6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.29773719730051607,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742559337615967,
|
|
"epoch": 0.0004962286621675268,
|
|
"grad_norm": 5.21875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7925,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 11448.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742586231231689,
|
|
"epoch": 0.0009924573243350536,
|
|
"grad_norm": 5.0,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7825,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 22646.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742585563659668,
|
|
"epoch": 0.0014886859865025804,
|
|
"grad_norm": 4.75,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7537,
|
|
"mean_token_accuracy": 0.00020905662677250803,
|
|
"num_tokens": 34011.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742612171173096,
|
|
"epoch": 0.001984914648670107,
|
|
"grad_norm": 4.6875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6968,
|
|
"mean_token_accuracy": 0.0005319510004483163,
|
|
"num_tokens": 44566.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742612838745117,
|
|
"epoch": 0.002481143310837634,
|
|
"grad_norm": 4.3125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6222,
|
|
"mean_token_accuracy": 0.004951164737576619,
|
|
"num_tokens": 56274.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742339897155762,
|
|
"epoch": 0.0029773719730051607,
|
|
"grad_norm": 3.875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4853,
|
|
"mean_token_accuracy": 0.024180484376847743,
|
|
"num_tokens": 66584.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741417789459229,
|
|
"epoch": 0.0034736006351726877,
|
|
"grad_norm": 3.0625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3696,
|
|
"mean_token_accuracy": 0.035252702049911024,
|
|
"num_tokens": 76948.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.738713264465332,
|
|
"epoch": 0.003969829297340214,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2243,
|
|
"mean_token_accuracy": 0.03839292954653502,
|
|
"num_tokens": 87253.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.734406661987304,
|
|
"epoch": 0.004466057959507741,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.1158,
|
|
"mean_token_accuracy": 0.04111976623535156,
|
|
"num_tokens": 98273.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.730075454711914,
|
|
"epoch": 0.004962286621675268,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.0069,
|
|
"mean_token_accuracy": 0.04187291283160448,
|
|
"num_tokens": 107788.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.727989768981933,
|
|
"epoch": 0.0054585152838427945,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9971,
|
|
"mean_token_accuracy": 0.03726662136614323,
|
|
"num_tokens": 118632.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.727395725250243,
|
|
"epoch": 0.0059547439460103215,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8873,
|
|
"mean_token_accuracy": 0.04101524651050568,
|
|
"num_tokens": 128649.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.725363159179688,
|
|
"epoch": 0.0064509726081778485,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8548,
|
|
"mean_token_accuracy": 0.04273284543305635,
|
|
"num_tokens": 138958.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.721133422851562,
|
|
"epoch": 0.0069472012703453755,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8217,
|
|
"mean_token_accuracy": 0.04243846274912357,
|
|
"num_tokens": 151445.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.715934658050537,
|
|
"epoch": 0.007443429932512902,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7492,
|
|
"mean_token_accuracy": 0.043674266710877416,
|
|
"num_tokens": 162239.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.70990047454834,
|
|
"epoch": 0.007939658594680429,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.7155,
|
|
"mean_token_accuracy": 0.03882305920124054,
|
|
"num_tokens": 173261.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.70549554824829,
|
|
"epoch": 0.008435887256847956,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.6222,
|
|
"mean_token_accuracy": 0.037595166079699996,
|
|
"num_tokens": 183857.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.701297950744628,
|
|
"epoch": 0.008932115919015483,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.646,
|
|
"mean_token_accuracy": 0.03864660933613777,
|
|
"num_tokens": 196635.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.692024230957031,
|
|
"epoch": 0.00942834458118301,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4935,
|
|
"mean_token_accuracy": 0.04372961949557066,
|
|
"num_tokens": 208700.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.67728853225708,
|
|
"epoch": 0.009924573243350537,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3747,
|
|
"mean_token_accuracy": 0.04096841104328632,
|
|
"num_tokens": 218915.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.659735870361327,
|
|
"epoch": 0.010420801905518062,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2746,
|
|
"mean_token_accuracy": 0.04906293600797653,
|
|
"num_tokens": 230211.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.6349609375,
|
|
"epoch": 0.010917030567685589,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1754,
|
|
"mean_token_accuracy": 0.05304342210292816,
|
|
"num_tokens": 241174.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.604046154022218,
|
|
"epoch": 0.011413259229853116,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.1068,
|
|
"mean_token_accuracy": 0.04802950657904148,
|
|
"num_tokens": 251708.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.571563243865967,
|
|
"epoch": 0.011909487892020643,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.1035,
|
|
"mean_token_accuracy": 0.04602174609899521,
|
|
"num_tokens": 263287.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.524694061279297,
|
|
"epoch": 0.01240571655418817,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.9308,
|
|
"mean_token_accuracy": 0.04466274194419384,
|
|
"num_tokens": 275179.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.459578609466552,
|
|
"epoch": 0.012901945216355697,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.731,
|
|
"mean_token_accuracy": 0.049411849677562715,
|
|
"num_tokens": 287223.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.371681499481202,
|
|
"epoch": 0.013398173878523224,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.6072,
|
|
"mean_token_accuracy": 0.04893956743180752,
|
|
"num_tokens": 297868.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.280974102020263,
|
|
"epoch": 0.013894402540690751,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.5167,
|
|
"mean_token_accuracy": 0.05017189048230648,
|
|
"num_tokens": 309702.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.183215141296387,
|
|
"epoch": 0.014390631202858276,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.483,
|
|
"mean_token_accuracy": 0.04685804210603237,
|
|
"num_tokens": 321613.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.04955472946167,
|
|
"epoch": 0.014886859865025803,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.2837,
|
|
"mean_token_accuracy": 0.053469978272914886,
|
|
"num_tokens": 332668.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.918639755249023,
|
|
"epoch": 0.01538308852719333,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.2455,
|
|
"mean_token_accuracy": 0.054456012323498726,
|
|
"num_tokens": 342579.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.765514373779297,
|
|
"epoch": 0.015879317189360857,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.0663,
|
|
"mean_token_accuracy": 0.054721662402153017,
|
|
"num_tokens": 352271.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.54785737991333,
|
|
"epoch": 0.016375545851528384,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.0234,
|
|
"mean_token_accuracy": 0.0642885323613882,
|
|
"num_tokens": 364033.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.348752498626709,
|
|
"epoch": 0.01687177451369591,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.9279,
|
|
"mean_token_accuracy": 0.06007566004991531,
|
|
"num_tokens": 373788.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.17256269454956,
|
|
"epoch": 0.01736800317586344,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8747,
|
|
"mean_token_accuracy": 0.06030374839901924,
|
|
"num_tokens": 384883.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.98403024673462,
|
|
"epoch": 0.017864231838030965,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7558,
|
|
"mean_token_accuracy": 0.06696737073361873,
|
|
"num_tokens": 396044.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.754393672943115,
|
|
"epoch": 0.018360460500198492,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.8178,
|
|
"mean_token_accuracy": 0.06663747020065784,
|
|
"num_tokens": 406995.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.613575267791749,
|
|
"epoch": 0.01885668916236602,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.6515,
|
|
"mean_token_accuracy": 0.06824805773794651,
|
|
"num_tokens": 417238.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.565249347686768,
|
|
"epoch": 0.019352917824533546,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.7144,
|
|
"mean_token_accuracy": 0.06267430260777473,
|
|
"num_tokens": 427263.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.521870040893555,
|
|
"epoch": 0.019849146486701073,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.6025,
|
|
"mean_token_accuracy": 0.06824213974177837,
|
|
"num_tokens": 438357.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.434043884277344,
|
|
"epoch": 0.0203453751488686,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.6098,
|
|
"mean_token_accuracy": 0.06504783928394317,
|
|
"num_tokens": 448738.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.3219557762146,
|
|
"epoch": 0.020841603811036124,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.6115,
|
|
"mean_token_accuracy": 0.07722424305975437,
|
|
"num_tokens": 457873.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.32616958618164,
|
|
"epoch": 0.02133783247320365,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.5469,
|
|
"mean_token_accuracy": 0.0731696669012308,
|
|
"num_tokens": 468640.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.25720796585083,
|
|
"epoch": 0.021834061135371178,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.5443,
|
|
"mean_token_accuracy": 0.07317286394536496,
|
|
"num_tokens": 479922.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.193887042999268,
|
|
"epoch": 0.022330289797538705,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.5043,
|
|
"mean_token_accuracy": 0.06673288382589818,
|
|
"num_tokens": 491056.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.124149227142334,
|
|
"epoch": 0.022826518459706232,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.5857,
|
|
"mean_token_accuracy": 0.07121467962861061,
|
|
"num_tokens": 504365.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.186816024780274,
|
|
"epoch": 0.02332274712187376,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.4841,
|
|
"mean_token_accuracy": 0.06768220961093903,
|
|
"num_tokens": 514347.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.151187849044799,
|
|
"epoch": 0.023818975784041286,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.4982,
|
|
"mean_token_accuracy": 0.07583453841507434,
|
|
"num_tokens": 525912.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.13270378112793,
|
|
"epoch": 0.024315204446208813,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3789,
|
|
"mean_token_accuracy": 0.0720100063830614,
|
|
"num_tokens": 535566.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.082850551605224,
|
|
"epoch": 0.02481143310837634,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.4273,
|
|
"mean_token_accuracy": 0.07373461574316025,
|
|
"num_tokens": 546474.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.037993764877319,
|
|
"epoch": 0.025307661770543867,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.512,
|
|
"mean_token_accuracy": 0.06475581601262093,
|
|
"num_tokens": 558132.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.116750812530517,
|
|
"epoch": 0.025803890432711394,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.4896,
|
|
"mean_token_accuracy": 0.07219315581023693,
|
|
"num_tokens": 568309.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.967530822753906,
|
|
"epoch": 0.02630011909487892,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.3878,
|
|
"mean_token_accuracy": 0.07423482574522496,
|
|
"num_tokens": 579701.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.024831199645996,
|
|
"epoch": 0.026796347757046448,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3601,
|
|
"mean_token_accuracy": 0.0777493104338646,
|
|
"num_tokens": 590832.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.985195302963257,
|
|
"epoch": 0.027292576419213975,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.3025,
|
|
"mean_token_accuracy": 0.07602775730192661,
|
|
"num_tokens": 601094.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.902724742889404,
|
|
"epoch": 0.027788805081381502,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.3129,
|
|
"mean_token_accuracy": 0.08000940680503846,
|
|
"num_tokens": 612027.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.878976964950562,
|
|
"epoch": 0.02828503374354903,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.3466,
|
|
"mean_token_accuracy": 0.07674156315624714,
|
|
"num_tokens": 623698.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.936695146560669,
|
|
"epoch": 0.028781262405716553,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.359,
|
|
"mean_token_accuracy": 0.07654249593615532,
|
|
"num_tokens": 635631.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.894163751602173,
|
|
"epoch": 0.02927749106788408,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.3397,
|
|
"mean_token_accuracy": 0.0750736203044653,
|
|
"num_tokens": 647663.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.829724407196045,
|
|
"epoch": 0.029773719730051607,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.3044,
|
|
"mean_token_accuracy": 0.07420375496149063,
|
|
"num_tokens": 659352.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.924322032928467,
|
|
"epoch": 0.030269948392219134,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2742,
|
|
"mean_token_accuracy": 0.08252886533737183,
|
|
"num_tokens": 669379.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.784340810775757,
|
|
"epoch": 0.03076617705438666,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2618,
|
|
"mean_token_accuracy": 0.0796934574842453,
|
|
"num_tokens": 680144.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.935956048965454,
|
|
"epoch": 0.03126240571655419,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.2152,
|
|
"mean_token_accuracy": 0.07189621031284332,
|
|
"num_tokens": 691165.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.792111110687256,
|
|
"epoch": 0.031758634378721715,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2931,
|
|
"mean_token_accuracy": 0.07880425304174424,
|
|
"num_tokens": 703767.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.7711742401123045,
|
|
"epoch": 0.032254863040889245,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2808,
|
|
"mean_token_accuracy": 0.08353963568806648,
|
|
"num_tokens": 714298.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.765182256698608,
|
|
"epoch": 0.03275109170305677,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2558,
|
|
"mean_token_accuracy": 0.08778629601001739,
|
|
"num_tokens": 724135.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.761576843261719,
|
|
"epoch": 0.03324732036522429,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.3084,
|
|
"mean_token_accuracy": 0.07605698890984058,
|
|
"num_tokens": 735050.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.796260786056519,
|
|
"epoch": 0.03374354902739182,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2195,
|
|
"mean_token_accuracy": 0.0794803261756897,
|
|
"num_tokens": 746073.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.7471351146698,
|
|
"epoch": 0.034239777689559346,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.2147,
|
|
"mean_token_accuracy": 0.07722761482000351,
|
|
"num_tokens": 758998.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.7211285591125485,
|
|
"epoch": 0.03473600635172688,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1516,
|
|
"mean_token_accuracy": 0.07914387956261634,
|
|
"num_tokens": 771608.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.7531922340393065,
|
|
"epoch": 0.0352322350138944,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.2202,
|
|
"mean_token_accuracy": 0.07867058292031288,
|
|
"num_tokens": 782394.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.697312593460083,
|
|
"epoch": 0.03572846367606193,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.122,
|
|
"mean_token_accuracy": 0.08720502257347107,
|
|
"num_tokens": 792389.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.740397548675537,
|
|
"epoch": 0.036224692338229454,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.1616,
|
|
"mean_token_accuracy": 0.08682659715414047,
|
|
"num_tokens": 803219.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.670851850509644,
|
|
"epoch": 0.036720921000396985,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.2297,
|
|
"mean_token_accuracy": 0.08005850054323674,
|
|
"num_tokens": 814249.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.705906867980957,
|
|
"epoch": 0.03721714966256451,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.1186,
|
|
"mean_token_accuracy": 0.08939927890896797,
|
|
"num_tokens": 825138.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.6558843612670895,
|
|
"epoch": 0.03771337832473204,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1626,
|
|
"mean_token_accuracy": 0.08431188724935054,
|
|
"num_tokens": 836024.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.675263166427612,
|
|
"epoch": 0.03820960698689956,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.1514,
|
|
"mean_token_accuracy": 0.08133338317275048,
|
|
"num_tokens": 847765.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.624440860748291,
|
|
"epoch": 0.03870583564906709,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.1454,
|
|
"mean_token_accuracy": 0.08270616680383683,
|
|
"num_tokens": 859980.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.753679609298706,
|
|
"epoch": 0.039202064311234616,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1381,
