epoch": 6.351880877742946,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020238898467922129,
|
|
"loss": 4.6056,
|
|
"mean_token_accuracy": 0.23656764179468154,
|
|
"num_tokens": 14251140.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.403133344650269,
|
|
"epoch": 6.355799373040752,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020210435238030626,
|
|
"loss": 4.6441,
|
|
"mean_token_accuracy": 0.23169869035482407,
|
|
"num_tokens": 14259900.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.392304944992065,
|
|
"epoch": 6.359717868338558,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020181985035773304,
|
|
"loss": 4.5513,
|
|
"mean_token_accuracy": 0.23402901440858842,
|
|
"num_tokens": 14268414.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.387255477905273,
|
|
"epoch": 6.363636363636363,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020153547911995272,
|
|
"loss": 4.618,
|
|
"mean_token_accuracy": 0.22642457187175752,
|
|
"num_tokens": 14277339.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.386812734603882,
|
|
"epoch": 6.367554858934169,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020125123917518308,
|
|
"loss": 4.5451,
|
|
"mean_token_accuracy": 0.23846700191497802,
|
|
"num_tokens": 14285710.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.379901170730591,
|
|
"epoch": 6.371473354231975,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020096713103140673,
|
|
"loss": 4.5595,
|
|
"mean_token_accuracy": 0.23510075807571412,
|
|
"num_tokens": 14294625.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.319660186767578,
|
|
"epoch": 6.37539184952978,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020068315519637103,
|
|
"loss": 4.4253,
|
|
"mean_token_accuracy": 0.24543185383081437,
|
|
"num_tokens": 14303030.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.322494649887085,
|
|
"epoch": 6.379310344827586,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000200399312177587,
|
|
"loss": 4.5157,
|
|
"mean_token_accuracy": 0.23401750326156617,
|
|
"num_tokens": 14311826.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.433436012268066,
|
|
"epoch": 6.383228840125392,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 4.606,
|
|
"mean_token_accuracy": 0.2403976783156395,
|
|
"num_tokens": 14319600.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.3320167541503904,
|
|
"epoch": 6.387147335423197,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001998320266176294,
|
|
"loss": 4.5233,
|
|
"mean_token_accuracy": 0.235674712061882,
|
|
"num_tokens": 14328768.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.375443983078003,
|
|
"epoch": 6.391065830721003,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000199548585090287,
|
|
"loss": 4.6008,
|
|
"mean_token_accuracy": 0.22806597799062728,
|
|
"num_tokens": 14338236.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.4087769985198975,
|
|
"epoch": 6.394984326018808,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000199265278406857,
|
|
"loss": 4.6504,
|
|
"mean_token_accuracy": 0.22760077565908432,
|
|
"num_tokens": 14347190.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.401350498199463,
|
|
"epoch": 6.398902821316614,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00019898210707365428,
|
|
"loss": 4.5958,
|
|
"mean_token_accuracy": 0.23107689321041108,
|
|
"num_tokens": 14355847.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.326382112503052,
|
|
"epoch": 6.40282131661442,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019869907159675187,
|
|
"loss": 4.526,
|
|
"mean_token_accuracy": 0.2316027283668518,
|
|
"num_tokens": 14364280.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.420984172821045,
|
|
"epoch": 6.406739811912225,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019841617248198017,
|
|
"loss": 4.5689,
|
|
"mean_token_accuracy": 0.23186853379011155,
|
|
"num_tokens": 14372857.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.395731496810913,
|
|
"epoch": 6.410658307210031,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001981334102349258,
|
|
"loss": 4.5157,
|
|
"mean_token_accuracy": 0.23702372908592223,
|
|
"num_tokens": 14381157.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.425013637542724,
|
|
"epoch": 6.414576802507837,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019785078536093065,
|
|
"loss": 4.5998,
|
|
"mean_token_accuracy": 0.23190046101808548,
|
|
"num_tokens": 14390404.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.415174770355224,
|
|
"epoch": 6.418495297805642,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019756829836509117,
|
|
"loss": 4.5943,
|
|
"mean_token_accuracy": 0.23080805391073228,
|
|
"num_tokens": 14399426.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.38646092414856,
|
|
"epoch": 6.422413793103448,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001972859497522575,
|
|
"loss": 4.5146,
|
|
"mean_token_accuracy": 0.23038942068815232,
|
|
"num_tokens": 14408233.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.4170982837677,
|
|
"epoch": 6.4263322884012535,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001970037400270323,
|
|
"loss": 4.5663,
|
|
"mean_token_accuracy": 0.2343892365694046,
|
|
"num_tokens": 14416828.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.4153375148773195,
|
|
"epoch": 6.4302507836990594,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001967216696937702,
|
|
"loss": 4.5974,
|
|
"mean_token_accuracy": 0.23196161538362503,
|
|
"num_tokens": 14425590.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.426552057266235,
|
|
"epoch": 6.434169278996865,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019643973925657666,
|
|
"loss": 4.723,
|
|
"mean_token_accuracy": 0.2178611546754837,
|
|
"num_tokens": 14434646.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.472262716293335,
|
|
"epoch": 6.4380877742946705,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019615794921930697,
|
|
"loss": 4.6233,
|
|
"mean_token_accuracy": 0.23203636556863785,
|
|
"num_tokens": 14443862.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.3769574642181395,
|
|
"epoch": 6.4420062695924765,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019587630008556563,
|
|
"loss": 4.5451,
|
|
"mean_token_accuracy": 0.2314043954014778,
|
|
"num_tokens": 14452813.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.393668746948242,
|
|
"epoch": 6.445924764890282,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019559479235870537,
|
|
"loss": 4.6166,
|
|
"mean_token_accuracy": 0.22798816710710526,
|
|
"num_tokens": 14461912.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.416455984115601,
|
|
"epoch": 6.4498432601880875,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001953134265418261,
|
|
"loss": 4.567,
|
|
"mean_token_accuracy": 0.23445055931806563,
|
|
"num_tokens": 14470460.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.400208330154419,
|
|
"epoch": 6.4537617554858935,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001950322031377742,
|
|
"loss": 4.5718,
|
|
"mean_token_accuracy": 0.2362300232052803,
|
|
"num_tokens": 14479258.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.40599946975708,
|
|
"epoch": 6.4576802507836994,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019475112264914142,
|
|
"loss": 4.5737,
|
|
"mean_token_accuracy": 0.23231059014797212,
|
|
"num_tokens": 14488209.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.347379446029663,
|
|
"epoch": 6.4615987460815045,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019447018557826424,
|
|
"loss": 4.5285,
|
|
"mean_token_accuracy": 0.23444238901138306,
|
|
"num_tokens": 14497234.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.308634233474732,
|
|
"epoch": 6.4655172413793105,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001941893924272227,
|
|
"loss": 4.4397,
|
|
"mean_token_accuracy": 0.24518217146396637,
|
|
"num_tokens": 14506374.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.389552068710327,
|
|
"epoch": 6.469435736677116,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019390874369783974,
|
|
"loss": 4.5914,
|
|
"mean_token_accuracy": 0.22903480976819993,
|
|
"num_tokens": 14514796.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.367393541336059,
|
|
"epoch": 6.4733542319749215,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019362823989168006,
|
|
"loss": 4.5726,
|
|
"mean_token_accuracy": 0.23848891407251357,
|
|
"num_tokens": 14523771.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.313951206207276,
|
|
"epoch": 6.4772727272727275,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00019334788151004962,
|
|
"loss": 4.4824,
|
|
"mean_token_accuracy": 0.2445491924881935,
|
|
"num_tokens": 14532130.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.280497217178345,
|
|
"epoch": 6.481191222570533,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019306766905399415,
|
|
"loss": 4.4914,
|
|
"mean_token_accuracy": 0.23300529420375823,
|
|
"num_tokens": 14540278.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.373630666732788,
|
|
"epoch": 6.485109717868339,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019278760302429872,
|
|
"loss": 4.5915,
|
|
"mean_token_accuracy": 0.22372643202543258,
|
|
"num_tokens": 14549519.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.388624477386474,
|
|
"epoch": 6.4890282131661445,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019250768392148682,
|
|
"loss": 4.6286,
|
|
"mean_token_accuracy": 0.2330925866961479,
|
|
"num_tokens": 14558576.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.380887746810913,
|
|
"epoch": 6.49294670846395,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019222791224581925,
|
|
"loss": 4.533,
|
|
"mean_token_accuracy": 0.23420221954584122,
|
|
"num_tokens": 14567544.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.377732610702514,
|
|
"epoch": 6.496865203761756,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00019194828849729336,
|
|
"loss": 4.5426,
|
|
"mean_token_accuracy": 0.22353632003068924,
|
|
"num_tokens": 14576662.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.309522294998169,
|
|
"epoch": 6.500783699059561,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019166881317564217,
|
|
"loss": 4.5241,
|
|
"mean_token_accuracy": 0.23146914392709733,
|
|
"num_tokens": 14585163.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.289928531646728,
|
|
"epoch": 6.504702194357367,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00019138948678033336,
|
|
"loss": 4.4635,
|
|
"mean_token_accuracy": 0.2431421846151352,
|
|
"num_tokens": 14593213.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.355341672897339,
|
|
"epoch": 6.508620689655173,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019111030981056832,
|
|
"loss": 4.5644,
|
|
"mean_token_accuracy": 0.2401885211467743,
|
|
"num_tokens": 14601479.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.444458150863648,
|
|
"epoch": 6.512539184952978,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001908312827652819,
|
|
"loss": 4.5766,
|
|
"mean_token_accuracy": 0.22793640196323395,
|
|
"num_tokens": 14610418.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.3856096267700195,
|
|
"epoch": 6.516457680250784,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019055240614314047,
|
|
"loss": 4.5553,
|
|
"mean_token_accuracy": 0.23759782910346985,
|
|
"num_tokens": 14618387.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.352475595474243,
|
|
"epoch": 6.52037617554859,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019027368044254177,
|
|
"loss": 4.5032,
|
|
"mean_token_accuracy": 0.24216801673173904,
|
|
"num_tokens": 14627003.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.359757661819458,
|
|
"epoch": 6.524294670846395,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001899951061616138,
|
|
"loss": 4.5717,
|
|
"mean_token_accuracy": 0.23583895117044448,
|
|
"num_tokens": 14635825.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.363951683044434,
|
|
"epoch": 6.528213166144201,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018971668379821405,
|
|
"loss": 4.5222,
|
|
"mean_token_accuracy": 0.2378536805510521,
|
|
"num_tokens": 14644830.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.335562467575073,
|
|
"epoch": 6.532131661442007,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001894384138499284,
|
|
"loss": 4.5589,
|
|
"mean_token_accuracy": 0.2230205252766609,
|
|
"num_tokens": 14653633.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.3608142852783205,
|
|
"epoch": 6.536050156739812,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018916029681407032,
|
|
"loss": 4.4181,
|
|
"mean_token_accuracy": 0.2608729347586632,
|
|
"num_tokens": 14662212.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.402458429336548,
|
|
"epoch": 6.539968652037618,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001888823331876801,
|
|
"loss": 4.5635,
|
|
"mean_token_accuracy": 0.23562585711479186,
|
|
"num_tokens": 14670942.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.435511636734009,
|
|
"epoch": 6.543887147335424,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018860452346752382,
|
|
"loss": 4.5355,
|
|
"mean_token_accuracy": 0.23375871777534485,
|
|
"num_tokens": 14679167.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.314652824401856,
|
|
"epoch": 6.547805642633229,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00018832686815009242,
|
|
"loss": 4.5544,
|
|
"mean_token_accuracy": 0.2321159288287163,
|
|
"num_tokens": 14688188.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.419964265823364,
|
|
"epoch": 6.551724137931035,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018804936773160108,
|
|
"loss": 4.6334,
|
|
"mean_token_accuracy": 0.23162245154380798,
|
|
"num_tokens": 14696958.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.351092720031739,
|
|
"epoch": 6.55564263322884,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000187772022707988,
|
|
"loss": 4.5472,
|
|
"mean_token_accuracy": 0.22869348675012588,
|
|
"num_tokens": 14705524.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.34046483039856,
|
|
"epoch": 6.559561128526646,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018749483357491365,
|
|
"loss": 4.5954,
|
|
"mean_token_accuracy": 0.23115788996219636,
|
|
"num_tokens": 14714500.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.3775536060333256,
|
|
"epoch": 6.563479623824452,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018721780082775997,
|
|
"loss": 4.577,
|
|
"mean_token_accuracy": 0.23142960369586946,
|
|
"num_tokens": 14722816.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.365781831741333,
|
|
"epoch": 6.567398119122257,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 4.5841,
|
|
"mean_token_accuracy": 0.235097935795784,
|
|
"num_tokens": 14731453.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.431947040557861,
|
|
"epoch": 6.571316614420063,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00018666420647134419,
|
|
"loss": 4.5924,
|
|
"mean_token_accuracy": 0.24101402312517167,
|
|
"num_tokens": 14739924.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.285829162597656,
|
|
"epoch": 6.575235109717868,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001863876458514448,
|
|
"loss": 4.4379,
|
|
"mean_token_accuracy": 0.24832119643688202,
|
|
"num_tokens": 14748356.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.438350057601928,
|
|
"epoch": 6.579153605015674,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00018611124359619024,
|
|
"loss": 4.6536,
|
|
"mean_token_accuracy": 0.2288836121559143,
|
|
"num_tokens": 14758846.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.295593357086181,
|
|
"epoch": 6.58307210031348,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018583500019955595,
|
|
"loss": 4.4636,
|
|
"mean_token_accuracy": 0.24019184708595276,
|
|
"num_tokens": 14767030.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.423260879516602,
|
|
"epoch": 6.586990595611285,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018555891615523374,
|
|
"loss": 4.5302,
|
|
"mean_token_accuracy": 0.2371893271803856,
|
|
"num_tokens": 14776774.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.354645872116089,
|
|