|
|
"mean_token_accuracy": 0.08271029070019723,
|
|
"num_tokens": 871514.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.651297092437744,
|
|
"epoch": 0.03969829297340215,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1347,
|
|
"mean_token_accuracy": 0.08268468752503395,
|
|
"num_tokens": 881984.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.617497539520263,
|
|
"epoch": 0.04019452163556967,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0907,
|
|
"mean_token_accuracy": 0.0867445409297943,
|
|
"num_tokens": 894198.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.598975372314453,
|
|
"epoch": 0.0406907502977372,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.1054,
|
|
"mean_token_accuracy": 0.08495021611452103,
|
|
"num_tokens": 905394.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.668603801727295,
|
|
"epoch": 0.041186978959904724,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.1566,
|
|
"mean_token_accuracy": 0.07967745214700699,
|
|
"num_tokens": 917010.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.577554368972779,
|
|
"epoch": 0.04168320762207225,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.988,
|
|
"mean_token_accuracy": 0.09206651523709297,
|
|
"num_tokens": 926761.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.479721593856811,
|
|
"epoch": 0.04217943628423978,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.9942,
|
|
"mean_token_accuracy": 0.08749158009886741,
|
|
"num_tokens": 938385.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.532217884063721,
|
|
"epoch": 0.0426756649464073,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0277,
|
|
"mean_token_accuracy": 0.08903890922665596,
|
|
"num_tokens": 949709.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.584686422348023,
|
|
"epoch": 0.04317189360857483,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.0589,
|
|
"mean_token_accuracy": 0.0841300755739212,
|
|
"num_tokens": 960837.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.509622049331665,
|
|
"epoch": 0.043668122270742356,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8948,
|
|
"mean_token_accuracy": 0.09134786874055863,
|
|
"num_tokens": 971036.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.528602838516235,
|
|
"epoch": 0.044164350932909886,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0396,
|
|
"mean_token_accuracy": 0.08592119216918945,
|
|
"num_tokens": 983057.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.493656635284424,
|
|
"epoch": 0.04466057959507741,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9887,
|
|
"mean_token_accuracy": 0.09124278873205185,
|
|
"num_tokens": 994303.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.4633971691131595,
|
|
"epoch": 0.04515680825724494,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.9794,
|
|
"mean_token_accuracy": 0.09112128540873528,
|
|
"num_tokens": 1005033.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.55339617729187,
|
|
"epoch": 0.045653036919412464,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.0539,
|
|
"mean_token_accuracy": 0.0886804811656475,
|
|
"num_tokens": 1015940.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.4837071895599365,
|
|
"epoch": 0.046149265581579994,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0009,
|
|
"mean_token_accuracy": 0.0916803628206253,
|
|
"num_tokens": 1027582.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.472231149673462,
|
|
"epoch": 0.04664549424374752,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.9737,
|
|
"mean_token_accuracy": 0.08888428881764412,
|
|
"num_tokens": 1040046.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.467594766616822,
|
|
"epoch": 0.04714172290591505,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.0423,
|
|
"mean_token_accuracy": 0.08253751546144486,
|
|
"num_tokens": 1050636.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.522167110443116,
|
|
"epoch": 0.04763795156808257,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9296,
|
|
"mean_token_accuracy": 0.09664900973439217,
|
|
"num_tokens": 1060748.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.500424242019653,
|
|
"epoch": 0.0481341802302501,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.9719,
|
|
"mean_token_accuracy": 0.08677386716008187,
|
|
"num_tokens": 1071940.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.417007827758789,
|
|
"epoch": 0.048630408892417626,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.9827,
|
|
"mean_token_accuracy": 0.08790183737874031,
|
|
"num_tokens": 1082962.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.444004106521606,
|
|
"epoch": 0.04912663755458515,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.96,
|
|
"mean_token_accuracy": 0.09731238633394242,
|
|
"num_tokens": 1093082.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.518671083450317,
|
|
"epoch": 0.04962286621675268,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.0303,
|
|
"mean_token_accuracy": 0.09159182384610176,
|
|
"num_tokens": 1103528.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.388113021850586,
|
|
"epoch": 0.0501190948789202,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.9133,
|
|
"mean_token_accuracy": 0.08871109336614609,
|
|
"num_tokens": 1114813.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.351367425918579,
|
|
"epoch": 0.050615323541087734,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.9058,
|
|
"mean_token_accuracy": 0.08876910582184791,
|
|
"num_tokens": 1125345.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.353222036361695,
|
|
"epoch": 0.05111155220325526,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9205,
|
|
"mean_token_accuracy": 0.09047875106334687,
|
|
"num_tokens": 1134985.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.4535609722137455,
|
|
"epoch": 0.05160778086542279,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.9339,
|
|
"mean_token_accuracy": 0.09264354780316353,
|
|
"num_tokens": 1145796.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.47781138420105,
|
|
"epoch": 0.05210400952759031,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.0635,
|
|
"mean_token_accuracy": 0.08175969608128071,
|
|
"num_tokens": 1156936.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.3812775135040285,
|
|
"epoch": 0.05260023818975784,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.9329,
|
|
"mean_token_accuracy": 0.09363158792257309,
|
|
"num_tokens": 1167526.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.422448301315308,
|
|
"epoch": 0.053096466851925365,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.9746,
|
|
"mean_token_accuracy": 0.08542074412107467,
|
|
"num_tokens": 1177647.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.395486497879029,
|
|
"epoch": 0.053592695514092896,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.9292,
|
|
"mean_token_accuracy": 0.09526413604617119,
|
|
"num_tokens": 1188255.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.320084714889527,
|
|
"epoch": 0.05408892417626042,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8533,
|
|
"mean_token_accuracy": 0.09879972338676453,
|
|
"num_tokens": 1199433.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.263059377670288,
|
|
"epoch": 0.05458515283842795,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.9302,
|
|
"mean_token_accuracy": 0.0863654799759388,
|
|
"num_tokens": 1210067.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.307140684127807,
|
|
"epoch": 0.055081381500595473,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.8185,
|
|
"mean_token_accuracy": 0.0939325712621212,
|
|
"num_tokens": 1220639.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.349351215362549,
|
|
"epoch": 0.055577610162763004,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.861,
|
|
"mean_token_accuracy": 0.09410361275076866,
|
|
"num_tokens": 1233158.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.274934434890747,
|
|
"epoch": 0.05607383882493053,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.8381,
|
|
"mean_token_accuracy": 0.08947066888213158,
|
|
"num_tokens": 1244552.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.424350881576538,
|
|
"epoch": 0.05657006748709806,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.9567,
|
|
"mean_token_accuracy": 0.08916749060153961,
|
|
"num_tokens": 1255478.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.213148832321167,
|
|
"epoch": 0.05706629614926558,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7664,
|
|
"mean_token_accuracy": 0.09938948079943657,
|
|
"num_tokens": 1267507.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.141163396835327,
|
|
"epoch": 0.057562524811433105,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7759,
|
|
"mean_token_accuracy": 0.09177131280303001,
|
|
"num_tokens": 1279377.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.411661386489868,
|
|
"epoch": 0.058058753473600636,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.8635,
|
|
"mean_token_accuracy": 0.10038365721702576,
|
|
"num_tokens": 1290671.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.1382372856140135,
|
|
"epoch": 0.05855498213576816,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8309,
|
|
"mean_token_accuracy": 0.10017411112785339,
|
|
"num_tokens": 1302792.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.324809122085571,
|
|
"epoch": 0.05905121079793569,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7832,
|
|
"mean_token_accuracy": 0.08957966193556785,
|
|
"num_tokens": 1313315.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.144257354736328,
|
|
"epoch": 0.05954743946010321,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.8465,
|
|
"mean_token_accuracy": 0.0927205927670002,
|
|
"num_tokens": 1326358.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.283497524261475,
|
|
"epoch": 0.060043668122270744,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.7205,
|
|
"mean_token_accuracy": 0.09314653351902961,
|
|
"num_tokens": 1336266.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.164820337295533,
|
|
"epoch": 0.06053989678443827,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6683,
|
|
"mean_token_accuracy": 0.09883529841899871,
|
|
"num_tokens": 1347610.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.1178053855896,
|
|
"epoch": 0.0610361254466058,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6337,
|
|
"mean_token_accuracy": 0.10055704191327094,
|
|
"num_tokens": 1358317.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.149615907669068,
|
|
"epoch": 0.06153235410877332,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7493,
|
|
"mean_token_accuracy": 0.09263555109500884,
|
|
"num_tokens": 1370126.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.162906694412231,
|
|
"epoch": 0.06202858277094085,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7092,
|
|
"mean_token_accuracy": 0.09986358359456063,
|
|
"num_tokens": 1381571.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.126469230651855,
|
|
"epoch": 0.06252481143310838,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7397,
|
|
"mean_token_accuracy": 0.09763761013746261,
|
|
"num_tokens": 1392486.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.15671968460083,
|
|
"epoch": 0.0630210400952759,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.751,
|
|
"mean_token_accuracy": 0.09868052825331688,
|
|
"num_tokens": 1403086.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.13517165184021,
|
|
"epoch": 0.06351726875744343,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7055,
|
|
"mean_token_accuracy": 0.09947170913219452,
|
|
"num_tokens": 1414537.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.06722993850708,
|
|
"epoch": 0.06401349741961096,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.726,
|
|
"mean_token_accuracy": 0.09579051956534386,
|
|
"num_tokens": 1426998.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.146437740325927,
|
|
"epoch": 0.06450972608177849,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7544,
|
|
"mean_token_accuracy": 0.09660795778036117,
|
|
"num_tokens": 1437667.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.114442014694214,
|
|
"epoch": 0.065005954743946,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6146,
|
|
"mean_token_accuracy": 0.10034202635288239,
|
|
"num_tokens": 1448280.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.085196685791016,
|
|
"epoch": 0.06550218340611354,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.6777,
|
|
"mean_token_accuracy": 0.09931168034672737,
|
|
"num_tokens": 1458608.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.150270652770996,
|
|
"epoch": 0.06599841206828107,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.8578,
|
|
"mean_token_accuracy": 0.09408899918198585,
|
|
"num_tokens": 1470435.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.095894765853882,
|
|
"epoch": 0.06649464073044858,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.7049,
|
|
"mean_token_accuracy": 0.098654755204916,
|
|
"num_tokens": 1481959.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.1083564281463625,
|
|
"epoch": 0.06699086939261611,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.7581,
|
|
"mean_token_accuracy": 0.09214193895459175,
|
|
"num_tokens": 1493798.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.06939730644226,
|
|
"epoch": 0.06748709805478365,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5739,
|
|
"mean_token_accuracy": 0.09953450039029121,
|
|
"num_tokens": 1502969.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.057254981994629,
|
|
"epoch": 0.06798332671695118,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.7168,
|
|
"mean_token_accuracy": 0.09586154073476791,
|
|
"num_tokens": 1513825.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.048639583587646,
|
|
"epoch": 0.06847955537911869,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6118,
|
|
"mean_token_accuracy": 0.10043377801775932,
|
|
"num_tokens": 1525328.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.052638721466065,
|
|
"epoch": 0.06897578404128622,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6426,
|
|
"mean_token_accuracy": 0.09776772186160088,
|
|
"num_tokens": 1537383.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.164321613311768,
|
|
"epoch": 0.06947201270345375,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.7385,
|
|
"mean_token_accuracy": 0.1011982411146164,
|
|
"num_tokens": 1549671.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.959526395797729,
|
|
"epoch": 0.06996824136562128,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.623,
|
|
"mean_token_accuracy": 0.09803913906216621,
|
|
"num_tokens": 1559974.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.045153188705444,
|
|
"epoch": 0.0704644700277888,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.725,
|
|
"mean_token_accuracy": 0.09499528259038925,
|
|
"num_tokens": 1570890.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.018730640411377,
|
|
"epoch": 0.07096069868995633,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.622,
|
|
"mean_token_accuracy": 0.10477519556879997,
|
|
"num_tokens": 1581434.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.0257142066955565,
|
|
"epoch": 0.07145692735212386,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6871,
|
|
"mean_token_accuracy": 0.09542187601327896,
|
|
"num_tokens": 1593710.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.100641775131225,
|
|
"epoch": 0.07195315601429139,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6523,
|
|
"mean_token_accuracy": 0.10131741389632225,
|
|
"num_tokens": 1604635.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.126769018173218,
|
|
"epoch": 0.07244938467645891,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.7134,
|
|
"mean_token_accuracy": 0.10091805458068848,
|
|
"num_tokens": 1615258.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.9821710109710695,
|
|
"epoch": 0.07294561333862644,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6178,
|
|
"mean_token_accuracy": 0.09964116811752319,
|
|
"num_tokens": 1625942.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.951982069015503,
|
|
"epoch": 0.07344184200079397,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5998,
|
|
"mean_token_accuracy": 0.105032167583704,
|
|
"num_tokens": 1636270.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.968512439727784,
|
|
"epoch": 0.07393807066296149,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.5799,
|
|
"mean_token_accuracy": 0.10235915258526802,
|
|
"num_tokens": 1647029.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.01955018043518,
|
|
"epoch": 0.07443429932512902,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7457,
|
|
"mean_token_accuracy": 0.0975896142423153,
|
|
"num_tokens": 1659483.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.89774866104126,
|
|
"epoch": 0.07493052798729655,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6003,
|
|
"mean_token_accuracy": 0.0963432714343071,
|
|
"num_tokens": 1669425.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.036856985092163,
|
|
"epoch": 0.07542675664946408,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6063,
|
|
"mean_token_accuracy": 0.09760197177529335,
|
|
"num_tokens": 1680426.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.869618892669678,
|
|
"epoch": 0.0759229853116316,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6246,
|
|
"mean_token_accuracy": 0.09728450551629067,
|
|
"num_tokens": 1692803.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.04201602935791,
|
|
"epoch": 0.07641921397379912,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.6441,
|
|
"mean_token_accuracy": 0.09765328615903854,
|
|
"num_tokens": 1703919.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.969884634017944,
|
|
"epoch": 0.07691544263596665,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.6147,
|
|
"mean_token_accuracy": 0.10524557158350945,
|
|
"num_tokens": 1714487.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.979475402832032,
|
|
"epoch": 0.07741167129813419,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6626,
|
|
"mean_token_accuracy": 0.09930093437433243,
|
|
"num_tokens": 1726601.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.933769798278808,
|
|
"epoch": 0.0779078999603017,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.554,
|
|
"mean_token_accuracy": 0.10119575038552284,
|
|
"num_tokens": 1738743.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.9056957244873045,
|
|
"epoch": 0.07840412862246923,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.6345,
|
|
"mean_token_accuracy": 0.10144512727856636,
|
|