"epoch": 6.590909090909091,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018528299195663046,
|
|
"loss": 4.5241,
|
|
"mean_token_accuracy": 0.23514657765626906,
|
|
"num_tokens": 14785662.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.435997581481933,
|
|
"epoch": 6.594827586206897,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001850072280968676,
|
|
"loss": 4.6601,
|
|
"mean_token_accuracy": 0.2216114044189453,
|
|
"num_tokens": 14795070.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.370920133590698,
|
|
"epoch": 6.598746081504702,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018473162506877977,
|
|
"loss": 4.5783,
|
|
"mean_token_accuracy": 0.23041531443595886,
|
|
"num_tokens": 14805452.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.3769683837890625,
|
|
"epoch": 6.602664576802508,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018445618336491415,
|
|
"loss": 4.5476,
|
|
"mean_token_accuracy": 0.23657834827899932,
|
|
"num_tokens": 14815632.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.332512426376343,
|
|
"epoch": 6.606583072100314,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018418090347752997,
|
|
"loss": 4.4983,
|
|
"mean_token_accuracy": 0.23888321071863175,
|
|
"num_tokens": 14824228.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.369360446929932,
|
|
"epoch": 6.610501567398119,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001839057858985968,
|
|
"loss": 4.6203,
|
|
"mean_token_accuracy": 0.22393569946289063,
|
|
"num_tokens": 14833609.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.376604843139648,
|
|
"epoch": 6.614420062695925,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018363083111979452,
|
|
"loss": 4.5607,
|
|
"mean_token_accuracy": 0.2368788316845894,
|
|
"num_tokens": 14841967.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.397101545333863,
|
|
"epoch": 6.618338557993731,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018335603963251184,
|
|
"loss": 4.6155,
|
|
"mean_token_accuracy": 0.22643832713365555,
|
|
"num_tokens": 14851472.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.4742025375366214,
|
|
"epoch": 6.622257053291536,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018308141192784573,
|
|
"loss": 4.5835,
|
|
"mean_token_accuracy": 0.23251916766166686,
|
|
"num_tokens": 14859672.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.393527030944824,
|
|
"epoch": 6.626175548589342,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018280694849660034,
|
|
"loss": 4.509,
|
|
"mean_token_accuracy": 0.24504079520702363,
|
|
"num_tokens": 14867919.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.362518167495727,
|
|
"epoch": 6.630094043887147,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00018253264982928636,
|
|
"loss": 4.4915,
|
|
"mean_token_accuracy": 0.24172884821891785,
|
|
"num_tokens": 14875670.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.3848419189453125,
|
|
"epoch": 6.634012539184953,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018225851641612007,
|
|
"loss": 4.5565,
|
|
"mean_token_accuracy": 0.23705655336380005,
|
|
"num_tokens": 14883835.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.314952182769775,
|
|
"epoch": 6.637931034482759,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001819845487470223,
|
|
"loss": 4.5018,
|
|
"mean_token_accuracy": 0.2371320366859436,
|
|
"num_tokens": 14892296.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.3302477359771725,
|
|
"epoch": 6.641849529780564,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018171074731161758,
|
|
"loss": 4.5307,
|
|
"mean_token_accuracy": 0.2396359920501709,
|
|
"num_tokens": 14901334.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.321045064926148,
|
|
"epoch": 6.64576802507837,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00018143711259923363,
|
|
"loss": 4.5448,
|
|
"mean_token_accuracy": 0.2356014758348465,
|
|
"num_tokens": 14910327.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.429177284240723,
|
|
"epoch": 6.649686520376176,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001811636450989,
|
|
"loss": 4.6473,
|
|
"mean_token_accuracy": 0.22881703376770018,
|
|
"num_tokens": 14919513.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.308483695983886,
|
|
"epoch": 6.653605015673981,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001808903452993474,
|
|
"loss": 4.4293,
|
|
"mean_token_accuracy": 0.23806800693273544,
|
|
"num_tokens": 14928127.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.4016190528869625,
|
|
"epoch": 6.657523510971787,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001806172136890068,
|
|
"loss": 4.5269,
|
|
"mean_token_accuracy": 0.23862918019294738,
|
|
"num_tokens": 14936771.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.395028162002563,
|
|
"epoch": 6.661442006269592,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001803442507560089,
|
|
"loss": 4.5174,
|
|
"mean_token_accuracy": 0.24067414402961732,
|
|
"num_tokens": 14945423.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 6.661442006269592,
|
|
"eval_entropy": 5.200874075409054,
|
|
"eval_loss": 5.646055221557617,
|
|
"eval_mean_token_accuracy": 0.17362669633715883,
|
|
"eval_num_tokens": 14945423.0,
|
|
"eval_runtime": 2.1578,
|
|
"eval_samples_per_second": 1910.296,
|
|
"eval_steps_per_second": 239.135,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.318533897399902,
|
|
"epoch": 6.665360501567398,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001800714569881825,
|
|
"loss": 4.5296,
|
|
"mean_token_accuracy": 0.23779328167438507,
|
|
"num_tokens": 14953859.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.308100461959839,
|
|
"epoch": 6.669278996865204,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017979883287305454,
|
|
"loss": 4.5064,
|
|
"mean_token_accuracy": 0.24611194878816606,
|
|
"num_tokens": 14962598.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.376848888397217,
|
|
"epoch": 6.673197492163009,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017952637889784824,
|
|
"loss": 4.5183,
|
|
"mean_token_accuracy": 0.2369800552725792,
|
|
"num_tokens": 14971606.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.329435014724732,
|
|
"epoch": 6.677115987460815,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00017925409554948329,
|
|
"loss": 4.5515,
|
|
"mean_token_accuracy": 0.23839427381753922,
|
|
"num_tokens": 14979686.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.385403299331665,
|
|
"epoch": 6.681034482758621,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00017898198331457393,
|
|
"loss": 4.6482,
|
|
"mean_token_accuracy": 0.22305527180433274,
|
|
"num_tokens": 14988977.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.3881550312042235,
|
|
"epoch": 6.684952978056426,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017871004267942908,
|
|
"loss": 4.6466,
|
|
"mean_token_accuracy": 0.22244464755058288,
|
|
"num_tokens": 14996871.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.3922523021697994,
|
|
"epoch": 6.688871473354232,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00017843827413005064,
|
|
"loss": 4.5598,
|
|
"mean_token_accuracy": 0.23447040915489198,
|
|
"num_tokens": 15006022.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.32819709777832,
|
|
"epoch": 6.692789968652038,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001781666781521331,
|
|
"loss": 4.5809,
|
|
"mean_token_accuracy": 0.2346392259001732,
|
|
"num_tokens": 15014827.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.3879491806030275,
|
|
"epoch": 6.696708463949843,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00017789525523106243,
|
|
"loss": 4.5011,
|
|
"mean_token_accuracy": 0.23627623915672302,
|
|
"num_tokens": 15023157.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.423385238647461,
|
|
"epoch": 6.700626959247649,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017762400585191545,
|
|
"loss": 4.5635,
|
|
"mean_token_accuracy": 0.23084907978773117,
|
|
"num_tokens": 15032539.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.461049270629883,
|
|
"epoch": 6.704545454545455,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00017735293049945888,
|
|
"loss": 4.7025,
|
|
"mean_token_accuracy": 0.21890240609645845,
|
|
"num_tokens": 15041527.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.316124963760376,
|
|
"epoch": 6.70846394984326,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00017708202965814807,
|
|
"loss": 4.5556,
|
|
"mean_token_accuracy": 0.23203188478946685,
|
|
"num_tokens": 15050794.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.350312232971191,
|
|
"epoch": 6.712382445141066,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00017681130381212707,
|
|
"loss": 4.5451,
|
|
"mean_token_accuracy": 0.23581001311540603,
|
|
"num_tokens": 15059357.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.443229246139526,
|
|
"epoch": 6.716300940438871,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00017654075344522652,
|
|
"loss": 4.6429,
|
|
"mean_token_accuracy": 0.22093008756637572,
|
|
"num_tokens": 15067668.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.4117018699646,
|
|
"epoch": 6.720219435736677,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017627037904096416,
|
|
"loss": 4.5417,
|
|
"mean_token_accuracy": 0.22675979882478714,
|
|
"num_tokens": 15076463.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.326123428344727,
|
|
"epoch": 6.724137931034483,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017600018108254257,
|
|
"loss": 4.5195,
|
|
"mean_token_accuracy": 0.23136935830116273,
|
|
"num_tokens": 15085024.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.3365785598754885,
|
|
"epoch": 6.728056426332288,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017573016005284958,
|
|
"loss": 4.5217,
|
|
"mean_token_accuracy": 0.23426428884267808,
|
|
"num_tokens": 15093735.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.3547638893127445,
|
|
"epoch": 6.731974921630094,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017546031643445636,
|
|
"loss": 4.522,
|
|
"mean_token_accuracy": 0.23850995302200317,
|
|
"num_tokens": 15102274.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.384570789337158,
|
|
"epoch": 6.735893416927899,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001751906507096172,
|
|
"loss": 4.5306,
|
|
"mean_token_accuracy": 0.2404734656214714,
|
|
"num_tokens": 15110762.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.326816987991333,
|
|
"epoch": 6.739811912225705,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001749211633602687,
|
|
"loss": 4.5066,
|
|
"mean_token_accuracy": 0.24196906685829161,
|
|
"num_tokens": 15119104.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.37403678894043,
|
|
"epoch": 6.743730407523511,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017465185486802826,
|
|
"loss": 4.5285,
|
|
"mean_token_accuracy": 0.23840103447437286,
|
|
"num_tokens": 15127351.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.43798451423645,
|
|
"epoch": 6.747648902821316,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017438272571419373,
|
|
"loss": 4.6687,
|
|
"mean_token_accuracy": 0.22918438762426377,
|
|
"num_tokens": 15136246.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.365797805786133,
|
|
"epoch": 6.751567398119122,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 4.5885,
|
|
"mean_token_accuracy": 0.23257775157690047,
|
|
"num_tokens": 15144633.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.352477169036865,
|
|
"epoch": 6.755485893416928,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00017384500734533137,
|
|
"loss": 4.5594,
|
|
"mean_token_accuracy": 0.23000601977109908,
|
|
"num_tokens": 15153496.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.359050273895264,
|
|
"epoch": 6.759404388714733,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017357641909129324,
|
|
"loss": 4.5479,
|
|
"mean_token_accuracy": 0.2362291008234024,
|
|
"num_tokens": 15162242.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.41942195892334,
|
|
"epoch": 6.763322884012539,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017330801209763938,
|
|
"loss": 4.6116,
|
|
"mean_token_accuracy": 0.23407375663518906,
|
|
"num_tokens": 15170001.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.421488094329834,
|
|
"epoch": 6.767241379310345,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017303978684405635,
|
|
"loss": 4.5373,
|
|
"mean_token_accuracy": 0.23607497662305832,
|
|
"num_tokens": 15178086.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.3931841373443605,
|
|
"epoch": 6.77115987460815,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017277174380990647,
|
|
"loss": 4.6077,
|
|
"mean_token_accuracy": 0.2275182992219925,
|
|
"num_tokens": 15187087.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.404901123046875,
|
|
"epoch": 6.775078369905956,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017250388347422578,
|
|
"loss": 4.5729,
|
|
"mean_token_accuracy": 0.24127262383699416,
|
|
"num_tokens": 15195850.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.413230228424072,
|
|
"epoch": 6.778996865203762,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017223620631572433,
|
|
"loss": 4.5996,
|
|
"mean_token_accuracy": 0.23583441823720933,
|
|
"num_tokens": 15204174.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.406825828552246,
|
|
"epoch": 6.782915360501567,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017196871281278465,
|
|
"loss": 4.479,
|
|
"mean_token_accuracy": 0.24435777813196183,
|
|
"num_tokens": 15212640.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.44074444770813,
|
|
"epoch": 6.786833855799373,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00017170140344346086,
|
|
"loss": 4.5445,
|
|
"mean_token_accuracy": 0.22679882496595383,
|
|
"num_tokens": 15220996.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.356796836853027,
|
|
"epoch": 6.790752351097178,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001714342786854781,
|
|
"loss": 4.5759,
|
|
"mean_token_accuracy": 0.23033391535282136,
|
|
"num_tokens": 15229511.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.382933521270752,
|
|
"epoch": 6.794670846394984,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017116733901623182,
|
|
"loss": 4.6259,
|
|
"mean_token_accuracy": 0.2252475768327713,
|
|
"num_tokens": 15238673.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.381862497329712,
|
|
"epoch": 6.79858934169279,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017090058491278637,
|
|
"loss": 4.5663,
|
|
"mean_token_accuracy": 0.2270931199193001,
|
|
"num_tokens": 15246717.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.371231555938721,
|
|
"epoch": 6.802507836990595,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017063401685187457,
|
|
"loss": 4.5539,
|
|
"mean_token_accuracy": 0.23174057006835938,
|
|
"num_tokens": 15255543.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.320664501190185,
|
|
"epoch": 6.806426332288401,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017036763530989688,
|
|
"loss": 4.4771,
|
|
"mean_token_accuracy": 0.24049075543880463,
|
|
"num_tokens": 15264275.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.376600456237793,
|
|
"epoch": 6.810344827586206,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017010144076292012,
|
|
"loss": 4.5838,
|
|
"mean_token_accuracy": 0.2393489509820938,
|
|
"num_tokens": 15272376.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.404627656936645,
|
|
"epoch": 6.814263322884012,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016983543368667722,
|
|
"loss": 4.6377,
|
|
"mean_token_accuracy": 0.22406622171401977,
|
|
"num_tokens": 15282362.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.359277057647705,
|
|
"epoch": 6.818181818181818,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00016956961455656607,
|
|
"loss": 4.5913,
|
|
"mean_token_accuracy": 0.23112136870622635,
|
|
"num_tokens": 15290934.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.370607995986939,
|
|
"epoch": 6.822100313479623,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016930398384764836,
|
|
"loss": 4.494,
|
|
"mean_token_accuracy": 0.2498026207089424,
|
|
"num_tokens": 15299743.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.408206176757813,