"num_tokens": 1748286.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.91564908027649,
|
|
"epoch": 0.07890035728463676,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.6636,
|
|
"mean_token_accuracy": 0.09903955459594727,
|
|
"num_tokens": 1758917.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.979661178588867,
|
|
"epoch": 0.0793965859468043,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4992,
|
|
"mean_token_accuracy": 0.10489165559411048,
|
|
"num_tokens": 1769179.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.849152421951294,
|
|
"epoch": 0.07989281460897181,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.6226,
|
|
"mean_token_accuracy": 0.09758250415325165,
|
|
"num_tokens": 1782369.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.871382761001587,
|
|
"epoch": 0.08038904327113934,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.6281,
|
|
"mean_token_accuracy": 0.10324807316064835,
|
|
"num_tokens": 1795992.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.90522723197937,
|
|
"epoch": 0.08088527193330687,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5238,
|
|
"mean_token_accuracy": 0.10905546322464943,
|
|
"num_tokens": 1806938.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.812549114227295,
|
|
"epoch": 0.0813815005954744,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4573,
|
|
"mean_token_accuracy": 0.10598495304584503,
|
|
"num_tokens": 1817663.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.936794233322144,
|
|
"epoch": 0.08187772925764192,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6127,
|
|
"mean_token_accuracy": 0.10083081871271134,
|
|
"num_tokens": 1828254.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.896241903305054,
|
|
"epoch": 0.08237395791980945,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5339,
|
|
"mean_token_accuracy": 0.10185903310775757,
|
|
"num_tokens": 1839001.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.864541244506836,
|
|
"epoch": 0.08287018658197698,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.6313,
|
|
"mean_token_accuracy": 0.09963079243898391,
|
|
"num_tokens": 1849906.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.797179365158081,
|
|
"epoch": 0.0833664152441445,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5266,
|
|
"mean_token_accuracy": 0.09986047223210334,
|
|
"num_tokens": 1860982.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.9268652439117435,
|
|
"epoch": 0.08386264390631203,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4752,
|
|
"mean_token_accuracy": 0.10353965610265732,
|
|
"num_tokens": 1872177.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.7672933578491214,
|
|
"epoch": 0.08435887256847956,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.455,
|
|
"mean_token_accuracy": 0.10329471975564956,
|
|
"num_tokens": 1882783.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.962534761428833,
|
|
"epoch": 0.08485510123064709,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.6426,
|
|
"mean_token_accuracy": 0.09471128433942795,
|
|
"num_tokens": 1893706.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.8037285804748535,
|
|
"epoch": 0.0853513298928146,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5138,
|
|
"mean_token_accuracy": 0.10021722540259362,
|
|
"num_tokens": 1903852.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.835017728805542,
|
|
"epoch": 0.08584755855498213,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.5342,
|
|
"mean_token_accuracy": 0.10754364952445031,
|
|
"num_tokens": 1914157.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.85911717414856,
|
|
"epoch": 0.08634378721714966,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4726,
|
|
"mean_token_accuracy": 0.10122424364089966,
|
|
"num_tokens": 1924684.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.886322259902954,
|
|
"epoch": 0.0868400158793172,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.6263,
|
|
"mean_token_accuracy": 0.09748353064060211,
|
|
"num_tokens": 1935657.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.7788018703460695,
|
|
"epoch": 0.08733624454148471,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5028,
|
|
"mean_token_accuracy": 0.10384351089596748,
|
|
"num_tokens": 1946735.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.872425079345703,
|
|
"epoch": 0.08783247320365224,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.6691,
|
|
"mean_token_accuracy": 0.09842884838581085,
|
|
"num_tokens": 1957871.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.7883913040161135,
|
|
"epoch": 0.08832870186581977,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.5061,
|
|
"mean_token_accuracy": 0.10142220854759217,
|
|
"num_tokens": 1969032.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.859635829925537,
|
|
"epoch": 0.0888249305279873,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5319,
|
|
"mean_token_accuracy": 0.10258013382554054,
|
|
"num_tokens": 1980212.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.746867752075195,
|
|
"epoch": 0.08932115919015482,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4491,
|
|
"mean_token_accuracy": 0.10580305159091949,
|
|
"num_tokens": 1990333.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.797193908691407,
|
|
"epoch": 0.08981738785232235,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5216,
|
|
"mean_token_accuracy": 0.10975843593478203,
|
|
"num_tokens": 2001486.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.801945495605469,
|
|
"epoch": 0.09031361651448988,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5084,
|
|
"mean_token_accuracy": 0.10368723571300506,
|
|
"num_tokens": 2012535.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.765622854232788,
|
|
"epoch": 0.0908098451766574,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4628,
|
|
"mean_token_accuracy": 0.10697698220610619,
|
|
"num_tokens": 2024116.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.651122283935547,
|
|
"epoch": 0.09130607383882493,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5028,
|
|
"mean_token_accuracy": 0.10319509655237198,
|
|
"num_tokens": 2036341.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.888702630996704,
|
|
"epoch": 0.09180230250099246,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4747,
|
|
"mean_token_accuracy": 0.10404421612620354,
|
|
"num_tokens": 2046768.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.6753270626068115,
|
|
"epoch": 0.09229853116315999,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4364,
|
|
"mean_token_accuracy": 0.10831291154026985,
|
|
"num_tokens": 2057315.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.732809257507324,
|
|
"epoch": 0.0927947598253275,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4229,
|
|
"mean_token_accuracy": 0.10807890966534614,
|
|
"num_tokens": 2068267.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.747051239013672,
|
|
"epoch": 0.09329098848749504,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.4919,
|
|
"mean_token_accuracy": 0.1026596762239933,
|
|
"num_tokens": 2078271.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.686890411376953,
|
|
"epoch": 0.09378721714966257,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4104,
|
|
"mean_token_accuracy": 0.10279989019036292,
|
|
"num_tokens": 2088381.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.7383002758026125,
|
|
"epoch": 0.0942834458118301,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3729,
|
|
"mean_token_accuracy": 0.10531853288412094,
|
|
"num_tokens": 2098899.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.682248830795288,
|
|
"epoch": 0.09477967447399761,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5398,
|
|
"mean_token_accuracy": 0.10515285804867744,
|
|
"num_tokens": 2110057.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.719785499572754,
|
|
"epoch": 0.09527590313616514,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4403,
|
|
"mean_token_accuracy": 0.10757240056991577,
|
|
"num_tokens": 2121758.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.719501113891601,
|
|
"epoch": 0.09577213179833267,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4557,
|
|
"mean_token_accuracy": 0.10785956755280494,
|
|
"num_tokens": 2131840.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.616270351409912,
|
|
"epoch": 0.0962683604605002,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4294,
|
|
"mean_token_accuracy": 0.10218973308801652,
|
|
"num_tokens": 2142084.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.70799412727356,
|
|
"epoch": 0.09676458912266772,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4822,
|
|
"mean_token_accuracy": 0.10346104353666305,
|
|
"num_tokens": 2152785.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.7580810546875,
|
|
"epoch": 0.09726081778483525,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4715,
|
|
"mean_token_accuracy": 0.10466889962553978,
|
|
"num_tokens": 2163774.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.600415515899658,
|
|
"epoch": 0.09775704644700278,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3865,
|
|
"mean_token_accuracy": 0.10628859028220176,
|
|
"num_tokens": 2175378.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.872743368148804,
|
|
"epoch": 0.0982532751091703,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.6002,
|
|
"mean_token_accuracy": 0.10284787714481354,
|
|
"num_tokens": 2186510.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.675116109848022,
|
|
"epoch": 0.09874950377133783,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3875,
|
|
"mean_token_accuracy": 0.11210239455103874,
|
|
"num_tokens": 2197482.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.613697242736817,
|
|
"epoch": 0.09924573243350536,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3886,
|
|
"mean_token_accuracy": 0.1066878505051136,
|
|
"num_tokens": 2208329.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.565951251983643,
|
|
"epoch": 0.09974196109567289,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999999982145551,
|
|
"loss": 6.2974,
|
|
"mean_token_accuracy": 0.10837465524673462,
|
|
"num_tokens": 2218632.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.640917587280273,
|
|
"epoch": 0.1002381897578404,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004999999909611855,
|
|
"loss": 6.4173,
|
|
"mean_token_accuracy": 0.10387595295906067,
|
|
"num_tokens": 2231335.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.693672466278076,
|
|
"epoch": 0.10073441842000794,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999999781283008,
|
|
"loss": 6.3927,
|
|
"mean_token_accuracy": 0.10786194428801536,
|
|
"num_tokens": 2241704.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.652690649032593,
|
|
"epoch": 0.10123064708217547,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999999597159017,
|
|
"loss": 6.448,
|
|
"mean_token_accuracy": 0.1010373942553997,
|
|
"num_tokens": 2252310.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.650144433975219,
|
|
"epoch": 0.101726875744343,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999999357239882,
|
|
"loss": 6.3674,
|
|
"mean_token_accuracy": 0.10643667429685592,
|
|
"num_tokens": 2262270.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.674994850158692,
|
|
"epoch": 0.10222310440651051,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999999061525612,
|
|
"loss": 6.5075,
|
|
"mean_token_accuracy": 0.10479943379759789,
|
|
"num_tokens": 2273986.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.751373052597046,
|
|
"epoch": 0.10271933306867805,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999998710016214,
|
|
"loss": 6.4006,
|
|
"mean_token_accuracy": 0.10447262153029442,
|
|
"num_tokens": 2283430.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.647673654556274,
|
|
"epoch": 0.10321556173084558,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999998302711696,
|
|
"loss": 6.4082,
|
|
"mean_token_accuracy": 0.09957063123583794,
|
|
"num_tokens": 2294201.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.568925905227661,
|
|
"epoch": 0.1037117903930131,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999997839612069,
|
|
"loss": 6.309,
|
|
"mean_token_accuracy": 0.10835110917687416,
|
|
"num_tokens": 2303876.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.604097175598144,
|
|
"epoch": 0.10420801905518062,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999997320717342,
|
|
"loss": 6.3495,
|
|
"mean_token_accuracy": 0.1118161328136921,
|
|
"num_tokens": 2313713.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.559231519699097,
|
|
"epoch": 0.10470424771734815,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999996746027531,
|
|
"loss": 6.3595,
|
|
"mean_token_accuracy": 0.10454496443271637,
|
|
"num_tokens": 2324862.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.71983904838562,
|
|
"epoch": 0.10520047637951568,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499999611554265,
|
|
"loss": 6.431,
|
|
"mean_token_accuracy": 0.10956530421972274,
|
|
"num_tokens": 2334974.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.596372747421265,
|
|
"epoch": 0.10569670504168321,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999995429262712,
|
|
"loss": 6.4574,
|
|
"mean_token_accuracy": 0.1073372520506382,
|
|
"num_tokens": 2345945.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.653563451766968,
|
|
"epoch": 0.10619293370385073,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999994687187737,
|
|
"loss": 6.397,
|
|
"mean_token_accuracy": 0.10704459249973297,
|
|
"num_tokens": 2356749.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.710749959945678,
|
|
"epoch": 0.10668916236601826,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999993889317742,
|
|
"loss": 6.4721,
|
|
"mean_token_accuracy": 0.1034046970307827,
|
|
"num_tokens": 2368181.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.5815818309783936,
|
|
"epoch": 0.10718539102818579,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999993035652746,
|
|
"loss": 6.38,
|
|
"mean_token_accuracy": 0.10739323794841767,
|
|
"num_tokens": 2377737.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.599908828735352,
|
|
"epoch": 0.10768161969035331,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999992126192771,
|
|
"loss": 6.3006,
|
|
"mean_token_accuracy": 0.10967165976762772,
|
|
"num_tokens": 2388944.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.593209838867187,
|
|
"epoch": 0.10817784835252084,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499999116093784,
|
|
"loss": 6.3809,
|
|
"mean_token_accuracy": 0.10844719260931016,
|
|
"num_tokens": 2400758.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.568376064300537,
|
|
"epoch": 0.10867407701468837,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999990139887976,
|
|
"loss": 6.3747,
|
|
"mean_token_accuracy": 0.11013728603720666,
|
|
"num_tokens": 2411916.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.728429841995239,
|
|
"epoch": 0.1091703056768559,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999989063043206,
|
|
"loss": 6.4014,
|
|
"mean_token_accuracy": 0.10465751141309738,
|
|
"num_tokens": 2423310.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.530461025238037,
|
|
"epoch": 0.10966653433902342,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999987930403555,
|
|
"loss": 6.3331,
|
|
"mean_token_accuracy": 0.10839267894625663,
|
|
"num_tokens": 2432100.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.574756145477295,
|
|
"epoch": 0.11016276300119095,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999986741969051,
|
|
"loss": 6.4049,
|
|
"mean_token_accuracy": 0.1044836975634098,
|
|
"num_tokens": 2443563.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.649301338195801,
|
|
"epoch": 0.11065899166335848,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999985497739724,
|
|
"loss": 6.4178,
|
|
"mean_token_accuracy": 0.10569542646408081,
|
|
"num_tokens": 2455435.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.570544004440308,
|
|
"epoch": 0.11115522032552601,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999984197715606,
|
|
"loss": 6.3493,
|
|
"mean_token_accuracy": 0.10882077589631081,
|
|
"num_tokens": 2466173.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.516915893554687,
|
|
"epoch": 0.11165144898769352,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999982841896728,
|
|
"loss": 6.3036,
|
|
"mean_token_accuracy": 0.10695828124880791,
|
|
"num_tokens": 2476349.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.61347975730896,
|
|
"epoch": 0.11214767764986106,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999981430283123,
|
|
"loss": 6.4251,
|
|
"mean_token_accuracy": 0.11261754408478737,
|
|
"num_tokens": 2487864.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.566452312469482,
|
|
"epoch": 0.11264390631202859,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999979962874827,
|
|
"loss": 6.361,
|
|
"mean_token_accuracy": 0.10681881308555603,
|
|
"num_tokens": 2499188.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.645647430419922,
|
|
"epoch": 0.11314013497419612,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999978439671877,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.10881530791521073,
|
|
"num_tokens": 2509813.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.521003484725952,
|
|
"epoch": 0.11363636363636363,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999976860674309,
|
|
"loss": 6.3642,
|
|
"mean_token_accuracy": 0.10348583310842514,
|
|
"num_tokens": 2520813.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.54542760848999,
|
|
"epoch": 0.11413259229853116,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999975225882164,
|
|
"loss": 6.363,
|
|
"mean_token_accuracy": 0.10692540258169174,
|
|
"num_tokens": 2531802.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.528800964355469,
|
|
"epoch": 0.1146288209606987,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499997353529548,
|
|
"loss": 6.1676,
|
|
"mean_token_accuracy": 0.12081934958696365,
|
|
"num_tokens": 2541935.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.474112510681152,
|
|
"epoch": 0.11512504962286621,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999971788914303,
|
|
"loss": 6.2718,
|
|
"mean_token_accuracy": 0.1114352934062481,
|
|
"num_tokens": 2553224.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.483311462402344,
|
|
"epoch": 0.11562127828503374,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004999969986738672,
|
|
"loss": 6.3156,
|
|
"mean_token_accuracy": 0.10600073337554931,
|
|
"num_tokens": 2565355.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.61416974067688,