|
|
"epoch": 6.826018808777429,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016903854203464939,
|
|
"loss": 4.5698,
|
|
"mean_token_accuracy": 0.2326970636844635,
|
|
"num_tokens": 15307858.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.3404402256011965,
|
|
"epoch": 6.829937304075235,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00016877328959195653,
|
|
"loss": 4.5367,
|
|
"mean_token_accuracy": 0.23066382706165314,
|
|
"num_tokens": 15316829.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.405491638183594,
|
|
"epoch": 6.83385579937304,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016850822699361913,
|
|
"loss": 4.6526,
|
|
"mean_token_accuracy": 0.22798590511083602,
|
|
"num_tokens": 15325394.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.36154613494873,
|
|
"epoch": 6.837774294670846,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001682433547133469,
|
|
"loss": 4.5561,
|
|
"mean_token_accuracy": 0.23089941293001176,
|
|
"num_tokens": 15333988.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.405565881729126,
|
|
"epoch": 6.841692789968652,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016797867322450948,
|
|
"loss": 4.5998,
|
|
"mean_token_accuracy": 0.2328247010707855,
|
|
"num_tokens": 15343107.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.405721044540405,
|
|
"epoch": 6.8456112852664575,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001677141830001358,
|
|
"loss": 4.535,
|
|
"mean_token_accuracy": 0.23322147578001023,
|
|
"num_tokens": 15351440.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.331574821472168,
|
|
"epoch": 6.849529780564263,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016744988451291266,
|
|
"loss": 4.4977,
|
|
"mean_token_accuracy": 0.23977696299552917,
|
|
"num_tokens": 15360426.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.356007289886475,
|
|
"epoch": 6.853448275862069,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016718577823518426,
|
|
"loss": 4.5634,
|
|
"mean_token_accuracy": 0.2369026854634285,
|
|
"num_tokens": 15369252.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.368553924560547,
|
|
"epoch": 6.8573667711598745,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016692186463895136,
|
|
"loss": 4.5801,
|
|
"mean_token_accuracy": 0.22651129812002183,
|
|
"num_tokens": 15377672.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.354849481582642,
|
|
"epoch": 6.86128526645768,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016665814419587055,
|
|
"loss": 4.5107,
|
|
"mean_token_accuracy": 0.23912533223628998,
|
|
"num_tokens": 15385585.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.415370321273803,
|
|
"epoch": 6.8652037617554855,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001663946173772528,
|
|
"loss": 4.5773,
|
|
"mean_token_accuracy": 0.22974804639816285,
|
|
"num_tokens": 15394193.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.41090874671936,
|
|
"epoch": 6.8691222570532915,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016613128465406329,
|
|
"loss": 4.5937,
|
|
"mean_token_accuracy": 0.23047085106372833,
|
|
"num_tokens": 15402498.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.306142997741699,
|
|
"epoch": 6.8730407523510975,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001658681464969204,
|
|
"loss": 4.4576,
|
|
"mean_token_accuracy": 0.24284121096134187,
|
|
"num_tokens": 15410893.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.387901449203492,
|
|
"epoch": 6.8769592476489025,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016560520337609465,
|
|
"loss": 4.634,
|
|
"mean_token_accuracy": 0.22664025723934172,
|
|
"num_tokens": 15420024.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.276748180389404,
|
|
"epoch": 6.8808777429467085,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016534245576150787,
|
|
"loss": 4.5507,
|
|
"mean_token_accuracy": 0.22926532626152038,
|
|
"num_tokens": 15428474.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.3228942394256595,
|
|
"epoch": 6.8847962382445145,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001650799041227329,
|
|
"loss": 4.4764,
|
|
"mean_token_accuracy": 0.2393030419945717,
|
|
"num_tokens": 15437019.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.425300741195679,
|
|
"epoch": 6.88871473354232,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00016481754892899193,
|
|
"loss": 4.595,
|
|
"mean_token_accuracy": 0.24194916486740112,
|
|
"num_tokens": 15445838.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.395278978347778,
|
|
"epoch": 6.8926332288401255,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001645553906491563,
|
|
"loss": 4.5804,
|
|
"mean_token_accuracy": 0.23492428064346313,
|
|
"num_tokens": 15454378.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.381695985794067,
|
|
"epoch": 6.896551724137931,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001642934297517452,
|
|
"loss": 4.5766,
|
|
"mean_token_accuracy": 0.2318781614303589,
|
|
"num_tokens": 15462447.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.321694040298462,
|
|
"epoch": 6.900470219435737,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001640316667049254,
|
|
"loss": 4.5586,
|
|
"mean_token_accuracy": 0.2371809810400009,
|
|
"num_tokens": 15471646.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.390234756469726,
|
|
"epoch": 6.9043887147335425,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00016377010197650993,
|
|
"loss": 4.5689,
|
|
"mean_token_accuracy": 0.239595165848732,
|
|
"num_tokens": 15481945.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.392079305648804,
|
|
"epoch": 6.908307210031348,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001635087360339572,
|
|
"loss": 4.5266,
|
|
"mean_token_accuracy": 0.24613254964351655,
|
|
"num_tokens": 15490558.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.393307209014893,
|
|
"epoch": 6.912225705329154,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016324756934437074,
|
|
"loss": 4.5333,
|
|
"mean_token_accuracy": 0.2366451859474182,
|
|
"num_tokens": 15499150.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.347663021087646,
|
|
"epoch": 6.91614420062696,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001629866023744977,
|
|
"loss": 4.5233,
|
|
"mean_token_accuracy": 0.23416864573955537,
|
|
"num_tokens": 15508736.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.427065467834472,
|
|
"epoch": 6.920062695924765,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016272583559072832,
|
|
"loss": 4.5954,
|
|
"mean_token_accuracy": 0.23729897141456605,
|
|
"num_tokens": 15517445.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.40287561416626,
|
|
"epoch": 6.923981191222571,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016246526945909505,
|
|
"loss": 4.5802,
|
|
"mean_token_accuracy": 0.23840786665678024,
|
|
"num_tokens": 15526282.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.425543737411499,
|
|
"epoch": 6.927899686520377,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016220490444527198,
|
|
"loss": 4.5843,
|
|
"mean_token_accuracy": 0.23946866542100906,
|
|
"num_tokens": 15534776.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.366584825515747,
|
|
"epoch": 6.931818181818182,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001619447410145735,
|
|
"loss": 4.5444,
|
|
"mean_token_accuracy": 0.24108180403709412,
|
|
"num_tokens": 15543173.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.324645757675171,
|
|
"epoch": 6.935736677115988,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016168477963195382,
|
|
"loss": 4.4803,
|
|
"mean_token_accuracy": 0.23840233385562898,
|
|
"num_tokens": 15551603.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.330155944824218,
|
|
"epoch": 6.939655172413794,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016142502076200622,
|
|
"loss": 4.5534,
|
|
"mean_token_accuracy": 0.23660384714603425,
|
|
"num_tokens": 15560412.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.317119884490967,
|
|
"epoch": 6.943573667711599,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016116546486896183,
|
|
"loss": 4.5442,
|
|
"mean_token_accuracy": 0.23395816534757613,
|
|
"num_tokens": 15569781.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.366128921508789,
|
|
"epoch": 6.947492163009405,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00016090611241668904,
|
|
"loss": 4.5039,
|
|
"mean_token_accuracy": 0.2435334101319313,
|
|
"num_tokens": 15577640.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.411202335357666,
|
|
"epoch": 6.95141065830721,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001606469638686928,
|
|
"loss": 4.6286,
|
|
"mean_token_accuracy": 0.22477778643369675,
|
|
"num_tokens": 15586950.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.328489351272583,
|
|
"epoch": 6.955329153605016,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001603880196881137,
|
|
"loss": 4.6001,
|
|
"mean_token_accuracy": 0.23528698086738586,
|
|
"num_tokens": 15595554.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.388863277435303,
|
|
"epoch": 6.959247648902822,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001601292803377269,
|
|
"loss": 4.5398,
|
|
"mean_token_accuracy": 0.2326338842511177,
|
|
"num_tokens": 15603998.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.4487920761108395,
|
|
"epoch": 6.963166144200627,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001598707462799414,
|
|
"loss": 4.6154,
|
|
"mean_token_accuracy": 0.23224805444478988,
|
|
"num_tokens": 15612977.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.411797142028808,
|
|
"epoch": 6.967084639498433,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00015961241797679977,
|
|
"loss": 4.6162,
|
|
"mean_token_accuracy": 0.2321353793144226,
|
|
"num_tokens": 15622914.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.393106698989868,
|
|
"epoch": 6.971003134796238,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015935429588997637,
|
|
"loss": 4.5639,
|
|
"mean_token_accuracy": 0.2346791595220566,
|
|
"num_tokens": 15631929.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.292469215393067,
|
|
"epoch": 6.974921630094044,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015909638048077724,
|
|
"loss": 4.449,
|
|
"mean_token_accuracy": 0.24069613516330718,
|
|
"num_tokens": 15640400.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.459117317199707,
|
|
"epoch": 6.97884012539185,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015883867221013915,
|
|
"loss": 4.6379,
|
|
"mean_token_accuracy": 0.23515487909317018,
|
|
"num_tokens": 15648934.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.4205221176147464,
|
|
"epoch": 6.982758620689655,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001585811715386285,
|
|
"loss": 4.5868,
|
|
"mean_token_accuracy": 0.22875691056251526,
|
|
"num_tokens": 15658461.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.317831897735596,
|
|
"epoch": 6.986677115987461,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015832387892644072,
|
|
"loss": 4.4951,
|
|
"mean_token_accuracy": 0.2419518917798996,
|
|
"num_tokens": 15667182.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.38176965713501,
|
|
"epoch": 6.990595611285267,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001580667948333994,
|
|
"loss": 4.6102,
|
|
"mean_token_accuracy": 0.22848259210586547,
|
|
"num_tokens": 15675951.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.3952858448028564,
|
|
"epoch": 6.994514106583072,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001578099197189555,
|
|
"loss": 4.5623,
|
|
"mean_token_accuracy": 0.23417815566062927,
|
|
"num_tokens": 15684171.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.333796644210816,
|
|
"epoch": 6.998432601880878,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00015755325404218662,
|
|
"loss": 4.5598,
|
|
"mean_token_accuracy": 0.24021217674016954,
|
|
"num_tokens": 15692986.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.440787935256958,
|
|
"epoch": 7.002351097178684,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00015729679826179577,
|
|
"loss": 4.5461,
|
|
"mean_token_accuracy": 0.2385655641555786,
|
|
"num_tokens": 15702591.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.382565021514893,
|
|
"epoch": 7.006269592476489,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015704055283611123,
|
|
"loss": 4.4592,
|
|
"mean_token_accuracy": 0.2423474982380867,
|
|
"num_tokens": 15711186.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.329565620422363,
|
|
"epoch": 7.010188087774295,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015678451822308495,
|
|
"loss": 4.5057,
|
|
"mean_token_accuracy": 0.2520110845565796,
|
|
"num_tokens": 15719831.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.401710462570191,
|
|
"epoch": 7.0141065830721,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001565286948802923,
|
|
"loss": 4.5298,
|
|
"mean_token_accuracy": 0.23961152732372284,
|
|
"num_tokens": 15727926.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.338213157653809,
|
|
"epoch": 7.018025078369906,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015627308326493118,
|
|
"loss": 4.405,
|
|
"mean_token_accuracy": 0.2539877325296402,
|
|
"num_tokens": 15735934.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.289963579177856,
|
|
"epoch": 7.021943573667712,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001560176838338209,
|
|
"loss": 4.4644,
|
|
"mean_token_accuracy": 0.24006322175264358,
|
|
"num_tokens": 15745014.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.2990635395050045,
|
|
"epoch": 7.025862068965517,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015576249704340173,
|
|
"loss": 4.407,
|
|
"mean_token_accuracy": 0.2474314883351326,
|
|
"num_tokens": 15753360.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.336004066467285,
|
|
"epoch": 7.029780564263323,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00015550752334973368,
|
|
"loss": 4.4505,
|
|
"mean_token_accuracy": 0.24653201699256896,
|
|
"num_tokens": 15762395.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.350561809539795,
|
|
"epoch": 7.033699059561129,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015525276320849628,
|
|
"loss": 4.3751,
|
|
"mean_token_accuracy": 0.2573044627904892,
|
|
"num_tokens": 15771292.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.39741587638855,
|
|
"epoch": 7.037617554858934,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015499821707498702,
|
|
"loss": 4.4772,
|
|
"mean_token_accuracy": 0.2485238328576088,
|
|
"num_tokens": 15779879.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.334306240081787,
|
|
"epoch": 7.04153605015674,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015474388540412138,
|
|
"loss": 4.463,
|
|
"mean_token_accuracy": 0.2476590469479561,
|
|
"num_tokens": 15788960.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.402884817123413,
|
|
"epoch": 7.045454545454546,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000154489768650431,
|
|
"loss": 4.4996,
|
|
"mean_token_accuracy": 0.24270211905241013,
|
|
"num_tokens": 15798260.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.3795098781585695,
|
|
"epoch": 7.049373040752351,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015423586726806394,
|
|
"loss": 4.4857,
|
|
"mean_token_accuracy": 0.2480814889073372,
|
|
"num_tokens": 15807480.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.410169219970703,
|
|
"epoch": 7.053291536050157,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00015398218171078305,
|
|
"loss": 4.5579,
|
|
"mean_token_accuracy": 0.23652213513851167,
|
|
"num_tokens": 15816330.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 7.053291536050157,
|
|
"eval_entropy": 5.213407396346099,
|
|
"eval_loss": 5.650358200073242,
|
|
"eval_mean_token_accuracy": 0.17421094881182955,
|
|
"eval_num_tokens": 15816330.0,
|
|
"eval_runtime": 2.3503,
|
|
"eval_samples_per_second": 1753.816,
|
|
"eval_steps_per_second": 219.546,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.351854467391968,
|
|
"epoch": 7.057210031347962,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015372871243196545,
|
|
"loss": 4.4611,
|
|
"mean_token_accuracy": 0.2453000009059906,
|
|
"num_tokens": 15824600.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.38481879234314,
|
|
"epoch": 7.061128526645768,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001534754598846021,
|
|
"loss": 4.4973,
|
|
"mean_token_accuracy": 0.2386948436498642,
|
|
"num_tokens": 15834170.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.418598747253418,
|
|
"epoch": 7.065047021943574,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015322242452129616,
|
|
"loss": 4.5208,
|
|
"mean_token_accuracy": 0.2356327325105667,
|
|
"num_tokens": 15843061.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.374097537994385,
|
|
"epoch": 7.068965517241379,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015296960679426285,
|
|
"loss": 4.4804,
|
|
"mean_token_accuracy": 0.24357059299945832,
|
|
"num_tokens": 15851788.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.416475820541382,
|
|
"epoch": 7.072884012539185,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015271700715532854,
|
|
"loss": 4.5146,
|
|
"mean_token_accuracy": 0.2418033465743065,
|
|
"num_tokens": 15860216.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 5.36783537864685,
|
|
"epoch": 7.076802507836991,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015246462605592972,
|
|
"loss": 4.3641,
|
|
"mean_token_accuracy": 0.2551824957132339,
|
|
"num_tokens": 15869498.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.349653244018555,
|
|
"epoch": 7.080721003134796,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015221246394711224,
|
|
"loss": 4.4243,
|
|
"mean_token_accuracy": 0.24866150170564652,
|
|
"num_tokens": 15877981.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.380398464202881,
|
|
"epoch": 7.084639498432602,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00015196052127953085,
|
|
"loss": 4.4788,
|
|
"mean_token_accuracy": 0.24613707065582274,
|
|
"num_tokens": 15886454.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.290103721618652,
|
|
"epoch": 7.088557993730408,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001517087985034477,
|
|
"loss": 4.4037,
|
|
"mean_token_accuracy": 0.2512354001402855,
|
|
"num_tokens": 15895025.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.396518898010254,
|
|
"epoch": 7.092476489028213,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001514572960687325,
|
|
"loss": 4.5212,
|
|
"mean_token_accuracy": 0.2370706394314766,
|
|
"num_tokens": 15903611.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.4300117015838625,
|
|
"epoch": 7.096394984326019,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015120601442486061,
|
|
"loss": 4.505,
|
|
"mean_token_accuracy": 0.24787195175886154,
|
|
"num_tokens": 15912078.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.337722444534302,
|
|
"epoch": 7.100313479623824,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00015095495402091329,
|
|
"loss": 4.4203,
|
|
"mean_token_accuracy": 0.2462344914674759,
|
|
"num_tokens": 15920914.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.374953365325927,
|
|
"epoch": 7.10423197492163,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001507041153055761,
|
|
"loss": 4.4122,
|
|
"mean_token_accuracy": 0.25229266434907915,
|
|
"num_tokens": 15929199.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.451293087005615,
|
|
"epoch": 7.108150470219436,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00015045349872713838,
|
|
"loss": 4.5538,
|
|
"mean_token_accuracy": 0.2362870454788208,
|
|
"num_tokens": 15937817.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.368142557144165,
|
|
"epoch": 7.112068965517241,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00015020310473349275,
|
|
"loss": 4.4764,
|
|
"mean_token_accuracy": 0.2515915781259537,
|
|
"num_tokens": 15945997.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.3515660762786865,
|
|
"epoch": 7.115987460815047,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00014995293377213388,
|
|
"loss": 4.4507,
|
|
"mean_token_accuracy": 0.24186779260635377,
|
|
"num_tokens": 15954215.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.378357458114624,
|
|
"epoch": 7.119905956112853,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001497029862901578,
|
|
"loss": 4.4942,
|
|
"mean_token_accuracy": 0.24566491693258286,
|
|
"num_tokens": 15962222.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.400769948959351,
|
|
"epoch": 7.123824451410658,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00014945326273426102,
|
|
"loss": 4.4146,
|
|
"mean_token_accuracy": 0.24633834064006804,
|
|
"num_tokens": 15970712.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.398710250854492,
|
|
"epoch": 7.127742946708464,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00014920376355074034,
|
|
"loss": 4.5038,
|
|
"mean_token_accuracy": 0.2400316119194031,
|
|
"num_tokens": 15979049.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.324611663818359,
|
|
"epoch": 7.131661442006269,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014895448918549097,
|
|
"loss": 4.464,
|
|
"mean_token_accuracy": 0.2415042296051979,
|
|
"num_tokens": 15988221.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.357416677474975,
|
|
"epoch": 7.135579937304075,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00014870544008400681,
|
|
"loss": 4.4504,
|
|
"mean_token_accuracy": 0.23646633327007294,
|
|
"num_tokens": 15996579.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.345903062820435,
|
|
"epoch": 7.139498432601881,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001484566166913789,
|
|
"loss": 4.4779,
|
|
"mean_token_accuracy": 0.24050430804491044,
|
|
"num_tokens": 16005489.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.335462331771851,
|
|
"epoch": 7.143416927899686,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00014820801945229504,
|
|
"loss": 4.4363,
|
|
"mean_token_accuracy": 0.2470606356859207,
|
|
"num_tokens": 16014167.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.344165802001953,
|
|
"epoch": 7.147335423197492,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001479596488110387,
|
|
"loss": 4.359,
|
|
"mean_token_accuracy": 0.25832580775022507,
|
|
"num_tokens": 16022184.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.388113117218017,
|
|
"epoch": 7.151253918495298,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014771150521148863,
|
|
"loss": 4.4815,
|
|
"mean_token_accuracy": 0.2414539024233818,
|
|
"num_tokens": 16031704.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.370992994308471,
|
|
"epoch": 7.155172413793103,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001474635890971176,
|
|
"loss": 4.5304,
|
|
"mean_token_accuracy": 0.24115773290395737,
|
|
"num_tokens": 16040913.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.345702457427978,
|
|
"epoch": 7.159090909090909,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001472159009109919,
|
|
"loss": 4.4273,
|
|
"mean_token_accuracy": 0.2395051747560501,
|
|
"num_tokens": 16048943.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.375765943527222,
|
|
"epoch": 7.163009404388715,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00014696844109577043,
|
|
"loss": 4.4166,
|
|
"mean_token_accuracy": 0.2439316764473915,
|
|
"num_tokens": 16057999.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 5.407761096954346,
|
|
"epoch": 7.16692789968652,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014672121009370415,
|
|
"loss": 4.556,
|
|
"mean_token_accuracy": 0.23141980171203613,
|
|
"num_tokens": 16066920.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 5.314061832427979,
|
|
"epoch": 7.170846394984326,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014647420834663482,
|
|
"loss": 4.4916,
|
|
"mean_token_accuracy": 0.23647740632295608,
|
|
"num_tokens": 16075128.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 5.326253843307495,
|
|
"epoch": 7.174764890282131,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001462274362959946,
|
|
"loss": 4.4845,
|
|
"mean_token_accuracy": 0.24234554022550583,
|
|
"num_tokens": 16083949.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 5.380868530273437,
|
|
"epoch": 7.178683385579937,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001459808943828053,
|
|
"loss": 4.4411,
|
|
"mean_token_accuracy": 0.2473629891872406,
|
|
"num_tokens": 16092891.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.413130903244019,
|
|
"epoch": 7.182601880877743,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00014573458304767707,
|
|
"loss": 4.5405,
|
|
"mean_token_accuracy": 0.23372244983911514,
|
|
"num_tokens": 16101602.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.417646408081055,
|
|
"epoch": 7.186520376175548,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014548850273080839,
|
|
"loss": 4.5181,
|
|
"mean_token_accuracy": 0.23602350950241088,
|
|
"num_tokens": 16110430.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.3205465316772464,
|
|
"epoch": 7.190438871473354,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00014524265387198465,
|
|
"loss": 4.4464,
|
|
"mean_token_accuracy": 0.24494406580924988,
|
|
"num_tokens": 16119444.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.329060745239258,
|
|
"epoch": 7.19435736677116,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014499703691057766,
|
|
"loss": 4.4301,
|
|
"mean_token_accuracy": 0.2463176056742668,
|
|
"num_tokens": 16128721.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.397345304489136,
|
|
"epoch": 7.198275862068965,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001447516522855447,
|
|
"loss": 4.5236,
|
|
"mean_token_accuracy": 0.23731179982423783,
|
|
"num_tokens": 16137508.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.40913953781128,
|
|
"epoch": 7.202194357366771,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00014450650043542775,
|
|
"loss": 4.4648,
|
|
"mean_token_accuracy": 0.24815330803394317,
|
|
"num_tokens": 16146239.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.393718624114991,
|
|
"epoch": 7.206112852664576,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014426158179835317,
|
|
"loss": 4.4353,
|
|
"mean_token_accuracy": 0.2448173940181732,
|
|
"num_tokens": 16154769.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.273973512649536,
|
|
"epoch": 7.210031347962382,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001440168968120301,
|
|
"loss": 4.3423,
|
|
"mean_token_accuracy": 0.25829260498285295,
|
|
"num_tokens": 16163307.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.362701559066773,
|
|
"epoch": 7.213949843260188,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00014377244591375033,
|
|
"loss": 4.4445,
|
|
"mean_token_accuracy": 0.24364533126354218,
|
|
"num_tokens": 16171533.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.306896686553955,
|
|
"epoch": 7.217868338557993,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001435282295403873,
|
|
"loss": 4.4254,
|
|
"mean_token_accuracy": 0.25574268102645875,
|
|
"num_tokens": 16180163.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 5.311830663681031,
|
|
"epoch": 7.221786833855799,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00014328424812839525,
|
|
"loss": 4.4201,
|
|
"mean_token_accuracy": 0.24944522231817245,
|
|
"num_tokens": 16188060.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 5.3369323253631595,
|
|
"epoch": 7.225705329153605,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014304050211380836,
|
|
"loss": 4.5239,
|
|
"mean_token_accuracy": 0.2351473644375801,
|
|
"num_tokens": 16197597.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.3946469783782955,
|
|
"epoch": 7.22962382445141,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00014279699193224047,
|
|
"loss": 4.5157,
|
|
"mean_token_accuracy": 0.2263515219092369,
|
|
"num_tokens": 16206288.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 5.358144474029541,
|
|
"epoch": 7.233542319749216,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014255371801888379,
|
|
"loss": 4.4749,
|
|
"mean_token_accuracy": 0.23528446704149247,
|
|
"num_tokens": 16215321.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.357288980484009,
|
|
"epoch": 7.237460815047022,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00014231068080850813,
|
|
"loss": 4.4,
|
|
"mean_token_accuracy": 0.25356293469667435,
|
|
"num_tokens": 16224445.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.351481628417969,
|
|
"epoch": 7.241379310344827,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014206788073546035,
|
|
"loss": 4.4465,
|
|
"mean_token_accuracy": 0.24855274856090545,
|
|
"num_tokens": 16233189.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.341161298751831,
|
|
"epoch": 7.245297805642633,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00014182531823366372,
|
|
"loss": 4.434,
|
|
"mean_token_accuracy": 0.2545219734311104,
|
|
"num_tokens": 16241664.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 5.340350914001465,
|
|
"epoch": 7.2492163009404385,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00014158299373661666,
|
|
"loss": 4.4791,
|
|
"mean_token_accuracy": 0.24678215384483337,
|
|
"num_tokens": 16251293.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.398111152648926,
|
|
"epoch": 7.253134796238244,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00014134090767739223,
|
|
"loss": 4.4627,
|
|
"mean_token_accuracy": 0.2457396313548088,
|
|
"num_tokens": 16260134.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.409440040588379,
|
|
"epoch": 7.25705329153605,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00014109906048863772,
|
|
"loss": 4.4747,
|
|
"mean_token_accuracy": 0.2451981410384178,
|
|
"num_tokens": 16269027.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.356630516052246,
|
|
"epoch": 7.2609717868338555,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001408574526025731,
|
|
"loss": 4.4914,
|
|
"mean_token_accuracy": 0.2458111360669136,
|
|
"num_tokens": 16277692.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.3549261569976805,
|
|
"epoch": 7.264890282131661,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014061608445099083,
|
|
"loss": 4.4994,
|
|
"mean_token_accuracy": 0.2390519767999649,
|
|
"num_tokens": 16287059.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.282725286483765,
|
|
"epoch": 7.268808777429467,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00014037495646525495,
|
|
"loss": 4.4425,
|
|
"mean_token_accuracy": 0.2474353864789009,
|
|
"num_tokens": 16294880.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.33547887802124,
|
|
"epoch": 7.2727272727272725,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001401340690763003,
|
|
"loss": 4.4123,
|
|
"mean_token_accuracy": 0.25003220438957213,
|
|
"num_tokens": 16304101.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.391659593582153,
|
|
"epoch": 7.2766457680250785,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013989342271463173,
|
|
"loss": 4.5639,
|
|
"mean_token_accuracy": 0.23170257955789567,
|
|
"num_tokens": 16312887.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.32423996925354,
|
|
"epoch": 7.280564263322884,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001396530178103232,
|
|
"loss": 4.4608,
|
|
"mean_token_accuracy": 0.2440595492720604,
|
|
"num_tokens": 16322256.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.3475456714630125,
|
|
"epoch": 7.2844827586206895,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001394128547930174,
|
|
"loss": 4.4849,
|
|
"mean_token_accuracy": 0.2395765781402588,
|
|
"num_tokens": 16331810.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.419012641906738,
|
|
"epoch": 7.2884012539184955,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001391729340919245,
|
|
"loss": 4.4909,
|
|
"mean_token_accuracy": 0.2402353748679161,
|
|
"num_tokens": 16341385.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.352398300170899,
|
|
"epoch": 7.2923197492163006,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00013893325613582172,
|
|
"loss": 4.4399,
|
|
"mean_token_accuracy": 0.2459324687719345,
|
|
"num_tokens": 16350231.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.39335560798645,
|
|
"epoch": 7.2962382445141065,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001386938213530523,