|
|
"epoch": 0.11611750694720127,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999968128768634,
|
|
"loss": 6.4,
|
|
"mean_token_accuracy": 0.10448111891746521,
|
|
"num_tokens": 2575962.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.583769941329956,
|
|
"epoch": 0.1166137356093688,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999966215004236,
|
|
"loss": 6.4055,
|
|
"mean_token_accuracy": 0.11200469061732292,
|
|
"num_tokens": 2588136.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.557573032379151,
|
|
"epoch": 0.11710996427153632,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999964245445522,
|
|
"loss": 6.2965,
|
|
"mean_token_accuracy": 0.11258405596017837,
|
|
"num_tokens": 2597785.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.516626834869385,
|
|
"epoch": 0.11760619293370385,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999962220092545,
|
|
"loss": 6.337,
|
|
"mean_token_accuracy": 0.11605330407619477,
|
|
"num_tokens": 2609738.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.4262982368469235,
|
|
"epoch": 0.11810242159587138,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004999960138945351,
|
|
"loss": 6.2232,
|
|
"mean_token_accuracy": 0.11857966929674149,
|
|
"num_tokens": 2620179.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.496797704696656,
|
|
"epoch": 0.11859865025803891,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999958002003996,
|
|
"loss": 6.2665,
|
|
"mean_token_accuracy": 0.11352286636829376,
|
|
"num_tokens": 2631550.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.50876030921936,
|
|
"epoch": 0.11909487892020643,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004999955809268529,
|
|
"loss": 6.3027,
|
|
"mean_token_accuracy": 0.10872863009572029,
|
|
"num_tokens": 2642467.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.478094530105591,
|
|
"epoch": 0.11959110758237396,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999953560739008,
|
|
"loss": 6.2868,
|
|
"mean_token_accuracy": 0.11035665571689605,
|
|
"num_tokens": 2652132.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.516570568084717,
|
|
"epoch": 0.12008733624454149,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999951256415485,
|
|
"loss": 6.3359,
|
|
"mean_token_accuracy": 0.1081702619791031,
|
|
"num_tokens": 2663336.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.608760070800781,
|
|
"epoch": 0.12058356490670902,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499994889629802,
|
|
"loss": 6.4131,
|
|
"mean_token_accuracy": 0.10076656267046928,
|
|
"num_tokens": 2674201.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.543395376205444,
|
|
"epoch": 0.12107979356887653,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000499994648038667,
|
|
"loss": 6.3385,
|
|
"mean_token_accuracy": 0.11510411649942398,
|
|
"num_tokens": 2685073.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.572003412246704,
|
|
"epoch": 0.12157602223104406,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999944008681496,
|
|
"loss": 6.2812,
|
|
"mean_token_accuracy": 0.11140692159533501,
|
|
"num_tokens": 2695024.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.5019573211669925,
|
|
"epoch": 0.1220722508932116,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999941481182559,
|
|
"loss": 6.3588,
|
|
"mean_token_accuracy": 0.10714156180620193,
|
|
"num_tokens": 2706232.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.5704296112060545,
|
|
"epoch": 0.12256847955537911,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999938897889921,
|
|
"loss": 6.3373,
|
|
"mean_token_accuracy": 0.11231422051787376,
|
|
"num_tokens": 2715874.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.454069900512695,
|
|
"epoch": 0.12306470821754664,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999936258803647,
|
|
"loss": 6.2863,
|
|
"mean_token_accuracy": 0.11209510788321495,
|
|
"num_tokens": 2727409.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.470958852767945,
|
|
"epoch": 0.12356093687971417,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999933563923802,
|
|
"loss": 6.1948,
|
|
"mean_token_accuracy": 0.11663963794708251,
|
|
"num_tokens": 2737259.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.504468393325806,
|
|
"epoch": 0.1240571655418817,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004999930813250451,
|
|
"loss": 6.3227,
|
|
"mean_token_accuracy": 0.11047348752617836,
|
|
"num_tokens": 2748416.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.485745906829834,
|
|
"epoch": 0.12455339420404922,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999928006783667,
|
|
"loss": 6.2973,
|
|
"mean_token_accuracy": 0.11183110401034355,
|
|
"num_tokens": 2760470.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.535139369964599,
|
|
"epoch": 0.12504962286621676,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999925144523515,
|
|
"loss": 6.3279,
|
|
"mean_token_accuracy": 0.10748472735285759,
|
|
"num_tokens": 2770222.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.476829242706299,
|
|
"epoch": 0.12554585152838427,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999922226470066,
|
|
"loss": 6.2505,
|
|
"mean_token_accuracy": 0.11203066259622574,
|
|
"num_tokens": 2780212.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.390298366546631,
|
|
"epoch": 0.1260420801905518,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999919252623397,
|
|
"loss": 6.2144,
|
|
"mean_token_accuracy": 0.11430977880954743,
|
|
"num_tokens": 2790239.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.52067379951477,
|
|
"epoch": 0.12653830885271933,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999916222983577,
|
|
"loss": 6.254,
|
|
"mean_token_accuracy": 0.11566178128123283,
|
|
"num_tokens": 2801902.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.4108202934265135,
|
|
"epoch": 0.12703453751488686,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999913137550682,
|
|
"loss": 6.2473,
|
|
"mean_token_accuracy": 0.11208073645830155,
|
|
"num_tokens": 2813268.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.478806877136231,
|
|
"epoch": 0.1275307661770544,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999909996324791,
|
|
"loss": 6.32,
|
|
"mean_token_accuracy": 0.10953322052955627,
|
|
"num_tokens": 2824506.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.505747509002686,
|
|
"epoch": 0.12802699483922192,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000499990679930598,
|
|
"loss": 6.2224,
|
|
"mean_token_accuracy": 0.11515569612383843,
|
|
"num_tokens": 2835144.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.419563961029053,
|
|
"epoch": 0.12852322350138945,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999903546494327,
|
|
"loss": 6.2472,
|
|
"mean_token_accuracy": 0.11156807765364647,
|
|
"num_tokens": 2845382.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.44638466835022,
|
|
"epoch": 0.12901945216355698,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999900237889916,
|
|
"loss": 6.2666,
|
|
"mean_token_accuracy": 0.11455247327685356,
|
|
"num_tokens": 2854697.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.497555685043335,
|
|
"epoch": 0.12951568082572448,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999896873492827,
|
|
"loss": 6.2643,
|
|
"mean_token_accuracy": 0.10822557806968688,
|
|
"num_tokens": 2866216.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.422132730484009,
|
|
"epoch": 0.130011909487892,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999893453303143,
|
|
"loss": 6.2308,
|
|
"mean_token_accuracy": 0.11659785062074661,
|
|
"num_tokens": 2877734.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.505683135986328,
|
|
"epoch": 0.13050813815005954,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999889977320949,
|
|
"loss": 6.289,
|
|
"mean_token_accuracy": 0.1148169457912445,
|
|
"num_tokens": 2889445.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.354604768753052,
|
|
"epoch": 0.13100436681222707,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999886445546332,
|
|
"loss": 6.2237,
|
|
"mean_token_accuracy": 0.1161403276026249,
|
|
"num_tokens": 2900485.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.544115304946899,
|
|
"epoch": 0.1315005954743946,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499988285797938,
|
|
"loss": 6.2087,
|
|
"mean_token_accuracy": 0.11861399188637733,
|
|
"num_tokens": 2909795.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.294916772842408,
|
|
"epoch": 0.13199682413656214,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000499987921462018,
|
|
"loss": 6.2314,
|
|
"mean_token_accuracy": 0.12291175946593284,
|
|
"num_tokens": 2920153.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.524097633361817,
|
|
"epoch": 0.13249305279872967,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999875515468825,
|
|
"loss": 6.2228,
|
|
"mean_token_accuracy": 0.1149907335639,
|
|
"num_tokens": 2930457.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.369509267807007,
|
|
"epoch": 0.13298928146089717,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999871760525403,
|
|
"loss": 6.2798,
|
|
"mean_token_accuracy": 0.11520167663693429,
|
|
"num_tokens": 2941305.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.499384498596191,
|
|
"epoch": 0.1334855101230647,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999867949790011,
|
|
"loss": 6.2212,
|
|
"mean_token_accuracy": 0.11666619777679443,
|
|
"num_tokens": 2951798.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.377152729034424,
|
|
"epoch": 0.13398173878523223,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999864083262743,
|
|
"loss": 6.1253,
|
|
"mean_token_accuracy": 0.11854363307356834,
|
|
"num_tokens": 2962486.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.414649248123169,
|
|
"epoch": 0.13447796744739976,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999860160943691,
|
|
"loss": 6.2323,
|
|
"mean_token_accuracy": 0.1163373351097107,
|
|
"num_tokens": 2974317.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.405454921722412,
|
|
"epoch": 0.1349741961095673,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999856182832958,
|
|
"loss": 6.1428,
|
|
"mean_token_accuracy": 0.11802050322294236,
|
|
"num_tokens": 2985440.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.40319504737854,
|
|
"epoch": 0.13547042477173482,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999852148930637,
|
|
"loss": 6.222,
|
|
"mean_token_accuracy": 0.11579161584377289,
|
|
"num_tokens": 2997044.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.445984601974487,
|
|
"epoch": 0.13596665343390235,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999848059236832,
|
|
"loss": 6.2333,
|
|
"mean_token_accuracy": 0.11098912060260772,
|
|
"num_tokens": 3005965.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.295212173461914,
|
|
"epoch": 0.13646288209606988,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999843913751643,
|
|
"loss": 6.1123,
|
|
"mean_token_accuracy": 0.11798060461878776,
|
|
"num_tokens": 3015031.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.371025562286377,
|
|
"epoch": 0.13695911075823738,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999839712475174,
|
|
"loss": 6.1708,
|
|
"mean_token_accuracy": 0.11246290653944016,
|
|
"num_tokens": 3026275.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.35689868927002,
|
|
"epoch": 0.13745533942040492,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999835455407526,
|
|
"loss": 6.2198,
|
|
"mean_token_accuracy": 0.11414360404014587,
|
|
"num_tokens": 3036778.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.4954687595367435,
|
|
"epoch": 0.13795156808257245,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999831142548808,
|
|
"loss": 6.3258,
|
|
"mean_token_accuracy": 0.11213771849870682,
|
|
"num_tokens": 3048149.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.3874505996704105,
|
|
"epoch": 0.13844779674473998,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999826773899124,
|
|
"loss": 6.1823,
|
|
"mean_token_accuracy": 0.11371054351329804,
|
|
"num_tokens": 3059149.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.409183120727539,
|
|
"epoch": 0.1389440254069075,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004999822349458586,
|
|
"loss": 6.2059,
|
|
"mean_token_accuracy": 0.1147544652223587,
|
|
"num_tokens": 3071613.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.439931964874267,
|
|
"epoch": 0.13944025406907504,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999817869227301,
|
|
"loss": 6.3355,
|
|
"mean_token_accuracy": 0.11122817322611808,
|
|
"num_tokens": 3084742.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.420901536941528,
|
|
"epoch": 0.13993648273124257,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999813333205381,
|
|
"loss": 6.2625,
|
|
"mean_token_accuracy": 0.11637563481926919,
|
|
"num_tokens": 3095356.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.433026123046875,
|
|
"epoch": 0.14043271139341007,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999808741392937,
|
|
"loss": 6.2222,
|
|
"mean_token_accuracy": 0.10983924940228462,
|
|
"num_tokens": 3107717.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.381127834320068,
|
|
"epoch": 0.1409289400555776,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999804093790086,
|
|
"loss": 6.2182,
|
|
"mean_token_accuracy": 0.11328766494989395,
|
|
"num_tokens": 3119205.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.441397428512573,
|
|
"epoch": 0.14142516871774513,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000499979939039694,
|
|
"loss": 6.2685,
|
|
"mean_token_accuracy": 0.11661531254649163,
|
|
"num_tokens": 3130564.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.354721212387085,
|
|
"epoch": 0.14192139737991266,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999794631213618,
|
|
"loss": 6.2024,
|
|
"mean_token_accuracy": 0.11099046766757965,
|
|
"num_tokens": 3141883.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.4592526912689205,
|
|
"epoch": 0.1424176260420802,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999789816240236,
|
|
"loss": 6.208,
|
|
"mean_token_accuracy": 0.11153334975242615,
|
|
"num_tokens": 3152817.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.309853506088257,
|
|
"epoch": 0.14291385470424772,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999784945476915,
|
|
"loss": 6.1224,
|
|
"mean_token_accuracy": 0.12020867317914963,
|
|
"num_tokens": 3163546.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.413205289840699,
|
|
"epoch": 0.14341008336641525,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999780018923775,
|
|
"loss": 6.1713,
|
|
"mean_token_accuracy": 0.1094489723443985,
|
|
"num_tokens": 3173899.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.2561510562896725,
|
|
"epoch": 0.14390631202858278,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999775036580939,
|
|
"loss": 6.0903,
|
|
"mean_token_accuracy": 0.11627468764781952,
|
|
"num_tokens": 3184433.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.487192440032959,
|
|
"epoch": 0.1444025406907503,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999769998448529,
|
|
"loss": 6.1973,
|
|
"mean_token_accuracy": 0.1131520263850689,
|
|
"num_tokens": 3196077.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.187653255462647,
|
|
"epoch": 0.14489876935291782,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999764904526672,
|
|
"loss": 6.1208,
|
|
"mean_token_accuracy": 0.11931058466434478,
|
|
"num_tokens": 3206645.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.462401914596557,
|
|
"epoch": 0.14539499801508535,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999759754815492,
|
|
"loss": 6.2386,
|
|
"mean_token_accuracy": 0.11986212581396102,
|
|
"num_tokens": 3216786.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.32322940826416,
|
|
"epoch": 0.14589122667725288,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999754549315118,
|
|
"loss": 6.1656,
|
|
"mean_token_accuracy": 0.11642653867602348,
|
|
"num_tokens": 3227384.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.288940668106079,
|
|
"epoch": 0.1463874553394204,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000499974928802568,
|
|
"loss": 6.0217,
|
|
"mean_token_accuracy": 0.1218691572546959,
|
|
"num_tokens": 3238800.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.292658805847168,
|
|
"epoch": 0.14688368400158794,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999743970947306,
|
|
"loss": 6.1753,
|
|
"mean_token_accuracy": 0.11594840288162231,
|
|
"num_tokens": 3248627.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.4044637203216555,
|
|
"epoch": 0.14737991266375547,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499973859808013,
|
|
"loss": 6.1955,
|
|
"mean_token_accuracy": 0.11360678747296334,
|
|
"num_tokens": 3258989.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.28885703086853,
|
|
"epoch": 0.14787614132592297,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999733169424284,
|
|
"loss": 6.175,
|
|
"mean_token_accuracy": 0.11919418796896934,
|
|
"num_tokens": 3269474.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.379557180404663,
|
|
"epoch": 0.1483723699880905,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999727684979904,
|
|
"loss": 6.1567,
|
|
"mean_token_accuracy": 0.11620205119252205,
|
|
"num_tokens": 3280453.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.28340425491333,
|
|
"epoch": 0.14886859865025803,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999722144747125,
|
|
"loss": 6.0847,
|
|
"mean_token_accuracy": 0.11536976397037506,
|
|
"num_tokens": 3290234.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.280434465408325,
|
|
"epoch": 0.14936482731242556,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999716548726085,
|
|
"loss": 6.0635,
|
|
"mean_token_accuracy": 0.12484975904226303,
|
|
"num_tokens": 3300943.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.274162673950196,
|
|
"epoch": 0.1498610559745931,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999710896916921,
|
|
"loss": 6.0263,
|
|
"mean_token_accuracy": 0.11883052140474319,
|
|
"num_tokens": 3311296.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.32963080406189,
|
|
"epoch": 0.15035728463676062,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999705189319774,
|
|
"loss": 6.1816,
|
|
"mean_token_accuracy": 0.11108686327934265,
|
|
"num_tokens": 3322310.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.295844268798828,
|
|
"epoch": 0.15085351329892815,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999699425934788,