|
|
"loss": 4.5203,
|
|
"mean_token_accuracy": 0.23949545323848725,
|
|
"num_tokens": 16359659.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 5.349349689483643,
|
|
"epoch": 7.3001567398119125,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00013845463017152527,
|
|
"loss": 4.4691,
|
|
"mean_token_accuracy": 0.2441416099667549,
|
|
"num_tokens": 16368606.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.398768758773803,
|
|
"epoch": 7.304075235109718,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013821568301871384,
|
|
"loss": 4.5282,
|
|
"mean_token_accuracy": 0.238643978536129,
|
|
"num_tokens": 16376428.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.337236452102661,
|
|
"epoch": 7.3079937304075235,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013797698032165534,
|
|
"loss": 4.4667,
|
|
"mean_token_accuracy": 0.24058870375156402,
|
|
"num_tokens": 16384983.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.3606503963470455,
|
|
"epoch": 7.3119122257053295,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013773852250695035,
|
|
"loss": 4.4868,
|
|
"mean_token_accuracy": 0.23594093024730683,
|
|
"num_tokens": 16393747.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 5.377978515625,
|
|
"epoch": 7.315830721003135,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00013750031000076145,
|
|
"loss": 4.4849,
|
|
"mean_token_accuracy": 0.2438811793923378,
|
|
"num_tokens": 16401899.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.362395143508911,
|
|
"epoch": 7.3197492163009406,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001372623432288131,
|
|
"loss": 4.4942,
|
|
"mean_token_accuracy": 0.23797068744897842,
|
|
"num_tokens": 16410787.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 5.408616018295288,
|
|
"epoch": 7.3236677115987465,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00013702462261639054,
|
|
"loss": 4.5098,
|
|
"mean_token_accuracy": 0.23776200860738755,
|
|
"num_tokens": 16420023.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.370176362991333,
|
|
"epoch": 7.327586206896552,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000136787148588339,
|
|
"loss": 4.4377,
|
|
"mean_token_accuracy": 0.24228774160146713,
|
|
"num_tokens": 16428394.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.375113821029663,
|
|
"epoch": 7.331504702194358,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001365499215690632,
|
|
"loss": 4.4481,
|
|
"mean_token_accuracy": 0.24545957744121552,
|
|
"num_tokens": 16436429.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.2817174911499025,
|
|
"epoch": 7.335423197492163,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00013631294198252603,
|
|
"loss": 4.3875,
|
|
"mean_token_accuracy": 0.25321238338947294,
|
|
"num_tokens": 16445140.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.345230436325073,
|
|
"epoch": 7.339341692789969,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013607621025224864,
|
|
"loss": 4.4376,
|
|
"mean_token_accuracy": 0.2413479819893837,
|
|
"num_tokens": 16453932.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.329945659637451,
|
|
"epoch": 7.343260188087775,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000135839726801309,
|
|
"loss": 4.4494,
|
|
"mean_token_accuracy": 0.24514175504446029,
|
|
"num_tokens": 16462539.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.362383031845093,
|
|
"epoch": 7.34717868338558,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013560349205234124,
|
|
"loss": 4.5108,
|
|
"mean_token_accuracy": 0.23988377004861833,
|
|
"num_tokens": 16471979.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.41093111038208,
|
|
"epoch": 7.351097178683386,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013536750642753538,
|
|
"loss": 4.5709,
|
|
"mean_token_accuracy": 0.23426455706357957,
|
|
"num_tokens": 16481266.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 5.3725179672241214,
|
|
"epoch": 7.355015673981192,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013513177034863588,
|
|
"loss": 4.5096,
|
|
"mean_token_accuracy": 0.24622242152690887,
|
|
"num_tokens": 16490360.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.441692352294922,
|
|
"epoch": 7.358934169278997,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001348962842369413,
|
|
"loss": 4.5417,
|
|
"mean_token_accuracy": 0.23306599706411363,
|
|
"num_tokens": 16498043.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.360684728622436,
|
|
"epoch": 7.362852664576803,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013466104851330346,
|
|
"loss": 4.4853,
|
|
"mean_token_accuracy": 0.2455366000533104,
|
|
"num_tokens": 16506343.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 5.353933334350586,
|
|
"epoch": 7.366771159874608,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001344260635981268,
|
|
"loss": 4.4779,
|
|
"mean_token_accuracy": 0.24376274198293685,
|
|
"num_tokens": 16515308.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.378349494934082,
|
|
"epoch": 7.370689655172414,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00013419132991136762,
|
|
"loss": 4.4718,
|
|
"mean_token_accuracy": 0.23848281502723695,
|
|
"num_tokens": 16524379.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.3918250560760494,
|
|
"epoch": 7.37460815047022,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013395684787253278,
|
|
"loss": 4.4804,
|
|
"mean_token_accuracy": 0.23886895030736924,
|
|
"num_tokens": 16532797.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.385506725311279,
|
|
"epoch": 7.378526645768025,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00013372261790068,
|
|
"loss": 4.5115,
|
|
"mean_token_accuracy": 0.2392851307988167,
|
|
"num_tokens": 16540984.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 5.338020086288452,
|
|
"epoch": 7.382445141065831,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013348864041441604,
|
|
"loss": 4.3983,
|
|
"mean_token_accuracy": 0.24625764787197113,
|
|
"num_tokens": 16549386.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.308428812026977,
|
|
"epoch": 7.386363636363637,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013325491583189664,
|
|
"loss": 4.4616,
|
|
"mean_token_accuracy": 0.24433789551258087,
|
|
"num_tokens": 16557673.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.387760305404663,
|
|
"epoch": 7.390282131661442,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013302144457082553,
|
|
"loss": 4.4926,
|
|
"mean_token_accuracy": 0.24186737686395646,
|
|
"num_tokens": 16566383.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.397812271118164,
|
|
"epoch": 7.394200626959248,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013278822704845372,
|
|
"loss": 4.4484,
|
|
"mean_token_accuracy": 0.24904673248529435,
|
|
"num_tokens": 16574819.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.3718208312988285,
|
|
"epoch": 7.398119122257054,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013255526368157876,
|
|
"loss": 4.4733,
|
|
"mean_token_accuracy": 0.24116274565458298,
|
|
"num_tokens": 16582571.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.431145286560058,
|
|
"epoch": 7.402037617554859,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013232255488654387,
|
|
"loss": 4.5451,
|
|
"mean_token_accuracy": 0.23890796005725862,
|
|
"num_tokens": 16592045.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.374118185043335,
|
|
"epoch": 7.405956112852665,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013209010107923747,
|
|
"loss": 4.4006,
|
|
"mean_token_accuracy": 0.24616035670042039,
|
|
"num_tokens": 16600864.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.37995491027832,
|
|
"epoch": 7.40987460815047,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00013185790267509218,
|
|
"loss": 4.5669,
|
|
"mean_token_accuracy": 0.24262983351945877,
|
|
"num_tokens": 16610193.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.356078100204468,
|
|
"epoch": 7.413793103448276,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00013162596008908423,
|
|
"loss": 4.5205,
|
|
"mean_token_accuracy": 0.2357272610068321,
|
|
"num_tokens": 16618663.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.365172338485718,
|
|
"epoch": 7.417711598746082,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013139427373573255,
|
|
"loss": 4.4381,
|
|
"mean_token_accuracy": 0.24956832081079483,
|
|
"num_tokens": 16626885.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.416658687591553,
|
|
"epoch": 7.421630094043887,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013116284402909834,
|
|
"loss": 4.5472,
|
|
"mean_token_accuracy": 0.23631460815668107,
|
|
"num_tokens": 16636592.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.409358167648316,
|
|
"epoch": 7.425548589341693,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000130931671382784,
|
|
"loss": 4.5345,
|
|
"mean_token_accuracy": 0.24192966818809508,
|
|
"num_tokens": 16645897.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.412053442001342,
|
|
"epoch": 7.429467084639499,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013070075620993233,
|
|
"loss": 4.4797,
|
|
"mean_token_accuracy": 0.24083305299282073,
|
|
"num_tokens": 16654034.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.34807858467102,
|
|
"epoch": 7.433385579937304,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013047009892322642,
|
|
"loss": 4.4718,
|
|
"mean_token_accuracy": 0.2501526653766632,
|
|
"num_tokens": 16662510.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.354178094863892,
|
|
"epoch": 7.43730407523511,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001302396999348882,
|
|
"loss": 4.446,
|
|
"mean_token_accuracy": 0.2495130106806755,
|
|
"num_tokens": 16671014.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.376804447174072,
|
|
"epoch": 7.441222570532915,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013000955965667788,
|
|
"loss": 4.522,
|
|
"mean_token_accuracy": 0.23666924387216567,
|
|
"num_tokens": 16680068.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.402717065811157,
|
|
"epoch": 7.445141065830721,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001297796784998936,
|
|
"loss": 4.5501,
|
|
"mean_token_accuracy": 0.2416461154818535,
|
|
"num_tokens": 16688478.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 7.445141065830721,
|
|
"eval_entropy": 5.219952181328175,
|
|
"eval_loss": 5.6512131690979,
|
|
"eval_mean_token_accuracy": 0.17356166133815928,
|
|
"eval_num_tokens": 16688478.0,
|
|
"eval_runtime": 2.1585,
|
|
"eval_samples_per_second": 1909.697,
|
|
"eval_steps_per_second": 239.06,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 5.422776460647583,
|
|
"epoch": 7.449059561128527,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00012955005687537025,
|
|
"loss": 4.5791,
|
|
"mean_token_accuracy": 0.23783110231161117,
|
|
"num_tokens": 16698913.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 5.3249043941497805,
|
|
"epoch": 7.452978056426332,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00012932069519347873,
|
|
"loss": 4.4307,
|
|
"mean_token_accuracy": 0.24400845915079117,
|
|
"num_tokens": 16708306.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 5.349935150146484,
|
|
"epoch": 7.456896551724138,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012909159386412578,
|
|
"loss": 4.4853,
|
|
"mean_token_accuracy": 0.2509059712290764,
|
|
"num_tokens": 16716345.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 5.335985231399536,
|
|
"epoch": 7.460815047021944,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012886275329675246,
|
|
"loss": 4.4975,
|
|
"mean_token_accuracy": 0.24026042073965073,
|
|
"num_tokens": 16725405.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 5.323128175735474,
|
|
"epoch": 7.464733542319749,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001286341739003341,
|
|
"loss": 4.4745,
|
|
"mean_token_accuracy": 0.24026674181222915,
|
|
"num_tokens": 16734771.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 5.3368408203125,
|
|
"epoch": 7.468652037617555,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00012840585608337907,
|
|
"loss": 4.5306,
|
|
"mean_token_accuracy": 0.23413396924734114,
|
|
"num_tokens": 16742403.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 5.3224304676055905,
|
|
"epoch": 7.472570532915361,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012817780025392833,
|
|
"loss": 4.4927,
|
|
"mean_token_accuracy": 0.24681653082370758,
|
|
"num_tokens": 16751450.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 5.339302682876587,
|
|
"epoch": 7.476489028213166,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00012795000681955468,
|
|
"loss": 4.4527,
|
|
"mean_token_accuracy": 0.2478245735168457,
|
|
"num_tokens": 16760216.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 5.368650674819946,
|
|
"epoch": 7.480407523510972,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012772247618736182,
|
|
"loss": 4.5225,
|
|
"mean_token_accuracy": 0.2414704069495201,
|
|
"num_tokens": 16768648.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 5.290320110321045,
|
|
"epoch": 7.484326018808777,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00012749520876398395,
|
|
"loss": 4.3867,
|
|
"mean_token_accuracy": 0.25117810517549516,
|
|
"num_tokens": 16776587.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 5.36309323310852,
|
|
"epoch": 7.488244514106583,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012726820495558483,
|
|
"loss": 4.4827,
|
|
"mean_token_accuracy": 0.24279263466596604,
|
|
"num_tokens": 16785937.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 5.345360803604126,
|
|
"epoch": 7.492163009404389,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012704146516785705,
|
|
"loss": 4.4434,
|
|
"mean_token_accuracy": 0.24300254732370377,
|
|
"num_tokens": 16794359.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 5.379220914840698,
|
|
"epoch": 7.496081504702194,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001268149898060213,
|
|
"loss": 4.488,
|
|
"mean_token_accuracy": 0.24235294610261918,
|
|
"num_tokens": 16803685.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 5.2875770092010494,
|
|
"epoch": 7.5,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00012658877927482586,
|
|
"loss": 4.4295,
|
|
"mean_token_accuracy": 0.2503228232264519,
|
|
"num_tokens": 16812267.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 5.391651725769043,
|
|
"epoch": 7.503918495297806,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001263628339785456,
|
|
"loss": 4.5559,
|
|
"mean_token_accuracy": 0.24211336821317672,
|
|
"num_tokens": 16820334.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 5.365380573272705,
|
|
"epoch": 7.507836990595611,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012613715432098138,
|
|
"loss": 4.4944,
|
|
"mean_token_accuracy": 0.2450470969080925,
|
|
"num_tokens": 16828908.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 5.413535833358765,
|
|
"epoch": 7.511755485893417,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00012591174070545932,
|
|
"loss": 4.5052,
|
|
"mean_token_accuracy": 0.24078808724880219,
|
|
"num_tokens": 16837772.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.383439922332764,
|
|
"epoch": 7.515673981191222,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00012568659353483024,
|
|
"loss": 4.5397,
|
|
"mean_token_accuracy": 0.23518583178520203,
|
|
"num_tokens": 16846714.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 5.378705835342407,
|
|
"epoch": 7.519592476489028,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012546171321146848,
|
|
"loss": 4.4543,
|
|
"mean_token_accuracy": 0.24314221143722534,
|
|
"num_tokens": 16854745.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 5.414407777786255,
|
|
"epoch": 7.523510971786834,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012523710013727173,
|
|
"loss": 4.4906,
|
|
"mean_token_accuracy": 0.24661173522472382,
|
|
"num_tokens": 16864181.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.399878358840942,
|
|
"epoch": 7.527429467084639,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00012501275471365998,
|
|
"loss": 4.526,
|
|
"mean_token_accuracy": 0.23486624658107758,
|
|
"num_tokens": 16873551.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 5.3849061012268065,
|
|
"epoch": 7.531347962382445,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012478867734157486,
|
|
"loss": 4.581,
|
|
"mean_token_accuracy": 0.23321934044361115,
|
|
"num_tokens": 16883108.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 5.336598443984985,
|
|
"epoch": 7.535266457680251,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012456486842147885,
|
|
"loss": 4.4532,
|
|
"mean_token_accuracy": 0.25436029583215714,
|
|
"num_tokens": 16891772.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 5.331082534790039,
|
|
"epoch": 7.539184952978056,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00012434132835335496,
|
|
"loss": 4.441,
|
|
"mean_token_accuracy": 0.24696197509765624,
|
|
"num_tokens": 16900519.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 5.400768375396728,
|
|
"epoch": 7.543103448275862,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00012411805753670543,
|
|
"loss": 4.5096,
|
|
"mean_token_accuracy": 0.24485050588846208,
|
|
"num_tokens": 16909228.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 5.342201614379883,
|
|
"epoch": 7.547021943573668,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001238950563705514,
|
|
"loss": 4.5129,
|
|
"mean_token_accuracy": 0.24116040915250778,
|
|
"num_tokens": 16918100.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 5.361485958099365,
|
|
"epoch": 7.550940438871473,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001236723252534322,
|
|
"loss": 4.4806,
|
|
"mean_token_accuracy": 0.24281463623046876,
|
|
"num_tokens": 16926911.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 5.400156497955322,
|
|
"epoch": 7.554858934169279,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012344986458340433,
|
|
"loss": 4.5232,
|
|
"mean_token_accuracy": 0.23811523318290712,
|
|
"num_tokens": 16935638.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 5.336216783523559,
|
|
"epoch": 7.558777429467085,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012322767475804122,
|
|
"loss": 4.3654,
|
|
"mean_token_accuracy": 0.26129130870103834,
|
|
"num_tokens": 16944351.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 5.4017571926116945,
|
|
"epoch": 7.56269592476489,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00012300575617443194,
|
|
"loss": 4.5117,
|
|
"mean_token_accuracy": 0.2431052267551422,
|
|
"num_tokens": 16952486.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 5.395779609680176,
|
|
"epoch": 7.566614420062696,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001227841092291812,
|
|
"loss": 4.5591,
|
|
"mean_token_accuracy": 0.23626385480165482,
|
|
"num_tokens": 16960877.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.360510206222534,
|
|
"epoch": 7.570532915360501,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00012256273431840792,
|
|
"loss": 4.5272,
|
|
"mean_token_accuracy": 0.24673281610012054,
|
|
"num_tokens": 16970868.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 5.341772794723511,
|
|
"epoch": 7.574451410658307,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012234163183774482,
|
|
"loss": 4.523,
|
|
"mean_token_accuracy": 0.24206652045249938,
|
|
"num_tokens": 16980339.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.35775842666626,
|
|
"epoch": 7.578369905956113,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012212080218233814,
|
|
"loss": 4.4503,
|
|
"mean_token_accuracy": 0.23950976729393006,
|
|
"num_tokens": 16988786.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 5.348165464401245,
|
|
"epoch": 7.582288401253918,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00012190024574684614,
|
|
"loss": 4.4741,
|
|
"mean_token_accuracy": 0.24090987145900727,
|
|
"num_tokens": 16998454.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 5.3573033809661865,
|
|
"epoch": 7.586206896551724,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00012167996292543889,
|
|
"loss": 4.4783,
|
|
"mean_token_accuracy": 0.2442306473851204,
|
|
"num_tokens": 17006951.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 5.420799732208252,
|
|
"epoch": 7.59012539184953,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001214599541117976,
|
|
"loss": 4.4971,
|
|
"mean_token_accuracy": 0.24409402161836624,
|
|
"num_tokens": 17015991.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 5.3700957775115965,
|
|
"epoch": 7.594043887147335,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00012124021969911366,
|
|
"loss": 4.5057,
|
|
"mean_token_accuracy": 0.24502206295728685,
|
|
"num_tokens": 17024493.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 5.4092669010162355,
|
|
"epoch": 7.597962382445141,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00012102076008008804,
|
|
"loss": 4.4999,
|
|
"mean_token_accuracy": 0.24524618834257125,
|
|
"num_tokens": 17033402.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 5.371424150466919,
|
|
"epoch": 7.601880877742946,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00012080157564693073,
|
|
"loss": 4.4464,
|
|
"mean_token_accuracy": 0.2475828170776367,
|
|
"num_tokens": 17042417.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 5.387950801849366,
|
|
"epoch": 7.605799373040752,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00012058266679135987,
|
|
"loss": 4.5998,
|
|
"mean_token_accuracy": 0.2316117137670517,
|
|
"num_tokens": 17050395.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 5.4202492237091064,
|
|
"epoch": 7.609717868338558,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012036403390460107,
|
|
"loss": 4.5475,
|
|
"mean_token_accuracy": 0.23720477372407914,
|
|
"num_tokens": 17060077.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 5.386623430252075,
|
|
"epoch": 7.613636363636363,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012014567737738667,
|
|
"loss": 4.4823,
|
|
"mean_token_accuracy": 0.23551006764173507,
|
|
"num_tokens": 17068099.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 5.353383159637451,
|
|
"epoch": 7.617554858934169,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011992759759995529,
|
|
"loss": 4.4516,
|
|
"mean_token_accuracy": 0.240041883289814,
|
|
"num_tokens": 17076846.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 5.3551304817199705,
|
|
"epoch": 7.621473354231975,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011970979496205078,
|
|
"loss": 4.4636,
|
|
"mean_token_accuracy": 0.24167897254228593,
|
|
"num_tokens": 17085960.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 5.392842149734497,
|
|
"epoch": 7.62539184952978,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011949226985292183,
|
|
"loss": 4.4436,
|
|
"mean_token_accuracy": 0.24874720424413682,
|
|
"num_tokens": 17095327.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 5.332487535476685,
|
|
"epoch": 7.629310344827586,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011927502266132098,
|
|
"loss": 4.4608,
|
|
"mean_token_accuracy": 0.24539865255355836,
|
|
"num_tokens": 17103961.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 5.333521175384521,
|
|
"epoch": 7.633228840125392,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011905805377550428,
|
|
"loss": 4.4431,
|
|
"mean_token_accuracy": 0.24534398317337036,
|
|
"num_tokens": 17112427.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 5.315281534194947,
|
|
"epoch": 7.637147335423197,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0001188413635832303,
|
|
"loss": 4.5165,
|
|
"mean_token_accuracy": 0.23658459335565568,
|
|
"num_tokens": 17120890.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 5.327602815628052,
|
|
"epoch": 7.641065830721003,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011862495247175947,
|
|
"loss": 4.3874,
|
|
"mean_token_accuracy": 0.25123307704925535,
|
|
"num_tokens": 17129881.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 5.421972990036011,
|
|
"epoch": 7.644984326018808,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011840882082785357,
|
|
"loss": 4.534,
|
|
"mean_token_accuracy": 0.2411178469657898,
|
|
"num_tokens": 17138900.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.3478100299835205,
|
|
"epoch": 7.648902821316614,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000118192969037775,
|
|
"loss": 4.4983,
|
|
"mean_token_accuracy": 0.24222891628742219,
|
|
"num_tokens": 17147023.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 5.443572568893432,
|
|
"epoch": 7.65282131661442,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011797739748728575,
|
|
"loss": 4.5551,
|
|
"mean_token_accuracy": 0.23083175271749495,
|
|
"num_tokens": 17156926.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 5.396751356124878,
|
|
"epoch": 7.656739811912225,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001177621065616473,
|
|
"loss": 4.5103,
|
|
"mean_token_accuracy": 0.2377547651529312,
|
|
"num_tokens": 17165079.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 5.293580102920532,
|
|
"epoch": 7.660658307210031,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011754709664561931,
|
|
"loss": 4.4436,
|
|
"mean_token_accuracy": 0.24877318888902664,
|
|
"num_tokens": 17173509.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 5.4073498249053955,
|
|
"epoch": 7.664576802507837,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011733236812345943,
|
|
"loss": 4.4977,
|
|
"mean_token_accuracy": 0.2395196422934532,
|
|
"num_tokens": 17182399.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.352904510498047,
|
|
"epoch": 7.668495297805642,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011711792137892221,
|
|
"loss": 4.482,
|
|
"mean_token_accuracy": 0.2389036610722542,
|
|
"num_tokens": 17190733.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 5.334116888046265,
|
|
"epoch": 7.672413793103448,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011690375679525896,
|
|
"loss": 4.4816,
|
|
"mean_token_accuracy": 0.23728373944759368,
|
|
"num_tokens": 17200078.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 5.364469051361084,
|
|
"epoch": 7.6763322884012535,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011668987475521642,
|
|
"loss": 4.5206,
|
|
"mean_token_accuracy": 0.23895889818668364,
|
|
"num_tokens": 17208430.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 5.370899724960327,
|
|
"epoch": 7.6802507836990594,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011647627564103636,
|
|
"loss": 4.5463,
|
|
"mean_token_accuracy": 0.23971864730119705,
|
|
"num_tokens": 17217309.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 5.3540764331817625,
|
|
"epoch": 7.684169278996865,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011626295983445525,
|
|
"loss": 4.4115,
|
|
"mean_token_accuracy": 0.25273463875055313,
|
|
"num_tokens": 17225700.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.280774974822998,
|
|
"epoch": 7.6880877742946705,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011604992771670292,
|
|
"loss": 4.4424,
|
|
"mean_token_accuracy": 0.2432604491710663,
|
|
"num_tokens": 17234582.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 5.240580224990845,
|
|
"epoch": 7.6920062695924765,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011583717966850219,
|
|
"loss": 4.456,
|
|
"mean_token_accuracy": 0.24674883633852004,
|
|
"num_tokens": 17243207.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 5.30293869972229,
|
|
"epoch": 7.695924764890282,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011562471607006845,
|
|
"loss": 4.4301,
|
|
"mean_token_accuracy": 0.2476469725370407,
|
|
"num_tokens": 17251658.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 5.392450284957886,
|
|
"epoch": 7.6998432601880875,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011541253730110866,
|
|
"loss": 4.5742,
|
|
"mean_token_accuracy": 0.23393143564462662,
|
|
"num_tokens": 17261175.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 5.340528202056885,
|
|
"epoch": 7.7037617554858935,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001152006437408206,
|
|
"loss": 4.3854,
|
|
"mean_token_accuracy": 0.25130711644887926,
|
|
"num_tokens": 17269302.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 5.3753856182098385,
|
|
"epoch": 7.7076802507836994,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011498903576789233,
|
|
"loss": 4.5371,
|
|
"mean_token_accuracy": 0.24030940383672714,
|
|
"num_tokens": 17277911.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 5.377471017837524,
|
|
"epoch": 7.7115987460815045,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011477771376050177,
|
|
"loss": 4.4823,
|
|
"mean_token_accuracy": 0.23887639939785005,
|
|
"num_tokens": 17287219.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 5.335998916625977,
|
|
"epoch": 7.7155172413793105,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011456667809631556,
|
|
"loss": 4.4467,
|
|
"mean_token_accuracy": 0.24633740186691283,
|
|
"num_tokens": 17295716.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 5.333749675750733,
|
|
"epoch": 7.7194357366771165,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00011435592915248855,
|
|
"loss": 4.4977,
|
|
"mean_token_accuracy": 0.2378844976425171,
|
|
"num_tokens": 17305870.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.352323389053344,
|
|
"epoch": 7.7233542319749215,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011414546730566342,
|
|
"loss": 4.4704,
|
|
"mean_token_accuracy": 0.2434654116630554,
|
|
"num_tokens": 17314130.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 5.3674328327178955,
|
|
"epoch": 7.7272727272727275,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011393529293196955,
|
|
"loss": 4.3807,
|
|
"mean_token_accuracy": 0.2581719785928726,
|
|
"num_tokens": 17322320.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 5.37611346244812,
|
|
"epoch": 7.731191222570533,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011372540640702263,
|
|
"loss": 4.494,
|
|
"mean_token_accuracy": 0.24809039235115052,
|
|
"num_tokens": 17331162.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.313843727111816,
|
|
"epoch": 7.735109717868339,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00011351580810592382,
|
|
"loss": 4.4038,
|
|
"mean_token_accuracy": 0.24903481602668762,
|
|
"num_tokens": 17339568.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 5.3781829357147215,
|
|
"epoch": 7.7390282131661445,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00011330649840325938,
|
|
"loss": 4.5212,
|
|
"mean_token_accuracy": 0.23818150758743287,
|
|
"num_tokens": 17349354.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 5.33710823059082,
|
|
"epoch": 7.74294670846395,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001130974776730997,
|
|
"loss": 4.5288,
|
|
"mean_token_accuracy": 0.2379002630710602,
|
|
"num_tokens": 17358668.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 5.348646783828736,
|
|
"epoch": 7.746865203761756,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011288874628899864,
|
|
"loss": 4.3965,
|
|
"mean_token_accuracy": 0.2469676837325096,
|
|
"num_tokens": 17366792.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 5.401514768600464,
|
|
"epoch": 7.750783699059561,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011268030462399309,
|
|
"loss": 4.4813,
|
|
"mean_token_accuracy": 0.23957546651363373,
|
|
"num_tokens": 17375589.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 5.4194807529449465,
|
|
"epoch": 7.754702194357367,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00011247215305060213,
|
|
"loss": 4.5637,
|
|
"mean_token_accuracy": 0.24057128131389618,
|
|
"num_tokens": 17385816.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 5.328887939453125,
|
|
"epoch": 7.758620689655173,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011226429194082627,
|
|
"loss": 4.4259,
|
|
"mean_token_accuracy": 0.25116683095693587,
|
|
"num_tokens": 17395267.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 5.374943542480469,
|
|
"epoch": 7.762539184952978,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000112056721666147,
|
|
"loss": 4.5069,
|
|
"mean_token_accuracy": 0.25013367980718615,
|
|
"num_tokens": 17403544.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 5.365045881271362,
|
|
"epoch": 7.766457680250784,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011184944259752622,