|
|
"loss": 6.1842,
|
|
"mean_token_accuracy": 0.11696541234850884,
|
|
"num_tokens": 3334176.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.382207250595092,
|
|
"epoch": 0.15134974196109569,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999693606762103,
|
|
"loss": 6.1791,
|
|
"mean_token_accuracy": 0.11649519056081772,
|
|
"num_tokens": 3344839.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.323039865493774,
|
|
"epoch": 0.1518459706232632,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999687731801865,
|
|
"loss": 6.1538,
|
|
"mean_token_accuracy": 0.11218968853354454,
|
|
"num_tokens": 3357046.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.362309741973877,
|
|
"epoch": 0.15234219928543072,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999681801054217,
|
|
"loss": 6.2222,
|
|
"mean_token_accuracy": 0.11561866924166679,
|
|
"num_tokens": 3367783.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.31308650970459,
|
|
"epoch": 0.15283842794759825,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499967581451931,
|
|
"loss": 6.0902,
|
|
"mean_token_accuracy": 0.11377239897847176,
|
|
"num_tokens": 3378529.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.354143381118774,
|
|
"epoch": 0.15333465660976578,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499966977219729,
|
|
"loss": 6.2817,
|
|
"mean_token_accuracy": 0.11187926903367043,
|
|
"num_tokens": 3389318.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.434057950973511,
|
|
"epoch": 0.1538308852719333,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999663674088306,
|
|
"loss": 6.2759,
|
|
"mean_token_accuracy": 0.11047122180461884,
|
|
"num_tokens": 3400603.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.393321371078491,
|
|
"epoch": 0.15432711393410084,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999657520192511,
|
|
"loss": 6.1748,
|
|
"mean_token_accuracy": 0.1146856315433979,
|
|
"num_tokens": 3410316.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.320789861679077,
|
|
"epoch": 0.15482334259626837,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999651310510058,
|
|
"loss": 6.1198,
|
|
"mean_token_accuracy": 0.12020543217658997,
|
|
"num_tokens": 3422155.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.328480195999146,
|
|
"epoch": 0.15531957125843587,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999645045041099,
|
|
"loss": 6.0594,
|
|
"mean_token_accuracy": 0.11951424777507783,
|
|
"num_tokens": 3432892.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.272742033004761,
|
|
"epoch": 0.1558157999206034,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999638723785791,
|
|
"loss": 6.1812,
|
|
"mean_token_accuracy": 0.11743228211998939,
|
|
"num_tokens": 3443967.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.284852838516235,
|
|
"epoch": 0.15631202858277093,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000499963234674429,
|
|
"loss": 6.069,
|
|
"mean_token_accuracy": 0.11911690831184388,
|
|
"num_tokens": 3454354.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.222921228408813,
|
|
"epoch": 0.15680825724493846,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999625913916754,
|
|
"loss": 6.0872,
|
|
"mean_token_accuracy": 0.11979979053139686,
|
|
"num_tokens": 3465232.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.206456661224365,
|
|
"epoch": 0.157304485907106,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999619425303343,
|
|
"loss": 6.0632,
|
|
"mean_token_accuracy": 0.12077862471342087,
|
|
"num_tokens": 3475443.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.376963663101196,
|
|
"epoch": 0.15780071456927353,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999612880904219,
|
|
"loss": 6.1005,
|
|
"mean_token_accuracy": 0.11797050833702087,
|
|
"num_tokens": 3486643.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.291903114318847,
|
|
"epoch": 0.15829694323144106,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999606280719542,
|
|
"loss": 6.1495,
|
|
"mean_token_accuracy": 0.11660699099302292,
|
|
"num_tokens": 3497557.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.405364274978638,
|
|
"epoch": 0.1587931718936086,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999599624749475,
|
|
"loss": 6.1583,
|
|
"mean_token_accuracy": 0.1143525168299675,
|
|
"num_tokens": 3509790.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.194210481643677,
|
|
"epoch": 0.1592894005557761,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999592912994187,
|
|
"loss": 6.077,
|
|
"mean_token_accuracy": 0.11786877140402793,
|
|
"num_tokens": 3521577.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.348882055282592,
|
|
"epoch": 0.15978562921794362,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999586145453842,
|
|
"loss": 6.1151,
|
|
"mean_token_accuracy": 0.11298025399446487,
|
|
"num_tokens": 3532126.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.160671663284302,
|
|
"epoch": 0.16028185788011115,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999579322128608,
|
|
"loss": 5.9537,
|
|
"mean_token_accuracy": 0.11721618697047234,
|
|
"num_tokens": 3542560.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.2272850513458256,
|
|
"epoch": 0.16077808654227868,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999572443018654,
|
|
"loss": 6.1069,
|
|
"mean_token_accuracy": 0.1176838219165802,
|
|
"num_tokens": 3553863.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.241751861572266,
|
|
"epoch": 0.1612743152044462,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999565508124151,
|
|
"loss": 6.1353,
|
|
"mean_token_accuracy": 0.11932577788829804,
|
|
"num_tokens": 3564953.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.313349819183349,
|
|
"epoch": 0.16177054386661374,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000499955851744527,
|
|
"loss": 6.1226,
|
|
"mean_token_accuracy": 0.12102637365460396,
|
|
"num_tokens": 3576247.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.362412357330323,
|
|
"epoch": 0.16226677252878127,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999551470982185,
|
|
"loss": 6.1641,
|
|
"mean_token_accuracy": 0.11501411274075508,
|
|
"num_tokens": 3587600.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.3714934349060055,
|
|
"epoch": 0.1627630011909488,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999544368735073,
|
|
"loss": 6.1295,
|
|
"mean_token_accuracy": 0.12266664281487465,
|
|
"num_tokens": 3597727.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.2838012218475345,
|
|
"epoch": 0.1632592298531163,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999537210704106,
|
|
"loss": 6.1149,
|
|
"mean_token_accuracy": 0.1231493391096592,
|
|
"num_tokens": 3608066.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.3129997730255125,
|
|
"epoch": 0.16375545851528384,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999529996889463,
|
|
"loss": 6.1644,
|
|
"mean_token_accuracy": 0.11525594145059585,
|
|
"num_tokens": 3618263.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.383852910995484,
|
|
"epoch": 0.16425168717745137,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999522727291325,
|
|
"loss": 6.1189,
|
|
"mean_token_accuracy": 0.12352551445364952,
|
|
"num_tokens": 3629465.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.246583700180054,
|
|
"epoch": 0.1647479158396189,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499951540190987,
|
|
"loss": 6.1389,
|
|
"mean_token_accuracy": 0.11551668420433998,
|
|
"num_tokens": 3639657.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.313644647598267,
|
|
"epoch": 0.16524414450178643,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999508020745281,
|
|
"loss": 6.1039,
|
|
"mean_token_accuracy": 0.11597631126642227,
|
|
"num_tokens": 3652374.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.2801806926727295,
|
|
"epoch": 0.16574037316395396,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999500583797739,
|
|
"loss": 6.0802,
|
|
"mean_token_accuracy": 0.11924066990613938,
|
|
"num_tokens": 3664083.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.2540826320648195,
|
|
"epoch": 0.1662366018261215,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999493091067431,
|
|
"loss": 6.1179,
|
|
"mean_token_accuracy": 0.12284742221236229,
|
|
"num_tokens": 3675314.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.218377304077149,
|
|
"epoch": 0.166732830488289,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999485542554541,
|
|
"loss": 6.081,
|
|
"mean_token_accuracy": 0.11232990324497223,
|
|
"num_tokens": 3686019.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.390085935592651,
|
|
"epoch": 0.16722905915045652,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999477938259256,
|
|
"loss": 6.1781,
|
|
"mean_token_accuracy": 0.11307597011327744,
|
|
"num_tokens": 3698170.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.218471336364746,
|
|
"epoch": 0.16772528781262405,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999470278181766,
|
|
"loss": 6.1603,
|
|
"mean_token_accuracy": 0.11149280816316605,
|
|
"num_tokens": 3709243.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.361375761032105,
|
|
"epoch": 0.16822151647479158,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999462562322261,
|
|
"loss": 6.1193,
|
|
"mean_token_accuracy": 0.11822729557752609,
|
|
"num_tokens": 3720196.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.330177640914917,
|
|
"epoch": 0.1687177451369591,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000499945479068093,
|
|
"loss": 6.1702,
|
|
"mean_token_accuracy": 0.1214469164609909,
|
|
"num_tokens": 3732714.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.278452777862549,
|
|
"epoch": 0.16921397379912664,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004999446963257969,
|
|
"loss": 6.14,
|
|
"mean_token_accuracy": 0.11547001376748085,
|
|
"num_tokens": 3746474.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.218746757507324,
|
|
"epoch": 0.16971020246129417,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999439080053571,
|
|
"loss": 6.0096,
|
|
"mean_token_accuracy": 0.12439236715435982,
|
|
"num_tokens": 3757528.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.3749387741088865,
|
|
"epoch": 0.1702064311234617,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999431141067928,
|
|
"loss": 6.1542,
|
|
"mean_token_accuracy": 0.11521807983517647,
|
|
"num_tokens": 3768360.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.213152885437012,
|
|
"epoch": 0.1707026597856292,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999423146301242,
|
|
"loss": 6.016,
|
|
"mean_token_accuracy": 0.12060007154941559,
|
|
"num_tokens": 3779809.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.188921022415161,
|
|
"epoch": 0.17119888844779674,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999415095753709,
|
|
"loss": 6.0182,
|
|
"mean_token_accuracy": 0.12153950557112694,
|
|
"num_tokens": 3791455.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.190573024749756,
|
|
"epoch": 0.17169511710996427,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999406989425527,
|
|
"loss": 6.051,
|
|
"mean_token_accuracy": 0.11963336691260337,
|
|
"num_tokens": 3803209.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.1931469440460205,
|
|
"epoch": 0.1721913457721318,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999398827316901,
|
|
"loss": 6.0878,
|
|
"mean_token_accuracy": 0.11741367131471633,
|
|
"num_tokens": 3813553.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.292501640319824,
|
|
"epoch": 0.17268757443429933,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000499939060942803,
|
|
"loss": 6.1305,
|
|
"mean_token_accuracy": 0.1166291169822216,
|
|
"num_tokens": 3824856.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.269052934646607,
|
|
"epoch": 0.17318380309646686,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999382335759119,
|
|
"loss": 6.2027,
|
|
"mean_token_accuracy": 0.11211237907409669,
|
|
"num_tokens": 3835489.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.338535118103027,
|
|
"epoch": 0.1736800317586344,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999374006310372,
|
|
"loss": 6.0367,
|
|
"mean_token_accuracy": 0.11971709057688713,
|
|
"num_tokens": 3844838.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.158118391036988,
|
|
"epoch": 0.1741762604208019,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999365621081999,
|
|
"loss": 6.0031,
|
|
"mean_token_accuracy": 0.11688855215907097,
|
|
"num_tokens": 3855288.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.210778141021729,
|
|
"epoch": 0.17467248908296942,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999357180074204,
|
|
"loss": 6.0566,
|
|
"mean_token_accuracy": 0.12111689299345016,
|
|
"num_tokens": 3865585.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.272298288345337,
|
|
"epoch": 0.17516871774513695,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999348683287199,
|
|
"loss": 6.0413,
|
|
"mean_token_accuracy": 0.11825228407979012,
|
|
"num_tokens": 3877694.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.2135413646697994,
|
|
"epoch": 0.17566494640730448,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999340130721192,
|
|
"loss": 6.0028,
|
|
"mean_token_accuracy": 0.13018031790852547,
|
|
"num_tokens": 3888802.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.1698943138122555,
|
|
"epoch": 0.17616117506947201,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999331522376396,
|
|
"loss": 5.9874,
|
|
"mean_token_accuracy": 0.12037508189678192,
|
|
"num_tokens": 3900294.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.191412830352784,
|
|
"epoch": 0.17665740373163955,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999322858253027,
|
|
"loss": 6.0106,
|
|
"mean_token_accuracy": 0.12115705609321595,
|
|
"num_tokens": 3910865.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.2171179294586185,
|
|
"epoch": 0.17715363239380708,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999314138351297,
|
|
"loss": 6.0731,
|
|
"mean_token_accuracy": 0.11963929235935211,
|
|
"num_tokens": 3921960.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.269604063034057,
|
|
"epoch": 0.1776498610559746,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999305362671424,
|
|
"loss": 6.0819,
|
|
"mean_token_accuracy": 0.11684257239103317,
|
|
"num_tokens": 3934085.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.16135573387146,
|
|
"epoch": 0.1781460897181421,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999296531213624,
|
|
"loss": 5.9371,
|
|
"mean_token_accuracy": 0.1304635725915432,
|
|
"num_tokens": 3944532.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.181241798400879,
|
|
"epoch": 0.17864231838030964,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999287643978117,
|
|
"loss": 5.9725,
|
|
"mean_token_accuracy": 0.12921013534069062,
|
|
"num_tokens": 3954995.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.115298318862915,
|
|
"epoch": 0.17913854704247717,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999278700965124,
|
|
"loss": 6.0287,
|
|
"mean_token_accuracy": 0.12031830176711082,
|
|
"num_tokens": 3966621.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.2137250900268555,
|
|
"epoch": 0.1796347757046447,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999269702174866,
|
|
"loss": 6.0895,
|
|
"mean_token_accuracy": 0.12122552394866944,
|
|
"num_tokens": 3977568.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.222836112976074,
|
|
"epoch": 0.18013100436681223,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999260647607565,
|
|
"loss": 6.1064,
|
|
"mean_token_accuracy": 0.11863134428858757,
|
|
"num_tokens": 3988571.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.24027271270752,
|
|
"epoch": 0.18062723302897976,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999251537263448,
|
|
"loss": 5.9314,
|
|
"mean_token_accuracy": 0.12621500641107558,
|
|
"num_tokens": 3998380.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.1338170051574705,
|
|
"epoch": 0.1811234616911473,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999242371142739,
|
|
"loss": 6.1195,
|
|
"mean_token_accuracy": 0.12032051682472229,
|
|
"num_tokens": 4010017.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.292723035812378,
|
|
"epoch": 0.1816196903533148,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999233149245665,
|
|
"loss": 6.0814,
|
|
"mean_token_accuracy": 0.11958788856863975,
|
|
"num_tokens": 4020922.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.293724489212036,
|
|
"epoch": 0.18211591901548232,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999223871572456,
|
|
"loss": 6.0459,
|
|
"mean_token_accuracy": 0.11999811977148056,
|
|
"num_tokens": 4030927.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.2004194259643555,
|
|
"epoch": 0.18261214767764986,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999214538123342,
|
|
"loss": 6.122,
|
|
"mean_token_accuracy": 0.11295817345380783,
|
|
"num_tokens": 4041375.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.214535427093506,
|
|
"epoch": 0.18310837633981739,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999205148898554,
|
|
"loss": 6.1229,
|
|
"mean_token_accuracy": 0.11756806150078773,
|
|
"num_tokens": 4053213.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.2128990650177,
|
|
"epoch": 0.18360460500198492,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999195703898325,
|
|
"loss": 5.9532,
|
|
"mean_token_accuracy": 0.12472454011440277,
|
|
"num_tokens": 4064536.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.201371765136718,
|
|
"epoch": 0.18410083366415245,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999186203122889,
|
|
"loss": 6.0295,
|
|
"mean_token_accuracy": 0.12741652950644494,
|
|
"num_tokens": 4075282.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.092880630493164,
|
|
"epoch": 0.18459706232631998,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999176646572482,
|
|
"loss": 6.0337,
|
|
"mean_token_accuracy": 0.12471091002225876,
|
|
"num_tokens": 4085501.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.233616590499878,
|
|
"epoch": 0.1850932909884875,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499916703424734,
|
|
"loss": 5.9813,
|
|
"mean_token_accuracy": 0.11598324999213219,
|
|
"num_tokens": 4095697.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.167621278762818,
|
|
"epoch": 0.185589519650655,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999157366147702,
|
|
"loss": 6.0469,
|
|
"mean_token_accuracy": 0.12036259993910789,