|
|
"loss": 4.468,
|
|
"mean_token_accuracy": 0.24823676198720931,
|
|
"num_tokens": 17412090.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 5.338887786865234,
|
|
"epoch": 7.77037617554859,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011164245510540512,
|
|
"loss": 4.4805,
|
|
"mean_token_accuracy": 0.2421285703778267,
|
|
"num_tokens": 17420558.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 5.324377918243409,
|
|
"epoch": 7.774294670846395,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011143575955970384,
|
|
"loss": 4.4927,
|
|
"mean_token_accuracy": 0.24352795481681824,
|
|
"num_tokens": 17429901.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 5.42926983833313,
|
|
"epoch": 7.778213166144201,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011122935632982101,
|
|
"loss": 4.5331,
|
|
"mean_token_accuracy": 0.2354401171207428,
|
|
"num_tokens": 17438543.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 5.470410919189453,
|
|
"epoch": 7.782131661442007,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001110232457846325,
|
|
"loss": 4.5334,
|
|
"mean_token_accuracy": 0.23847686797380446,
|
|
"num_tokens": 17447234.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 5.300828123092652,
|
|
"epoch": 7.786050156739812,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011081742829249142,
|
|
"loss": 4.4102,
|
|
"mean_token_accuracy": 0.2531544104218483,
|
|
"num_tokens": 17455542.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 5.377626609802246,
|
|
"epoch": 7.789968652037618,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011061190422122687,
|
|
"loss": 4.4423,
|
|
"mean_token_accuracy": 0.2507604032754898,
|
|
"num_tokens": 17465035.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 5.37003345489502,
|
|
"epoch": 7.793887147335424,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011040667393814383,
|
|
"loss": 4.4853,
|
|
"mean_token_accuracy": 0.24460896700620652,
|
|
"num_tokens": 17473748.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 5.310074806213379,
|
|
"epoch": 7.797805642633229,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011020173781002199,
|
|
"loss": 4.3712,
|
|
"mean_token_accuracy": 0.2577215403318405,
|
|
"num_tokens": 17481525.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 5.355268716812134,
|
|
"epoch": 7.801724137931035,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001099970962031154,
|
|
"loss": 4.5067,
|
|
"mean_token_accuracy": 0.2393914818763733,
|
|
"num_tokens": 17490044.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 5.415674781799316,
|
|
"epoch": 7.80564263322884,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010979274948315191,
|
|
"loss": 4.545,
|
|
"mean_token_accuracy": 0.2348024994134903,
|
|
"num_tokens": 17498094.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 5.368183183670044,
|
|
"epoch": 7.809561128526646,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010958869801533222,
|
|
"loss": 4.4921,
|
|
"mean_token_accuracy": 0.24397765696048737,
|
|
"num_tokens": 17507677.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 5.35997166633606,
|
|
"epoch": 7.813479623824452,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010938494216432928,
|
|
"loss": 4.453,
|
|
"mean_token_accuracy": 0.24595078974962234,
|
|
"num_tokens": 17516489.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 5.363349676132202,
|
|
"epoch": 7.817398119122257,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010918148229428796,
|
|
"loss": 4.4854,
|
|
"mean_token_accuracy": 0.24310804307460784,
|
|
"num_tokens": 17524972.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 5.353729629516602,
|
|
"epoch": 7.821316614420063,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000108978318768824,
|
|
"loss": 4.4919,
|
|
"mean_token_accuracy": 0.2423836663365364,
|
|
"num_tokens": 17533745.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 5.28822956085205,
|
|
"epoch": 7.825235109717868,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010877545195102347,
|
|
"loss": 4.4678,
|
|
"mean_token_accuracy": 0.23482342362403869,
|
|
"num_tokens": 17542812.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 5.355826711654663,
|
|
"epoch": 7.829153605015674,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010857288220344242,
|
|
"loss": 4.468,
|
|
"mean_token_accuracy": 0.24956086426973342,
|
|
"num_tokens": 17551123.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 5.3634504795074465,
|
|
"epoch": 7.83307210031348,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010837060988810571,
|
|
"loss": 4.5583,
|
|
"mean_token_accuracy": 0.23241806626319886,
|
|
"num_tokens": 17560729.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 5.428134727478027,
|
|
"epoch": 7.836990595611285,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010816863536650687,
|
|
"loss": 4.5331,
|
|
"mean_token_accuracy": 0.2344573587179184,
|
|
"num_tokens": 17569842.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 7.836990595611285,
|
|
"eval_entropy": 5.20120231772578,
|
|
"eval_loss": 5.647326946258545,
|
|
"eval_mean_token_accuracy": 0.17411183262633723,
|
|
"eval_num_tokens": 17569842.0,
|
|
"eval_runtime": 2.3343,
|
|
"eval_samples_per_second": 1765.846,
|
|
"eval_steps_per_second": 221.052,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 5.402283191680908,
|
|
"epoch": 7.840909090909091,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00010796695899960696,
|
|
"loss": 4.4386,
|
|
"mean_token_accuracy": 0.2500694587826729,
|
|
"num_tokens": 17578887.0,
|
|
"step": 10005
|
|
},
|
|
{
|
|
"entropy": 5.371236371994018,
|
|
"epoch": 7.844827586206897,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010776558114783455,
|
|
"loss": 4.4505,
|
|
"mean_token_accuracy": 0.23804078251123428,
|
|
"num_tokens": 17587321.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 5.338368463516235,
|
|
"epoch": 7.848746081504702,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010756450217108441,
|
|
"loss": 4.4519,
|
|
"mean_token_accuracy": 0.24530260413885116,
|
|
"num_tokens": 17596295.0,
|
|
"step": 10015
|
|
},
|
|
{
|
|
"entropy": 5.374414110183716,
|
|
"epoch": 7.852664576802508,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001073637224287172,
|
|
"loss": 4.4515,
|
|
"mean_token_accuracy": 0.2463448762893677,
|
|
"num_tokens": 17605033.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 5.297580337524414,
|
|
"epoch": 7.856583072100314,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010716324227955904,
|
|
"loss": 4.4596,
|
|
"mean_token_accuracy": 0.24218398481607437,
|
|
"num_tokens": 17613706.0,
|
|
"step": 10025
|
|
},
|
|
{
|
|
"entropy": 5.413496589660644,
|
|
"epoch": 7.860501567398119,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00010696306208190034,
|
|
"loss": 4.5279,
|
|
"mean_token_accuracy": 0.24280298501253128,
|
|
"num_tokens": 17622082.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 5.2806469917297365,
|
|
"epoch": 7.864420062695925,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010676318219349564,
|
|
"loss": 4.4079,
|
|
"mean_token_accuracy": 0.25069363564252856,
|
|
"num_tokens": 17630980.0,
|
|
"step": 10035
|
|
},
|
|
{
|
|
"entropy": 5.297147750854492,
|
|
"epoch": 7.868338557993731,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010656360297156262,
|
|
"loss": 4.4458,
|
|
"mean_token_accuracy": 0.2451459437608719,
|
|
"num_tokens": 17639679.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 5.317021274566651,
|
|
"epoch": 7.872257053291536,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001063643247727818,
|
|
"loss": 4.4101,
|
|
"mean_token_accuracy": 0.24923669099807738,
|
|
"num_tokens": 17648542.0,
|
|
"step": 10045
|
|
},
|
|
{
|
|
"entropy": 5.347730255126953,
|
|
"epoch": 7.876175548589342,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010616534795329553,
|
|
"loss": 4.3969,
|
|
"mean_token_accuracy": 0.24997721761465072,
|
|
"num_tokens": 17657305.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 5.3834343433380125,
|
|
"epoch": 7.880094043887147,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00010596667286870778,
|
|
"loss": 4.5033,
|
|
"mean_token_accuracy": 0.238279789686203,
|
|
"num_tokens": 17665876.0,
|
|
"step": 10055
|
|
},
|
|
{
|
|
"entropy": 5.355647802352905,
|
|
"epoch": 7.884012539184953,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010576829987408304,
|
|
"loss": 4.4301,
|
|
"mean_token_accuracy": 0.25094297975301744,
|
|
"num_tokens": 17674756.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 5.35017147064209,
|
|
"epoch": 7.887931034482759,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010557022932394611,
|
|
"loss": 4.402,
|
|
"mean_token_accuracy": 0.26533199697732923,
|
|
"num_tokens": 17683989.0,
|
|
"step": 10065
|
|
},
|
|
{
|
|
"entropy": 5.461822366714477,
|
|
"epoch": 7.891849529780564,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010537246157228104,
|
|
"loss": 4.494,
|
|
"mean_token_accuracy": 0.25128734707832334,
|
|
"num_tokens": 17692776.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 5.366007709503174,
|
|
"epoch": 7.89576802507837,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010517499697253103,
|
|
"loss": 4.3919,
|
|
"mean_token_accuracy": 0.24464598894119263,
|
|
"num_tokens": 17701428.0,
|
|
"step": 10075
|
|
},
|
|
{
|
|
"entropy": 5.328644514083862,
|
|
"epoch": 7.899686520376176,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010497783587759724,
|
|
"loss": 4.5007,
|
|
"mean_token_accuracy": 0.23496186435222627,
|
|
"num_tokens": 17710343.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 5.378355264663696,
|
|
"epoch": 7.903605015673981,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00010478097863983857,
|
|
"loss": 4.488,
|
|
"mean_token_accuracy": 0.24506693780422212,
|
|
"num_tokens": 17719218.0,
|
|
"step": 10085
|
|
},
|
|
{
|
|
"entropy": 5.391655731201172,
|
|
"epoch": 7.907523510971787,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001045844256110707,
|
|
"loss": 4.5211,
|
|
"mean_token_accuracy": 0.23660155832767488,
|
|
"num_tokens": 17727288.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 5.3951763153076175,
|
|
"epoch": 7.911442006269592,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010438817714256589,
|
|
"loss": 4.53,
|
|
"mean_token_accuracy": 0.2354256510734558,
|
|
"num_tokens": 17736130.0,
|
|
"step": 10095
|
|
},
|
|
{
|
|
"entropy": 5.323842859268188,
|
|
"epoch": 7.915360501567398,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010419223358505196,
|
|
"loss": 4.4085,
|
|
"mean_token_accuracy": 0.251513671875,
|
|
"num_tokens": 17745122.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 5.26958966255188,
|
|
"epoch": 7.919278996865204,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010399659528871169,
|
|
"loss": 4.4305,
|
|
"mean_token_accuracy": 0.2473854124546051,
|
|
"num_tokens": 17753817.0,
|
|
"step": 10105
|
|
},
|
|
{
|
|
"entropy": 5.366055107116699,
|
|
"epoch": 7.923197492163009,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010380126260318265,
|
|
"loss": 4.5005,
|
|
"mean_token_accuracy": 0.24229052513837815,
|
|
"num_tokens": 17762793.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 5.281316566467285,
|
|
"epoch": 7.927115987460815,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00010360623587755579,
|
|
"loss": 4.4567,
|
|
"mean_token_accuracy": 0.2465897396206856,
|
|
"num_tokens": 17772223.0,
|
|
"step": 10115
|
|
},
|
|
{
|
|
"entropy": 5.333925342559814,
|
|
"epoch": 7.931034482758621,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001034115154603757,
|
|
"loss": 4.4377,
|
|
"mean_token_accuracy": 0.24358387440443038,
|
|
"num_tokens": 17781543.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 5.34580020904541,
|
|
"epoch": 7.934952978056426,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010321710169963919,
|
|
"loss": 4.4635,
|
|
"mean_token_accuracy": 0.24091559052467346,
|
|
"num_tokens": 17790258.0,
|
|
"step": 10125
|
|
},
|
|
{
|
|
"entropy": 5.313479709625244,
|
|
"epoch": 7.938871473354232,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010302299494279527,
|
|
"loss": 4.4522,
|
|
"mean_token_accuracy": 0.2422591283917427,
|
|
"num_tokens": 17798862.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 5.3611541271209715,
|
|
"epoch": 7.942789968652038,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010282919553674415,
|
|
"loss": 4.5023,
|
|
"mean_token_accuracy": 0.241889089345932,
|
|
"num_tokens": 17807285.0,
|
|
"step": 10135
|
|
},
|
|
{
|
|
"entropy": 5.365668869018554,
|
|
"epoch": 7.946708463949843,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010263570382783675,
|
|
"loss": 4.5131,
|
|
"mean_token_accuracy": 0.23552487790584564,
|
|
"num_tokens": 17816891.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 5.385086250305176,
|
|
"epoch": 7.950626959247649,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010244252016187419,
|
|
"loss": 4.4997,
|
|
"mean_token_accuracy": 0.24144584983587264,
|
|
"num_tokens": 17825895.0,
|
|
"step": 10145
|
|
},
|
|
{
|
|
"entropy": 5.406665372848511,
|
|
"epoch": 7.954545454545455,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00010224964488410699,
|
|
"loss": 4.5772,
|
|
"mean_token_accuracy": 0.23969023674726486,
|
|
"num_tokens": 17834921.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 5.418076038360596,
|
|
"epoch": 7.95846394984326,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001020570783392345,
|
|
"loss": 4.4468,
|
|
"mean_token_accuracy": 0.2525796741247177,
|
|
"num_tokens": 17844310.0,
|
|
"step": 10155
|
|
},
|
|
{
|
|
"entropy": 5.4341919898986815,
|
|
"epoch": 7.962382445141066,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010186482087140439,
|
|
"loss": 4.5044,
|
|
"mean_token_accuracy": 0.2483218178153038,
|
|
"num_tokens": 17853561.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 5.3055885314941404,
|
|
"epoch": 7.966300940438871,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000101672872824212,
|
|
"loss": 4.4329,
|
|
"mean_token_accuracy": 0.23615690171718598,
|
|
"num_tokens": 17863077.0,
|
|
"step": 10165
|
|
},
|
|
{
|
|
"entropy": 5.364435434341431,
|
|
"epoch": 7.970219435736677,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010148123454069943,
|
|
"loss": 4.4027,
|
|
"mean_token_accuracy": 0.24786202907562255,
|
|
"num_tokens": 17871780.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 5.331765842437744,
|
|
"epoch": 7.974137931034483,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001012899063633555,
|
|
"loss": 4.4319,
|
|
"mean_token_accuracy": 0.24527778774499892,
|
|
"num_tokens": 17881144.0,
|
|
"step": 10175
|
|
},
|
|
{
|
|
"entropy": 5.3537064552307125,
|
|
"epoch": 7.978056426332288,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010109888863411473,
|
|
"loss": 4.4689,
|
|
"mean_token_accuracy": 0.24191929250955582,
|
|
"num_tokens": 17890754.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 5.3898804664611815,
|
|
"epoch": 7.981974921630094,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001009081816943567,
|
|
"loss": 4.5511,
|
|
"mean_token_accuracy": 0.23794765770435333,
|
|
"num_tokens": 17899992.0,
|
|
"step": 10185
|
|
},
|
|
{
|
|
"entropy": 5.305447912216186,
|
|
"epoch": 7.985893416927899,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010071778588490558,
|
|
"loss": 4.4434,
|
|
"mean_token_accuracy": 0.2479951113462448,
|
|
"num_tokens": 17907847.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 5.372005987167358,
|
|
"epoch": 7.989811912225705,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010052770154602968,
|
|
"loss": 4.4882,
|
|
"mean_token_accuracy": 0.24114797115325928,
|
|
"num_tokens": 17916566.0,
|
|
"step": 10195
|
|
},
|
|
{
|
|
"entropy": 5.33311653137207,
|
|
"epoch": 7.993730407523511,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010033792901744046,
|
|
"loss": 4.3617,
|
|
"mean_token_accuracy": 0.25517906099557874,
|
|
"num_tokens": 17924969.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 5.354409170150757,
|
|
"epoch": 7.997648902821316,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010014846863829221,
|
|
"loss": 4.3851,
|
|
"mean_token_accuracy": 0.24743763953447342,
|
|
"num_tokens": 17933776.0,
|
|
"step": 10205
|
|
},
|
|
{
|
|
"entropy": 5.286488151550293,
|
|
"epoch": 8.001567398119123,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.995932074718131e-05,
|
|
"loss": 4.3907,
|
|
"mean_token_accuracy": 0.2490192636847496,
|
|
"num_tokens": 17941651.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 5.364273166656494,
|
|
"epoch": 8.005485893416928,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.977048568214578e-05,
|
|
"loss": 4.5313,
|
|
"mean_token_accuracy": 0.24043583422899245,
|
|
"num_tokens": 17951278.0,
|
|
"step": 10215
|
|
},
|
|
{
|
|
"entropy": 5.319476127624512,
|
|
"epoch": 8.009404388714733,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.958196378066444e-05,
|
|
"loss": 4.4197,
|
|
"mean_token_accuracy": 0.2502327531576157,
|
|
"num_tokens": 17960182.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 5.338903856277466,
|
|
"epoch": 8.013322884012538,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.93937553796565e-05,
|
|
"loss": 4.3889,
|
|
"mean_token_accuracy": 0.25357368886470794,
|
|
"num_tokens": 17968221.0,
|
|
"step": 10225
|
|
},
|
|
{
|
|
"entropy": 5.32127013206482,
|
|
"epoch": 8.017241379310345,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.920586081548089e-05,
|
|
"loss": 4.3096,
|
|
"mean_token_accuracy": 0.25324673056602476,
|
|
"num_tokens": 17975724.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 5.304115343093872,
|
|
"epoch": 8.02115987460815,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.901828042393573e-05,
|
|
"loss": 4.4919,
|
|
"mean_token_accuracy": 0.24371366798877717,
|
|
"num_tokens": 17984425.0,
|
|
"step": 10235
|
|
},
|
|
{
|
|
"entropy": 5.298713970184326,
|
|
"epoch": 8.025078369905955,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.88310145402575e-05,
|
|
"loss": 4.3272,
|
|
"mean_token_accuracy": 0.2556553691625595,
|
|
"num_tokens": 17993319.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 5.304515790939331,
|
|
"epoch": 8.028996865203762,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.864406349912081e-05,
|
|
"loss": 4.3644,
|
|
"mean_token_accuracy": 0.25019593834877013,
|
|
"num_tokens": 18002239.0,
|
|
"step": 10245
|
|
},
|
|
{
|
|
"entropy": 5.376190996170044,
|
|
"epoch": 8.032915360501567,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.845742763463739e-05,
|
|
"loss": 4.4629,
|
|
"mean_token_accuracy": 0.2528062015771866,
|
|
"num_tokens": 18010745.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 5.352078247070312,
|
|
"epoch": 8.036833855799372,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.827110728035584e-05,
|
|
"loss": 4.4285,
|
|
"mean_token_accuracy": 0.2519546240568161,
|
|
"num_tokens": 18019535.0,
|
|
"step": 10255
|
|
},
|
|
{
|
|
"entropy": 5.332774496078491,
|
|
"epoch": 8.04075235109718,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 9.808510276926075e-05,
|
|
"loss": 4.4263,
|
|
"mean_token_accuracy": 0.24920347481966018,
|
|
"num_tokens": 18029128.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 5.362235116958618,
|
|
"epoch": 8.044670846394984,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.789941443377249e-05,
|
|
"loss": 4.4469,
|
|
"mean_token_accuracy": 0.24880061745643617,
|
|
"num_tokens": 18038304.0,
|
|
"step": 10265
|
|
},
|
|
{
|
|
"entropy": 5.349996709823609,
|
|
"epoch": 8.04858934169279,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.771404260574613e-05,
|
|
"loss": 4.422,
|
|
"mean_token_accuracy": 0.24895030856132508,
|
|
"num_tokens": 18046628.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 5.397592496871948,
|
|
"epoch": 8.052507836990596,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.752898761647113e-05,
|
|
"loss": 4.4847,
|
|
"mean_token_accuracy": 0.24616074711084365,
|
|
"num_tokens": 18055654.0,
|
|
"step": 10275
|
|
},
|
|
{
|
|
"entropy": 5.31839075088501,
|
|
"epoch": 8.056426332288401,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.734424979667089e-05,
|
|
"loss": 4.4194,
|
|
"mean_token_accuracy": 0.2520749047398567,
|
|
"num_tokens": 18065844.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 5.371987056732178,
|
|
"epoch": 8.060344827586206,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.715982947650179e-05,
|
|
"loss": 4.4619,
|
|
"mean_token_accuracy": 0.2441060170531273,
|
|
"num_tokens": 18074497.0,
|
|
"step": 10285
|
|
},
|
|
{
|
|
"entropy": 5.390768957138062,
|
|
"epoch": 8.064263322884013,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.697572698555278e-05,
|
|
"loss": 4.4557,
|
|
"mean_token_accuracy": 0.24695109128952025,
|
|
"num_tokens": 18083967.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 5.333731174468994,
|
|
"epoch": 8.068181818181818,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.679194265284492e-05,
|
|
"loss": 4.4286,
|
|
"mean_token_accuracy": 0.25146859139204025,
|
|
"num_tokens": 18092263.0,
|
|
"step": 10295
|
|
},
|
|
{
|
|
"entropy": 5.418921566009521,
|
|
"epoch": 8.072100313479623,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.660847680683069e-05,
|
|
"loss": 4.4852,
|
|
"mean_token_accuracy": 0.24598128348588943,
|
|
"num_tokens": 18101559.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 5.400829887390136,
|
|
"epoch": 8.07601880877743,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.64253297753932e-05,
|
|
"loss": 4.4167,
|
|
"mean_token_accuracy": 0.2503548637032509,
|
|
"num_tokens": 18109850.0,
|
|
"step": 10305
|
|
},
|
|
{
|
|
"entropy": 5.368993330001831,
|
|
"epoch": 8.079937304075235,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.62425018858459e-05,
|
|
"loss": 4.4071,
|
|
"mean_token_accuracy": 0.25390606820583345,
|
|
"num_tokens": 18119205.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 5.303245496749878,
|
|
"epoch": 8.08385579937304,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.605999346493194e-05,
|
|
"loss": 4.3515,
|
|
"mean_token_accuracy": 0.2529494076967239,
|
|
"num_tokens": 18128713.0,
|
|
"step": 10315
|
|
},
|
|
{
|
|
"entropy": 5.333392190933227,
|
|
"epoch": 8.087774294670846,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.587780483882339e-05,
|
|
"loss": 4.3706,
|
|
"mean_token_accuracy": 0.2514773041009903,
|
|
"num_tokens": 18137080.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 5.388545131683349,
|
|
"epoch": 8.091692789968652,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.569593633312084e-05,
|
|
"loss": 4.4987,
|
|
"mean_token_accuracy": 0.2436349004507065,
|
|
"num_tokens": 18146449.0,
|
|
"step": 10325
|
|
},
|
|
{
|
|
"entropy": 5.405076169967652,
|
|
"epoch": 8.095611285266457,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.551438827285294e-05,
|
|
"loss": 4.5056,
|
|
"mean_token_accuracy": 0.23766676783561708,
|
|
"num_tokens": 18155394.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 5.379209995269775,
|
|
"epoch": 8.099529780564263,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.533316098247539e-05,
|
|
"loss": 4.5046,
|
|
"mean_token_accuracy": 0.23963469564914702,
|
|
"num_tokens": 18164538.0,
|
|
"step": 10335
|
|
},
|
|
{
|
|
"entropy": 5.375029134750366,
|
|
"epoch": 8.10344827586207,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.515225478587077e-05,
|
|
"loss": 4.4306,
|
|
"mean_token_accuracy": 0.2534207493066788,
|
|
"num_tokens": 18174396.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 5.3628815650939945,
|
|
"epoch": 8.107366771159874,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.497167000634773e-05,
|
|
"loss": 4.4058,
|
|
"mean_token_accuracy": 0.2505332499742508,
|
|
"num_tokens": 18183030.0,
|
|
"step": 10345
|
|
},
|
|
{
|
|
"entropy": 5.3625232696533205,
|
|
"epoch": 8.11128526645768,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.47914069666406e-05,
|
|
"loss": 4.4368,
|
|
"mean_token_accuracy": 0.24289096444845198,
|
|
"num_tokens": 18191139.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 5.288907861709594,
|
|
"epoch": 8.115203761755486,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.461146598890873e-05,
|
|
"loss": 4.3948,
|
|
"mean_token_accuracy": 0.25818361192941663,
|
|
"num_tokens": 18199429.0,
|
|
"step": 10355
|
|
},
|
|
{
|
|
"entropy": 5.345857048034668,
|
|
"epoch": 8.119122257053291,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 9.443184739473573e-05,
|
|
"loss": 4.4582,
|
|
"mean_token_accuracy": 0.24694569259881974,
|
|
"num_tokens": 18207765.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 5.297749471664429,
|
|
"epoch": 8.123040752351097,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 9.425255150512921e-05,
|
|
"loss": 4.4553,
|
|
"mean_token_accuracy": 0.24543828815221785,
|
|
"num_tokens": 18216524.0,
|
|
"step": 10365
|
|
},
|
|
{
|
|
"entropy": 5.393232631683349,
|
|
"epoch": 8.126959247648903,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 9.407357864051996e-05,
|
|
"loss": 4.4316,
|
|
"mean_token_accuracy": 0.25010311752557757,
|
|
"num_tokens": 18224647.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 5.3401425838470455,
|
|
"epoch": 8.130877742946709,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.389492912076156e-05,
|
|
"loss": 4.4226,
|
|
"mean_token_accuracy": 0.24939050674438476,
|
|
"num_tokens": 18234127.0,
|
|
"step": 10375
|
|
},
|
|
{
|
|
"entropy": 5.356734085083008,
|
|
"epoch": 8.134796238244514,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.371660326512952e-05,
|
|
"loss": 4.349,
|
|
"mean_token_accuracy": 0.25291299223899844,
|
|
"num_tokens": 18242455.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 5.396542358398437,
|
|
"epoch": 8.13871473354232,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.353860139232126e-05,
|
|
"loss": 4.4676,
|
|
"mean_token_accuracy": 0.2430454134941101,
|
|
"num_tokens": 18251201.0,
|
|
"step": 10385
|
|
},
|
|
{
|
|
"entropy": 5.363241863250733,
|
|
"epoch": 8.142633228840126,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.336092382045488e-05,
|
|
"loss": 4.4188,
|
|
"mean_token_accuracy": 0.2412657245993614,
|
|
"num_tokens": 18260651.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 5.393657302856445,
|
|
"epoch": 8.14655172413793,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.318357086706894e-05,
|
|
"loss": 4.4755,
|
|
"mean_token_accuracy": 0.25100196450948714,
|
|
"num_tokens": 18269218.0,
|
|
"step": 10395
|
|
},
|
|
{
|
|
"entropy": 5.350926256179809,
|
|
"epoch": 8.150470219435737,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.300654284912211e-05,
|
|
"loss": 4.4288,
|
|
"mean_token_accuracy": 0.24183020740747452,
|
|
"num_tokens": 18277371.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 5.350609970092774,
|
|
"epoch": 8.154388714733543,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.282984008299196e-05,
|
|
"loss": 4.4478,
|
|
"mean_token_accuracy": 0.2432093232870102,
|
|
"num_tokens": 18287503.0,
|
|
"step": 10405
|
|
},
|
|
{
|
|
"entropy": 5.344378423690796,
|
|
"epoch": 8.158307210031348,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 9.265346288447516e-05,
|
|
"loss": 4.4398,
|
|
"mean_token_accuracy": 0.24422506242990494,
|
|
"num_tokens": 18296753.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 5.405035448074341,
|
|
"epoch": 8.162225705329154,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.24774115687862e-05,
|
|
"loss": 4.4701,
|
|
"mean_token_accuracy": 0.2434030517935753,
|
|
"num_tokens": 18305068.0,
|
|
"step": 10415
|
|
},
|
|
{
|
|
"entropy": 5.3639904975891115,
|
|
"epoch": 8.16614420062696,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.23016864505575e-05,
|
|
"loss": 4.4674,
|
|
"mean_token_accuracy": 0.25033932775259016,
|
|
"num_tokens": 18313816.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 5.352625846862793,
|
|
"epoch": 8.170062695924765,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.212628784383825e-05,
|
|
"loss": 4.3976,
|
|
"mean_token_accuracy": 0.25259855538606646,
|
|
"num_tokens": 18322630.0,
|
|
"step": 10425
|
|
},
|
|
{
|
|
"entropy": 5.315604496002197,
|
|
"epoch": 8.17398119122257,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.195121606209411e-05,
|
|
"loss": 4.3781,
|
|
"mean_token_accuracy": 0.2530903860926628,
|
|
"num_tokens": 18330942.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 5.297631549835205,
|
|
"epoch": 8.177899686520377,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.177647141820693e-05,
|
|
"loss": 4.4359,
|
|
"mean_token_accuracy": 0.24657151401042937,
|
|
"num_tokens": 18339375.0,
|
|
"step": 10435
|
|
},
|
|
{
|
|
"entropy": 5.401091241836548,
|
|
"epoch": 8.181818181818182,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.160205422447357e-05,
|
|
"loss": 4.4493,
|
|
"mean_token_accuracy": 0.24811978340148927,
|
|
"num_tokens": 18347884.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 5.329320812225342,
|
|
"epoch": 8.185736677115987,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.142796479260588e-05,
|
|
"loss": 4.4433,
|
|
"mean_token_accuracy": 0.24646312594413758,
|
|
"num_tokens": 18357201.0,
|
|
"step": 10445
|
|
},
|
|
{
|
|
"entropy": 5.390034961700439,
|
|
"epoch": 8.189655172413794,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.125420343372995e-05,
|
|
"loss": 4.5113,
|
|
"mean_token_accuracy": 0.2422666847705841,
|
|
"num_tokens": 18365716.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 5.310503768920898,
|
|
"epoch": 8.193573667711599,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.108077045838546e-05,
|
|
"loss": 4.4099,
|
|
"mean_token_accuracy": 0.25830572843551636,
|
|
"num_tokens": 18374724.0,
|
|
"step": 10455
|
|
},
|
|
{
|
|
"entropy": 5.336271095275879,
|
|
"epoch": 8.197492163009404,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.090766617652525e-05,
|
|
"loss": 4.4468,
|
|
"mean_token_accuracy": 0.2512288153171539,
|
|
"num_tokens": 18383653.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 5.388662195205688,
|
|
"epoch": 8.20141065830721,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.073489089751468e-05,
|
|
"loss": 4.4631,
|
|
"mean_token_accuracy": 0.25155293196439743,
|
|
"num_tokens": 18392729.0,
|
|
"step": 10465
|
|
},
|
|
{
|
|
"entropy": 5.377563619613648,
|
|
"epoch": 8.205329153605016,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.056244493013129e-05,
|
|
"loss": 4.4811,
|
|
"mean_token_accuracy": 0.24001492261886598,
|
|
"num_tokens": 18402257.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 5.374493932723999,
|
|
"epoch": 8.20924764890282,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.039032858256401e-05,
|
|
"loss": 4.4627,
|
|
"mean_token_accuracy": 0.24414742588996888,
|
|
"num_tokens": 18410714.0,
|
|
"step": 10475
|
|
},
|
|
{
|
|
"entropy": 5.35172028541565,
|
|
"epoch": 8.213166144200628,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 9.021854216241251e-05,
|
|
"loss": 4.4757,
|
|
"mean_token_accuracy": 0.24495012313127518,
|
|
"num_tokens": 18419256.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 5.336600494384766,
|
|
"epoch": 8.217084639498433,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.004708597668716e-05,
|
|
"loss": 4.4374,
|
|
"mean_token_accuracy": 0.25062726736068724,
|
|
"num_tokens": 18428202.0,
|
|
"step": 10485
|
|
},
|
|
{
|
|
"entropy": 5.373058414459228,
|
|
"epoch": 8.221003134796238,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.987596033180784e-05,
|
|
"loss": 4.4289,
|
|
"mean_token_accuracy": 0.24316319078207016,
|
|
"num_tokens": 18437339.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 5.3915853023529055,
|
|
"epoch": 8.224921630094045,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.970516553360383e-05,
|
|
"loss": 4.4307,
|
|
"mean_token_accuracy": 0.2500095769762993,
|
|
"num_tokens": 18446667.0,
|
|
"step": 10495
|
|
},
|
|
{
|
|
"entropy": 5.358292102813721,
|
|
"epoch": 8.22884012539185,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 8.953470188731318e-05,
|
|
"loss": 4.3631,
|
|
"mean_token_accuracy": 0.253638057410717,
|
|
"num_tokens": 18455425.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"epoch": 8.22884012539185,
|
|
"eval_entropy": 5.205014356347018,
|
|
"eval_loss": 5.655328750610352,
|
|
"eval_mean_token_accuracy": 0.1735941723075717,
|
|
"eval_num_tokens": 18455425.0,
|
|
"eval_runtime": 2.159,
|
|
"eval_samples_per_second": 1909.183,
|
|
"eval_steps_per_second": 238.995,
|
|
"step": 10500
|
|
}
|
|
],
|
|
"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": 3680866824468480.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|