|
|
"num_tokens": 4106459.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.21561450958252,
|
|
"epoch": 0.18608574831282254,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999147642273809,
|
|
"loss": 5.9731,
|
|
"mean_token_accuracy": 0.12678501680493354,
|
|
"num_tokens": 4118320.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.1109779357910154,
|
|
"epoch": 0.18658197697499007,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00049991378626259,
|
|
"loss": 5.995,
|
|
"mean_token_accuracy": 0.13212494775652886,
|
|
"num_tokens": 4131249.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.219920682907104,
|
|
"epoch": 0.1870782056371576,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999128027204219,
|
|
"loss": 6.0718,
|
|
"mean_token_accuracy": 0.12198005616664886,
|
|
"num_tokens": 4141480.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.236648035049439,
|
|
"epoch": 0.18757443429932513,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499911813600901,
|
|
"loss": 6.1177,
|
|
"mean_token_accuracy": 0.11755191907286644,
|
|
"num_tokens": 4152344.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.261018133163452,
|
|
"epoch": 0.18807066296149266,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999108189040516,
|
|
"loss": 6.0806,
|
|
"mean_token_accuracy": 0.11942842453718186,
|
|
"num_tokens": 4164772.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.152463722229004,
|
|
"epoch": 0.1885668916236602,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999098186298986,
|
|
"loss": 6.0359,
|
|
"mean_token_accuracy": 0.11918087005615234,
|
|
"num_tokens": 4175511.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.117107248306274,
|
|
"epoch": 0.1890631202858277,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999088127784667,
|
|
"loss": 6.0012,
|
|
"mean_token_accuracy": 0.12256012931466102,
|
|
"num_tokens": 4187445.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.237027502059936,
|
|
"epoch": 0.18955934894799523,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000499907801349781,
|
|
"loss": 5.9748,
|
|
"mean_token_accuracy": 0.1233195923268795,
|
|
"num_tokens": 4198124.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.142972564697265,
|
|
"epoch": 0.19005557761016276,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999067843438664,
|
|
"loss": 6.0176,
|
|
"mean_token_accuracy": 0.12327835708856583,
|
|
"num_tokens": 4208954.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.2328629970550535,
|
|
"epoch": 0.1905518062723303,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999057617607482,
|
|
"loss": 6.0782,
|
|
"mean_token_accuracy": 0.12020684257149697,
|
|
"num_tokens": 4218896.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.11851167678833,
|
|
"epoch": 0.19104803493449782,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999047336004517,
|
|
"loss": 5.9858,
|
|
"mean_token_accuracy": 0.11666620373725892,
|
|
"num_tokens": 4228851.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.121028709411621,
|
|
"epoch": 0.19154426359666535,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999036998630024,
|
|
"loss": 5.9408,
|
|
"mean_token_accuracy": 0.12234764248132705,
|
|
"num_tokens": 4240377.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.136601066589355,
|
|
"epoch": 0.19204049225883288,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499902660548426,
|
|
"loss": 5.9324,
|
|
"mean_token_accuracy": 0.12323587760329247,
|
|
"num_tokens": 4251341.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.131585884094238,
|
|
"epoch": 0.1925367209210004,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999016156567482,
|
|
"loss": 6.039,
|
|
"mean_token_accuracy": 0.12415843605995178,
|
|
"num_tokens": 4261206.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.232227182388305,
|
|
"epoch": 0.1930329495831679,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000499900565187995,
|
|
"loss": 6.0428,
|
|
"mean_token_accuracy": 0.11449379473924637,
|
|
"num_tokens": 4272662.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.217731094360351,
|
|
"epoch": 0.19352917824533544,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998995091421924,
|
|
"loss": 6.0248,
|
|
"mean_token_accuracy": 0.12198412641882897,
|
|
"num_tokens": 4283302.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.128766965866089,
|
|
"epoch": 0.19402540690750297,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998984475193665,
|
|
"loss": 5.9794,
|
|
"mean_token_accuracy": 0.12389139384031296,
|
|
"num_tokens": 4296087.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.144579887390137,
|
|
"epoch": 0.1945216355696705,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998973803195439,
|
|
"loss": 5.9912,
|
|
"mean_token_accuracy": 0.12408131062984466,
|
|
"num_tokens": 4305918.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.167060136795044,
|
|
"epoch": 0.19501786423183803,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998963075427508,
|
|
"loss": 6.0475,
|
|
"mean_token_accuracy": 0.12526998519897461,
|
|
"num_tokens": 4316507.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.1785789966583256,
|
|
"epoch": 0.19551409289400556,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004998952291890138,
|
|
"loss": 5.9762,
|
|
"mean_token_accuracy": 0.12871691659092904,
|
|
"num_tokens": 4329091.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.164827680587768,
|
|
"epoch": 0.1960103215561731,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998941452583599,
|
|
"loss": 6.0421,
|
|
"mean_token_accuracy": 0.12194729819893838,
|
|
"num_tokens": 4339266.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.106686210632324,
|
|
"epoch": 0.1965065502183406,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998930557508156,
|
|
"loss": 5.9713,
|
|
"mean_token_accuracy": 0.12659700810909272,
|
|
"num_tokens": 4349300.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.051992893218994,
|
|
"epoch": 0.19700277888050813,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004998919606664082,
|
|
"loss": 5.8926,
|
|
"mean_token_accuracy": 0.13368902131915092,
|
|
"num_tokens": 4360314.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.151434469223022,
|
|
"epoch": 0.19749900754267566,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998908600051648,
|
|
"loss": 5.9413,
|
|
"mean_token_accuracy": 0.12665844783186914,
|
|
"num_tokens": 4370077.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.119269704818725,
|
|
"epoch": 0.1979952362048432,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998897537671126,
|
|
"loss": 5.951,
|
|
"mean_token_accuracy": 0.12383900433778763,
|
|
"num_tokens": 4381000.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.143065547943115,
|
|
"epoch": 0.19849146486701072,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998886419522792,
|
|
"loss": 6.1605,
|
|
"mean_token_accuracy": 0.11577931866049766,
|
|
"num_tokens": 4393485.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.28572907447815,
|
|
"epoch": 0.19898769352917825,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998875245606919,
|
|
"loss": 5.9903,
|
|
"mean_token_accuracy": 0.12323976159095765,
|
|
"num_tokens": 4404354.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.138240671157837,
|
|
"epoch": 0.19948392219134578,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998864015923788,
|
|
"loss": 6.1308,
|
|
"mean_token_accuracy": 0.12045308947563171,
|
|
"num_tokens": 4414818.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.158795976638794,
|
|
"epoch": 0.1999801508535133,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998852730473674,
|
|
"loss": 6.0079,
|
|
"mean_token_accuracy": 0.11828904002904891,
|
|
"num_tokens": 4426271.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.15863299369812,
|
|
"epoch": 0.2004763795156808,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998841389256859,
|
|
"loss": 6.0545,
|
|
"mean_token_accuracy": 0.12046427428722381,
|
|
"num_tokens": 4436981.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.142259788513184,
|
|
"epoch": 0.20097260817784834,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998829992273622,
|
|
"loss": 5.8818,
|
|
"mean_token_accuracy": 0.12613770440220834,
|
|
"num_tokens": 4446494.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.0647642612457275,
|
|
"epoch": 0.20146883684001587,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998818539524248,
|
|
"loss": 5.9738,
|
|
"mean_token_accuracy": 0.1264504000544548,
|
|
"num_tokens": 4458535.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.1495154857635494,
|
|
"epoch": 0.2019650655021834,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499880703100902,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.1266962394118309,
|
|
"num_tokens": 4469025.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.134205341339111,
|
|
"epoch": 0.20246129416435094,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998795466728223,
|
|
"loss": 5.9595,
|
|
"mean_token_accuracy": 0.12471545413136483,
|
|
"num_tokens": 4478431.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.171464109420777,
|
|
"epoch": 0.20295752282651847,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998783846682144,
|
|
"loss": 6.0815,
|
|
"mean_token_accuracy": 0.12394077926874161,
|
|
"num_tokens": 4490215.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.105607700347901,
|
|
"epoch": 0.203453751488686,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004998772170871072,
|
|
"loss": 5.963,
|
|
"mean_token_accuracy": 0.1244163066148758,
|
|
"num_tokens": 4502895.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.12325873374939,
|
|
"epoch": 0.20394998015085353,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004998760439295295,
|
|
"loss": 5.9871,
|
|
"mean_token_accuracy": 0.12174513414502144,
|
|
"num_tokens": 4514468.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.173321723937988,
|
|
"epoch": 0.20444620881302103,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998748651955105,
|
|
"loss": 6.007,
|
|
"mean_token_accuracy": 0.12350882440805436,
|
|
"num_tokens": 4525787.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.1624654769897464,
|
|
"epoch": 0.20494243747518856,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004998736808850795,
|
|
"loss": 6.0868,
|
|
"mean_token_accuracy": 0.11808518916368485,
|
|
"num_tokens": 4537536.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.173154306411743,
|
|
"epoch": 0.2054386661373561,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998724909982656,
|
|
"loss": 6.0371,
|
|
"mean_token_accuracy": 0.12236888855695724,
|
|
"num_tokens": 4549619.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.107865715026856,
|
|
"epoch": 0.20593489479952362,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004998712955350986,
|
|
"loss": 5.9509,
|
|
"mean_token_accuracy": 0.12667887806892394,
|
|
"num_tokens": 4560248.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.182792615890503,
|
|
"epoch": 0.20643112346169115,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004998700944956081,
|
|
"loss": 6.0598,
|
|
"mean_token_accuracy": 0.12352480441331863,
|
|
"num_tokens": 4571888.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.149948072433472,
|
|
"epoch": 0.20692735212385868,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004998688878798237,
|
|
"loss": 5.9082,
|
|
"mean_token_accuracy": 0.13024957329034806,
|
|
"num_tokens": 4583234.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.173293352127075,
|
|
"epoch": 0.2074235807860262,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004998676756877753,
|
|
"loss": 6.1228,
|
|
"mean_token_accuracy": 0.11778421998023987,
|
|
"num_tokens": 4595252.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.100701141357422,
|
|
"epoch": 0.20791980944819372,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998664579194934,
|
|
"loss": 5.9016,
|
|
"mean_token_accuracy": 0.12487494945526123,
|
|
"num_tokens": 4607091.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.117470121383667,
|
|
"epoch": 0.20841603811036125,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004998652345750077,
|
|
"loss": 5.8611,
|
|
"mean_token_accuracy": 0.12737861275672913,
|
|
"num_tokens": 4618003.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.059277963638306,
|
|
"epoch": 0.20891226677252878,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004998640056543487,
|
|
"loss": 5.8767,
|
|
"mean_token_accuracy": 0.1214037574827671,
|
|
"num_tokens": 4628429.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.125901508331299,
|
|
"epoch": 0.2094084954346963,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998627711575469,
|
|
"loss": 5.9361,
|
|
"mean_token_accuracy": 0.13086853697896003,
|
|
"num_tokens": 4639414.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.141415405273437,
|
|
"epoch": 0.20990472409686384,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000499861531084633,
|
|
"loss": 6.022,
|
|
"mean_token_accuracy": 0.12312455773353577,
|
|
"num_tokens": 4652594.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.039756822586059,
|
|
"epoch": 0.21040095275903137,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998602854356376,
|
|
"loss": 5.8612,
|
|
"mean_token_accuracy": 0.12929336726665497,
|
|
"num_tokens": 4662981.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.109541177749634,
|
|
"epoch": 0.2108971814211989,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998590342105917,
|
|
"loss": 5.9639,
|
|
"mean_token_accuracy": 0.1239814892411232,
|
|
"num_tokens": 4672469.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.082083654403687,
|
|
"epoch": 0.21139341008336643,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998577774095262,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.12228829637169839,
|
|
"num_tokens": 4683640.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.100391864776611,
|
|
"epoch": 0.21188963874553393,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998565150324724,
|
|
"loss": 5.8938,
|
|
"mean_token_accuracy": 0.12874594032764436,
|
|
"num_tokens": 4694032.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.9822530269622805,
|
|
"epoch": 0.21238586740770146,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998552470794616,
|
|
"loss": 5.7792,
|
|
"mean_token_accuracy": 0.1315452888607979,
|
|
"num_tokens": 4704253.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.03460578918457,
|
|
"epoch": 0.212882096069869,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998539735505251,
|
|
"loss": 5.9305,
|
|
"mean_token_accuracy": 0.1236830122768879,
|
|
"num_tokens": 4715327.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.228457880020142,
|
|
"epoch": 0.21337832473203652,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998526944456946,
|
|
"loss": 5.9449,
|
|
"mean_token_accuracy": 0.12152760252356529,
|
|
"num_tokens": 4726450.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.073284006118774,
|
|
"epoch": 0.21387455339420405,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004998514097650018,
|
|
"loss": 5.9736,
|
|
"mean_token_accuracy": 0.12102751210331916,
|
|
"num_tokens": 4736630.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.08151159286499,
|
|
"epoch": 0.21437078205637158,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998501195084786,
|
|
"loss": 5.8635,
|
|
"mean_token_accuracy": 0.12787755504250525,
|
|
"num_tokens": 4746592.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.100288772583008,
|
|
"epoch": 0.21486701071853911,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998488236761569,
|
|
"loss": 6.0417,
|
|
"mean_token_accuracy": 0.12549156099557876,
|
|
"num_tokens": 4757824.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.139098644256592,
|
|
"epoch": 0.21536323938070662,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998475222680688,
|
|
"loss": 5.9766,
|
|
"mean_token_accuracy": 0.12785990089178084,
|
|
"num_tokens": 4768453.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.9953094005584715,
|
|
"epoch": 0.21585946804287415,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998462152842468,
|
|
"loss": 5.956,
|
|
"mean_token_accuracy": 0.1242900937795639,
|
|
"num_tokens": 4779664.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.162226676940918,
|
|
"epoch": 0.21635569670504168,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000499844902724723,
|
|
"loss": 5.969,
|
|
"mean_token_accuracy": 0.1265193648636341,
|
|
"num_tokens": 4791012.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.099493169784546,
|
|
"epoch": 0.2168519253672092,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998435845895302,
|
|
"loss": 5.841,
|
|
"mean_token_accuracy": 0.1277057059109211,
|
|
"num_tokens": 4800578.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.023715972900391,
|
|
"epoch": 0.21734815402937674,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499842260878701,
|
|
"loss": 5.8514,
|
|
"mean_token_accuracy": 0.12200021669268608,
|
|
"num_tokens": 4810940.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.081690168380737,
|
|
"epoch": 0.21784438269154427,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004998409315922683,
|
|
"loss": 5.9947,
|
|
"mean_token_accuracy": 0.12390753328800201,
|
|
"num_tokens": 4822696.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.113512086868286,
|
|
"epoch": 0.2183406113537118,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998395967302649,
|
|
"loss": 5.8774,
|
|
"mean_token_accuracy": 0.13284592926502228,
|
|
"num_tokens": 4833669.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.167998266220093,
|
|
"epoch": 0.21883684001587933,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499838256292724,
|
|
"loss": 6.0529,
|
|
"mean_token_accuracy": 0.12145362943410873,
|
|
"num_tokens": 4844021.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.9783563137054445,
|
|
"epoch": 0.21933306867804683,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998369102796787,
|
|
"loss": 5.9243,
|
|
"mean_token_accuracy": 0.12736267521977424,
|
|
"num_tokens": 4855336.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.101631212234497,
|
|
"epoch": 0.21982929734021436,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004998355586911627,
|
|
"loss": 5.9299,
|
|
"mean_token_accuracy": 0.1268913999199867,
|
|
"num_tokens": 4867713.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.054467105865479,
|
|
"epoch": 0.2203255260023819,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004998342015272093,
|
|
"loss": 5.9026,
|
|
"mean_token_accuracy": 0.12375652343034745,
|
|
"num_tokens": 4879134.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.114092779159546,
|
|
"epoch": 0.22082175466454942,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004998328387878521,
|
|
"loss": 6.0599,
|
|
"mean_token_accuracy": 0.12487743720412255,
|
|
"num_tokens": 4891755.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.203889560699463,
|
|
"epoch": 0.22131798332671695,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998314704731249,
|
|
"loss": 6.0604,
|
|
"mean_token_accuracy": 0.11887223795056343,
|
|
"num_tokens": 4903264.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.066045093536377,
|
|
"epoch": 0.22181421198888449,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998300965830618,
|
|
"loss": 5.8843,
|
|
"mean_token_accuracy": 0.12301613837480545,
|
|
"num_tokens": 4912533.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.971320104598999,
|
|
"epoch": 0.22231044065105202,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004998287171176968,
|
|
"loss": 5.876,
|
|
"mean_token_accuracy": 0.12986253574490547,
|
|
"num_tokens": 4923288.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.049497413635254,
|
|
"epoch": 0.22280666931321952,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004998273320770641,
|
|
"loss": 5.8896,
|
|
"mean_token_accuracy": 0.12646428495645523,
|
|
"num_tokens": 4934759.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.087690019607544,
|
|
"epoch": 0.22330289797538705,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499825941461198,
|
|
"loss": 5.9523,
|
|
"mean_token_accuracy": 0.12756217569112777,
|
|
"num_tokens": 4947024.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.019892168045044,
|
|
"epoch": 0.22379912663755458,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004998245452701329,
|
|
"loss": 5.8962,
|
|
"mean_token_accuracy": 0.1269862487912178,
|
|
"num_tokens": 4959281.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.124982738494873,
|
|
"epoch": 0.2242953552997221,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998231435039036,
|
|
"loss": 5.9385,
|
|
"mean_token_accuracy": 0.11895076110959053,
|
|
"num_tokens": 4970466.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.050164318084716,
|
|
"epoch": 0.22479158396188964,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004998217361625449,
|
|
"loss": 5.8953,
|
|
"mean_token_accuracy": 0.13044259697198868,
|
|
"num_tokens": 4981135.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.087223243713379,
|
|
"epoch": 0.22528781262405717,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998203232460915,
|
|
"loss": 5.9574,
|
|
"mean_token_accuracy": 0.13070524409413337,
|
|
"num_tokens": 4991292.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.04678053855896,
|
|
"epoch": 0.2257840412862247,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998189047545785,
|
|
"loss": 5.9151,
|
|
"mean_token_accuracy": 0.12900153622031213,
|
|
"num_tokens": 5002806.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.015051984786988,
|
|
"epoch": 0.22628026994839223,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998174806880411,
|
|
"loss": 5.8519,
|
|
"mean_token_accuracy": 0.1365997314453125,
|
|
"num_tokens": 5013474.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.111403083801269,
|
|
"epoch": 0.22677649861055973,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004998160510465146,
|
|
"loss": 5.895,
|
|
"mean_token_accuracy": 0.13188145235180854,
|
|
"num_tokens": 5025151.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.943822288513184,
|
|
"epoch": 0.22727272727272727,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004998146158300345,
|
|
"loss": 5.8967,
|
|
"mean_token_accuracy": 0.13392265886068344,
|
|
"num_tokens": 5037166.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.117561721801758,
|
|
"epoch": 0.2277689559348948,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004998131750386364,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.12657614946365356,
|
|
"num_tokens": 5047452.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.033935308456421,
|
|
"epoch": 0.22826518459706233,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004998117286723559,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.1220039039850235,
|
|
"num_tokens": 5058501.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.084085893630982,
|
|
"epoch": 0.22876141325922986,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499810276731229,
|
|
"loss": 5.9417,
|
|
"mean_token_accuracy": 0.12538052052259446,
|
|
"num_tokens": 5069716.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.188976240158081,
|
|
"epoch": 0.2292576419213974,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998088192152918,
|
|
"loss": 5.9239,
|
|
"mean_token_accuracy": 0.12268051952123642,
|
|
"num_tokens": 5078509.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.029656267166137,
|
|
"epoch": 0.22975387058356492,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998073561245801,
|
|
"loss": 5.9546,
|
|
"mean_token_accuracy": 0.12687286511063575,
|
|
"num_tokens": 5090355.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.091899156570435,
|
|
"epoch": 0.23025009924573242,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004998058874591305,
|
|
"loss": 5.8536,
|
|
"mean_token_accuracy": 0.13038059920072556,
|
|
"num_tokens": 5100582.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.0615569114685055,
|
|
"epoch": 0.23074632790789995,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004998044132189793,
|
|
"loss": 5.9184,
|
|
"mean_token_accuracy": 0.12220206931233406,
|
|
"num_tokens": 5112176.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.061944389343262,
|
|
"epoch": 0.23124255657006748,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499802933404163,
|
|
"loss": 5.8503,
|
|
"mean_token_accuracy": 0.12976043820381164,
|
|
"num_tokens": 5122840.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.06622462272644,
|
|
"epoch": 0.231738785232235,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004998014480147183,
|
|
"loss": 5.83,
|
|
"mean_token_accuracy": 0.1355508543550968,
|
|
"num_tokens": 5132926.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.981128120422364,
|
|
"epoch": 0.23223501389440254,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997999570506823,
|
|
"loss": 5.8163,
|
|
"mean_token_accuracy": 0.12751471549272536,
|
|
"num_tokens": 5143394.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.014122343063354,
|
|
"epoch": 0.23273124255657007,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004997984605120917,
|
|
"loss": 5.8538,
|
|
"mean_token_accuracy": 0.12407195791602135,
|
|
"num_tokens": 5155710.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.8755535125732425,
|
|
"epoch": 0.2332274712187376,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997969583989835,
|
|
"loss": 5.7797,
|
|
"mean_token_accuracy": 0.12736281827092172,
|
|
"num_tokens": 5167065.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.026074457168579,
|
|
"epoch": 0.23372369988090513,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997954507113952,
|
|
"loss": 5.875,
|
|
"mean_token_accuracy": 0.12745090574026108,
|
|
"num_tokens": 5177535.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.040788745880127,
|
|
"epoch": 0.23421992854307264,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004997939374493642,
|
|
"loss": 5.8063,
|
|
"mean_token_accuracy": 0.13261814862489701,
|
|
"num_tokens": 5188324.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.968220281600952,
|
|
"epoch": 0.23471615720524017,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000499792418612928,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.1319134347140789,
|
|
"num_tokens": 5199223.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.008479738235474,
|
|
"epoch": 0.2352123858674077,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997908942021241,
|
|
"loss": 5.8617,
|
|
"mean_token_accuracy": 0.13253337219357492,
|
|
"num_tokens": 5210080.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.121827793121338,
|
|
"epoch": 0.23570861452957523,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997893642169904,
|
|
"loss": 5.9825,
|
|
"mean_token_accuracy": 0.12802610248327256,
|
|
"num_tokens": 5220867.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.06226658821106,
|
|
"epoch": 0.23620484319174276,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997878286575648,
|
|
"loss": 5.8316,
|
|
"mean_token_accuracy": 0.1282677799463272,
|
|
"num_tokens": 5231451.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.996644639968872,
|
|
"epoch": 0.2367010718539103,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997862875238855,
|
|
"loss": 5.9339,
|
|
"mean_token_accuracy": 0.12390372604131698,
|
|
"num_tokens": 5244643.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.055671977996826,
|
|
"epoch": 0.23719730051607782,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997847408159907,
|
|
"loss": 5.9503,
|
|
"mean_token_accuracy": 0.12081070691347122,
|
|
"num_tokens": 5256279.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.094895601272583,
|
|
"epoch": 0.23769352917824532,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004997831885339186,
|
|
"loss": 5.9201,
|
|
"mean_token_accuracy": 0.12661861553788184,
|
|
"num_tokens": 5268498.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.0079608917236325,
|
|
"epoch": 0.23818975784041285,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004997816306777077,
|
|
"loss": 5.8873,
|
|
"mean_token_accuracy": 0.13006770759820938,
|
|
"num_tokens": 5280869.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.153380107879639,
|
|
"epoch": 0.23868598650258038,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997800672473969,
|
|
"loss": 5.9915,
|
|
"mean_token_accuracy": 0.1220635175704956,
|
|
"num_tokens": 5293082.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.005958080291748,
|
|
"epoch": 0.2391822151647479,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997784982430247,
|
|
"loss": 5.7688,
|
|
"mean_token_accuracy": 0.1395465165376663,
|
|
"num_tokens": 5304039.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.969437026977539,
|
|
"epoch": 0.23967844382691544,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004997769236646301,
|
|
"loss": 5.8503,
|
|
"mean_token_accuracy": 0.13086041882634164,
|
|
"num_tokens": 5315016.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.144852113723755,
|
|
"epoch": 0.24017467248908297,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004997753435122522,
|
|
"loss": 5.9808,
|
|
"mean_token_accuracy": 0.12935518026351928,
|
|
"num_tokens": 5326090.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.986780261993408,
|
|
"epoch": 0.2406709011512505,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00049977375778593,
|
|
"loss": 5.778,
|
|
"mean_token_accuracy": 0.12515551522374152,
|
|
"num_tokens": 5336423.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.017197036743164,
|
|
"epoch": 0.24116712981341804,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004997721664857032,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.1286918543279171,
|
|
"num_tokens": 5346982.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.0123132228851315,
|
|
"epoch": 0.24166335847558554,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997705696116108,
|
|
"loss": 5.829,
|
|
"mean_token_accuracy": 0.13723039478063584,
|
|
"num_tokens": 5359210.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.017412328720093,
|
|
"epoch": 0.24215958713775307,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004997689671636926,
|
|
"loss": 5.8603,
|
|
"mean_token_accuracy": 0.12343929260969162,
|
|
"num_tokens": 5369891.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.015021753311157,
|
|
"epoch": 0.2426558157999206,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004997673591419884,
|
|
"loss": 5.8397,
|
|
"mean_token_accuracy": 0.12764633148908616,
|
|
"num_tokens": 5380850.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.030062103271485,
|
|
"epoch": 0.24315204446208813,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004997657455465381,
|
|
"loss": 5.8985,
|
|
"mean_token_accuracy": 0.12984421476721764,
|
|
"num_tokens": 5391903.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.048585891723633,
|
|
"epoch": 0.24364827312425566,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997641263773815,
|
|
"loss": 5.9466,
|
|
"mean_token_accuracy": 0.12775701135396958,
|
|
"num_tokens": 5402517.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.002290296554565,
|
|
"epoch": 0.2441445017864232,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000499762501634559,
|
|
"loss": 5.8711,
|
|
"mean_token_accuracy": 0.128414898365736,
|
|
"num_tokens": 5414110.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.061224746704101,
|
|
"epoch": 0.24464073044859072,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997608713181108,
|
|
"loss": 5.8641,
|
|
"mean_token_accuracy": 0.13066134005784988,
|
|
"num_tokens": 5424466.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.991046953201294,
|
|
"epoch": 0.24513695911075822,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997592354280772,
|
|
"loss": 5.8833,
|
|
"mean_token_accuracy": 0.1335628919303417,
|
|
"num_tokens": 5434461.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.993779897689819,
|
|
"epoch": 0.24563318777292575,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000499757593964499,
|
|
"loss": 5.8595,
|
|
"mean_token_accuracy": 0.12100402787327766,
|
|
"num_tokens": 5445921.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.966112470626831,
|
|
"epoch": 0.24612941643509328,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997559469274166,
|
|
"loss": 5.8419,
|
|
"mean_token_accuracy": 0.12944890931248665,
|
|
"num_tokens": 5456283.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.00761137008667,
|
|
"epoch": 0.24662564509726081,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004997542943168712,
|
|
"loss": 5.8772,
|
|
"mean_token_accuracy": 0.12770011499524117,
|
|
"num_tokens": 5467616.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.989387559890747,
|
|
"epoch": 0.24712187375942835,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004997526361329035,
|
|
"loss": 5.8305,
|
|
"mean_token_accuracy": 0.13280585035681725,
|
|
"num_tokens": 5478050.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.101069402694702,
|
|
"epoch": 0.24761810242159588,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997509723755547,
|
|
"loss": 5.8426,
|
|
"mean_token_accuracy": 0.13170866221189498,
|
|
"num_tokens": 5487185.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.979424858093262,
|
|
"epoch": 0.2481143310837634,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004997493030448662,
|
|
"loss": 5.8727,
|
|
"mean_token_accuracy": 0.1261827602982521,
|
|
"num_tokens": 5498802.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.974308633804322,
|
|
"epoch": 0.24861055974593094,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997476281408793,
|
|
"loss": 5.8782,
|
|
"mean_token_accuracy": 0.12776481211185456,
|
|
"num_tokens": 5510285.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.030932426452637,
|
|
"epoch": 0.24910678840809844,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004997459476636354,
|
|
"loss": 5.7601,
|
|
"mean_token_accuracy": 0.1376047044992447,
|
|
"num_tokens": 5520028.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.963772487640381,
|
|
"epoch": 0.24960301707026597,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997442616131763,
|
|
"loss": 5.8871,
|
|
"mean_token_accuracy": 0.1257816217839718,
|
|
"num_tokens": 5529271.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.042953777313232,
|
|
"epoch": 0.25009924573243353,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997425699895438,
|
|
"loss": 5.8057,
|
|
"mean_token_accuracy": 0.12424818351864815,
|
|
"num_tokens": 5539396.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.0163270950317385,
|
|
"epoch": 0.25059547439460106,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004997408727927798,
|
|
"loss": 5.9428,
|
|
"mean_token_accuracy": 0.12824917510151862,
|
|
"num_tokens": 5551674.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.025873517990112,
|
|
"epoch": 0.25109170305676853,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997391700229264,
|
|
"loss": 5.8561,
|
|
"mean_token_accuracy": 0.1312730759382248,
|
|
"num_tokens": 5562749.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.001038408279419,
|
|
"epoch": 0.25158793171893606,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000499737461680026,
|
|
"loss": 5.8772,
|
|
"mean_token_accuracy": 0.13404013216495514,
|
|
"num_tokens": 5573478.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.929586935043335,
|
|
"epoch": 0.2520841603811036,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997357477641207,
|
|
"loss": 5.8682,
|
|
"mean_token_accuracy": 0.135566084086895,
|
|
"num_tokens": 5584546.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.922780466079712,
|
|
"epoch": 0.2525803890432711,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997340282752532,
|
|
"loss": 5.7439,
|
|
"mean_token_accuracy": 0.13249244838953017,
|
|
"num_tokens": 5594004.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.023516273498535,
|
|
"epoch": 0.25307661770543866,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499732303213466,
|
|
"loss": 5.8103,
|
|
"mean_token_accuracy": 0.12287582755088806,
|
|
"num_tokens": 5606095.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.03965277671814,
|
|
"epoch": 0.2535728463676062,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997305725788019,
|
|
"loss": 5.8433,
|
|
"mean_token_accuracy": 0.12945516034960747,
|
|
"num_tokens": 5615596.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.911628723144531,
|
|
"epoch": 0.2540690750297737,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000499728836371304,
|
|
"loss": 5.7889,
|
|
"mean_token_accuracy": 0.13019108325242995,
|
|
"num_tokens": 5626753.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.890804100036621,
|
|
"epoch": 0.25456530369194125,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997270945910151,
|
|
"loss": 5.8127,
|
|
"mean_token_accuracy": 0.1270523451268673,
|
|
"num_tokens": 5636135.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.078077507019043,
|
|
"epoch": 0.2550615323541088,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004997253472379785,
|
|
"loss": 5.8847,
|
|
"mean_token_accuracy": 0.12609419897198676,
|
|
"num_tokens": 5647074.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.022387552261352,
|
|
"epoch": 0.2555577610162763,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004997235943122375,
|
|
"loss": 5.9047,
|
|
"mean_token_accuracy": 0.12348231971263886,
|
|
"num_tokens": 5658600.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.92790265083313,
|
|
"epoch": 0.25605398967844384,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997218358138356,
|
|
"loss": 5.8006,
|
|
"mean_token_accuracy": 0.12867250964045523,
|
|
"num_tokens": 5668453.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.022593784332275,
|
|
"epoch": 0.25655021834061137,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997200717428165,
|
|
"loss": 5.8552,
|
|
"mean_token_accuracy": 0.12913168221712112,
|
|
"num_tokens": 5678855.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.977180624008179,
|
|
"epoch": 0.2570464470027789,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997183020992237,
|
|
"loss": 5.9078,
|
|
"mean_token_accuracy": 0.12221302613615989,
|
|
"num_tokens": 5690063.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.008505201339721,
|
|
"epoch": 0.25754267566494643,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997165268831014,
|
|
"loss": 5.8082,
|
|
"mean_token_accuracy": 0.1332797534763813,
|
|
"num_tokens": 5701487.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.929228925704956,
|
|
"epoch": 0.25803890432711396,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004997147460944933,
|
|
"loss": 5.8037,
|
|
"mean_token_accuracy": 0.129875248670578,
|
|
"num_tokens": 5713034.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.02344069480896,
|
|
"epoch": 0.25853513298928144,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004997129597334438,
|
|
"loss": 5.8529,
|
|
"mean_token_accuracy": 0.13128544464707376,
|
|
"num_tokens": 5725469.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.925415468215943,
|
|
"epoch": 0.25903136165144897,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004997111677999971,
|
|
"loss": 5.8374,
|
|
"mean_token_accuracy": 0.1300521619617939,
|
|
"num_tokens": 5737099.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.973599290847778,
|
|
"epoch": 0.2595275903136165,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004997093702941976,
|
|
"loss": 5.8495,
|
|
"mean_token_accuracy": 0.12512658536434174,
|
|
"num_tokens": 5748756.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.009793472290039,
|
|
"epoch": 0.260023818975784,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00049970756721609,
|
|
"loss": 5.8378,
|
|
"mean_token_accuracy": 0.13358309268951415,
|
|
"num_tokens": 5759598.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.976476240158081,
|
|
"epoch": 0.26052004763795156,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000499705758565719,
|
|
"loss": 5.8817,
|
|
"mean_token_accuracy": 0.12957399263978003,
|
|
"num_tokens": 5770397.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.043560457229614,
|
|
"epoch": 0.2610162763001191,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997039443431292,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.12363751754164695,
|
|
"num_tokens": 5780935.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.038046789169312,
|
|
"epoch": 0.2615125049622866,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997021245483659,
|
|
"loss": 5.8193,
|
|
"mean_token_accuracy": 0.13055009618401528,
|
|
"num_tokens": 5791557.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.90911078453064,
|
|
"epoch": 0.26200873362445415,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000499700299181474,
|
|
"loss": 5.8476,
|
|
"mean_token_accuracy": 0.1279944323003292,
|
|
"num_tokens": 5802233.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.943614435195923,
|
|
"epoch": 0.2625049622866217,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004996984682424989,
|
|
"loss": 5.7656,
|
|
"mean_token_accuracy": 0.13612023070454599,
|
|
"num_tokens": 5813424.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.928361797332764,
|
|
"epoch": 0.2630011909487892,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000499696631731486,
|
|
"loss": 5.7455,
|
|
"mean_token_accuracy": 0.1320380948483944,
|
|
"num_tokens": 5825652.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.011695718765258,
|
|
"epoch": 0.26349741961095674,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996947896484809,
|
|
"loss": 5.8357,
|
|
"mean_token_accuracy": 0.13260062411427498,
|
|
"num_tokens": 5836603.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.954362440109253,
|
|
"epoch": 0.26399364827312427,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000499692941993529,
|
|
"loss": 5.8881,
|
|
"mean_token_accuracy": 0.13098345547914506,
|
|
"num_tokens": 5849365.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.022516059875488,
|
|
"epoch": 0.2644898769352918,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004996910887666765,
|
|
"loss": 5.8082,
|
|
"mean_token_accuracy": 0.13563585206866263,
|
|
"num_tokens": 5860305.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.918445682525634,
|
|
"epoch": 0.26498610559745933,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000499689229967969,
|
|
"loss": 5.7518,
|
|
"mean_token_accuracy": 0.13243596255779266,
|
|
"num_tokens": 5872272.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.965515995025635,
|
|
"epoch": 0.26548233425962686,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996873655974529,
|
|
"loss": 5.7807,
|
|
"mean_token_accuracy": 0.12951171174645423,
|
|
"num_tokens": 5884107.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.860142564773559,
|
|
"epoch": 0.26597856292179434,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004996854956551743,
|
|
"loss": 5.7719,
|
|
"mean_token_accuracy": 0.1379713624715805,
|
|
"num_tokens": 5894060.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.913548135757447,
|
|
"epoch": 0.26647479158396187,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004996836201411794,
|
|
"loss": 5.7915,
|
|
"mean_token_accuracy": 0.13487917557358742,
|
|
"num_tokens": 5904804.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.072554445266723,
|
|
"epoch": 0.2669710202461294,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499681739055515,
|
|
"loss": 5.8487,
|
|
"mean_token_accuracy": 0.13210411742329597,
|
|
"num_tokens": 5916496.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.941082620620728,
|
|
"epoch": 0.26746724890829693,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996798523982276,
|
|
"loss": 5.7944,
|
|
"mean_token_accuracy": 0.12776002287864685,
|
|
"num_tokens": 5926920.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.002683544158936,
|
|
"epoch": 0.26796347757046446,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000499677960169364,
|
|
"loss": 5.9145,
|
|
"mean_token_accuracy": 0.12489431127905845,
|
|
"num_tokens": 5940098.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.963934755325317,
|
|
"epoch": 0.268459706232632,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004996760623689711,
|
|
"loss": 5.7692,
|
|
"mean_token_accuracy": 0.13152637630701064,
|
|
"num_tokens": 5949687.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.962456846237183,
|
|
"epoch": 0.2689559348947995,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000499674158997096,
|
|
"loss": 5.8761,
|
|
"mean_token_accuracy": 0.12971905767917633,
|
|
"num_tokens": 5960777.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.018963384628296,
|
|
"epoch": 0.26945216355696705,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996722500537859,
|
|
"loss": 5.7802,
|
|
"mean_token_accuracy": 0.1312619134783745,
|
|
"num_tokens": 5972298.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.9298717975616455,
|
|
"epoch": 0.2699483922191346,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004996703355390881,
|
|
"loss": 5.8988,
|
|
"mean_token_accuracy": 0.12656683698296547,
|
|
"num_tokens": 5985613.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.017775201797486,
|
|
"epoch": 0.2704446208813021,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00049966841545305,
|
|
"loss": 5.954,
|
|
"mean_token_accuracy": 0.12761470898985863,
|
|
"num_tokens": 5997530.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.011219215393067,
|
|
"epoch": 0.27094084954346964,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004996664897957195,
|
|
"loss": 5.8209,
|
|
"mean_token_accuracy": 0.12617397606372832,
|
|
"num_tokens": 6008123.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.960656690597534,
|
|
"epoch": 0.2714370782056372,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004996645585671441,
|
|
"loss": 5.8355,
|
|
"mean_token_accuracy": 0.13165733143687247,
|
|
"num_tokens": 6020578.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.944126558303833,
|
|
"epoch": 0.2719333068678047,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004996626217673717,
|
|
"loss": 5.7613,
|
|
"mean_token_accuracy": 0.13792742118239404,
|
|
"num_tokens": 6031625.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.88871202468872,
|
|
"epoch": 0.27242953552997223,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004996606793964504,
|
|
"loss": 5.8008,
|
|
"mean_token_accuracy": 0.13623525723814964,
|
|
"num_tokens": 6041320.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.01530237197876,
|
|
"epoch": 0.27292576419213976,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004996587314544284,
|
|
"loss": 5.8187,
|
|
"mean_token_accuracy": 0.1311524450778961,
|
|
"num_tokens": 6052627.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.916905069351197,
|
|
"epoch": 0.27342199285430724,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004996567779413539,
|
|
"loss": 5.7701,
|
|
"mean_token_accuracy": 0.13783232718706132,
|
|
"num_tokens": 6063468.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.931315326690674,
|
|
"epoch": 0.27391822151647477,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004996548188572755,
|
|
"loss": 5.7497,
|
|
"mean_token_accuracy": 0.1367427311837673,
|
|
"num_tokens": 6074208.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.987172603607178,
|
|
"epoch": 0.2744144501786423,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004996528542022416,
|
|
"loss": 5.7967,
|
|
"mean_token_accuracy": 0.13130253031849862,
|
|
"num_tokens": 6085068.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.929187154769897,
|
|
"epoch": 0.27491067884080983,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000499650883976301,
|
|
"loss": 5.8003,
|
|
"mean_token_accuracy": 0.1321723371744156,
|
|
"num_tokens": 6096445.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.849560213088989,
|
|
"epoch": 0.27540690750297736,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004996489081795026,
|
|
"loss": 5.7075,
|
|
"mean_token_accuracy": 0.13662762194871902,
|
|
"num_tokens": 6106527.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.87761869430542,
|
|
"epoch": 0.2759031361651449,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004996469268118953,
|
|
"loss": 5.7365,
|
|
"mean_token_accuracy": 0.13877005353569985,
|
|
"num_tokens": 6115923.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.892449617385864,
|
|
"epoch": 0.2763993648273124,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004996449398735283,
|
|
"loss": 5.6192,
|
|
"mean_token_accuracy": 0.14092544242739677,
|
|
"num_tokens": 6125861.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.953143739700318,
|
|
"epoch": 0.27689559348947995,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004996429473644509,
|
|
"loss": 5.8347,
|
|
"mean_token_accuracy": 0.12965964078903197,
|
|
"num_tokens": 6137189.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.9353728771209715,
|
|
"epoch": 0.2773918221516475,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004996409492847124,
|
|
"loss": 5.7542,
|
|
"mean_token_accuracy": 0.13289039209485054,
|
|
"num_tokens": 6147920.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.957181549072265,
|
|
"epoch": 0.277888050813815,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004996389456343624,
|
|
"loss": 5.7504,
|
|
"mean_token_accuracy": 0.13782199770212172,
|
|
"num_tokens": 6159371.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.024093103408814,
|
|
"epoch": 0.27838427947598254,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004996369364134507,
|
|
"loss": 5.9497,
|
|
"mean_token_accuracy": 0.12122881636023522,
|
|
"num_tokens": 6172679.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.923732948303223,
|
|
"epoch": 0.2788805081381501,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499634921622027,
|
|
"loss": 5.7933,
|
|
"mean_token_accuracy": 0.12945920675992967,
|
|
"num_tokens": 6184244.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.903921222686767,
|
|
"epoch": 0.2793767368003176,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004996329012601413,
|
|
"loss": 5.681,
|
|
"mean_token_accuracy": 0.13114461600780486,
|
|
"num_tokens": 6194826.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.968857002258301,
|
|
"epoch": 0.27987296546248513,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004996308753278437,
|
|
"loss": 5.8212,
|
|
"mean_token_accuracy": 0.13206653594970702,
|
|
"num_tokens": 6206808.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.952521276473999,
|
|
"epoch": 0.28036919412465267,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004996288438251843,
|
|
"loss": 5.8217,
|
|
"mean_token_accuracy": 0.12705112993717194,
|
|
"num_tokens": 6217303.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.962418413162231,
|
|
"epoch": 0.28086542278682014,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996268067522137,
|
|
"loss": 5.8215,
|
|
"mean_token_accuracy": 0.12550831884145736,
|
|
"num_tokens": 6227965.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.0486168384552,
|
|
"epoch": 0.28136165144898767,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996247641089823,
|
|
"loss": 5.8799,
|
|
"mean_token_accuracy": 0.12784553319215775,
|
|
"num_tokens": 6239011.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.873768758773804,
|
|
"epoch": 0.2818578801111552,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004996227158955409,
|
|
"loss": 5.7375,
|
|
"mean_token_accuracy": 0.13439323455095292,
|
|
"num_tokens": 6249084.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.913141870498658,
|
|
"epoch": 0.28235410877332273,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00049962066211194,
|
|
"loss": 5.8041,
|
|
"mean_token_accuracy": 0.12883702740073205,
|
|
"num_tokens": 6260706.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.906771898269653,
|
|
"epoch": 0.28285033743549026,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996186027582308,
|
|
"loss": 5.7267,
|
|
"mean_token_accuracy": 0.13956455439329146,
|
|
"num_tokens": 6272848.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.91385931968689,
|
|
"epoch": 0.2833465660976578,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996165378344642,
|
|
"loss": 5.741,
|
|
"mean_token_accuracy": 0.13406383767724037,
|
|
"num_tokens": 6283820.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.939226245880127,
|
|
"epoch": 0.2838427947598253,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996144673406914,
|
|
"loss": 5.915,
|
|
"mean_token_accuracy": 0.12336270064115525,
|
|
"num_tokens": 6294193.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.937975311279297,
|
|
"epoch": 0.28433902342199285,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004996123912769639,
|
|
"loss": 5.7329,
|
|
"mean_token_accuracy": 0.14111651629209518,
|
|
"num_tokens": 6304407.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.87665205001831,
|
|
"epoch": 0.2848352520841604,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004996103096433332,
|
|
"loss": 5.8212,
|
|
"mean_token_accuracy": 0.13326509818434715,
|
|
"num_tokens": 6316311.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.942044496536255,
|
|
"epoch": 0.2853314807463279,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004996082224398506,
|
|
"loss": 5.7734,
|
|
"mean_token_accuracy": 0.13422515317797662,
|
|
"num_tokens": 6327248.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.964761114120483,
|
|
"epoch": 0.28582770940849545,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004996061296665682,
|
|
"loss": 5.7933,
|
|
"mean_token_accuracy": 0.131377661973238,
|
|
"num_tokens": 6338494.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.899775981903076,
|
|
"epoch": 0.286323938070663,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996040313235377,
|
|
"loss": 5.6797,
|
|
"mean_token_accuracy": 0.1386905901134014,
|
|
"num_tokens": 6349038.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.949228382110595,
|
|
"epoch": 0.2868201667328305,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004996019274108112,
|
|
"loss": 5.8067,
|
|
"mean_token_accuracy": 0.12893664464354515,
|
|
"num_tokens": 6360180.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.813185453414917,
|
|
"epoch": 0.28731639539499804,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499599817928441,
|
|
"loss": 5.7393,
|
|
"mean_token_accuracy": 0.13658856898546218,
|
|
"num_tokens": 6370931.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.841475248336792,
|
|
"epoch": 0.28781262405716557,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004995977028764792,
|
|
"loss": 5.7214,
|
|
"mean_token_accuracy": 0.13749603554606438,
|
|
"num_tokens": 6381810.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.949096298217773,
|
|
"epoch": 0.28830885271933304,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004995955822549783,
|
|
"loss": 5.7294,
|
|
"mean_token_accuracy": 0.1378832995891571,
|
|
"num_tokens": 6392268.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.916205501556396,
|
|
"epoch": 0.2888050813815006,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004995934560639909,
|
|
"loss": 5.7899,
|
|
"mean_token_accuracy": 0.13368103802204132,
|
|
"num_tokens": 6403650.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.920439100265503,
|
|
"epoch": 0.2893013100436681,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995913243035699,
|
|
"loss": 5.7577,
|
|
"mean_token_accuracy": 0.1311557449400425,
|
|
"num_tokens": 6414672.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.954520988464355,
|
|
"epoch": 0.28979753870583563,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995891869737678,
|
|
"loss": 5.8282,
|
|
"mean_token_accuracy": 0.13221259862184526,
|
|
"num_tokens": 6424916.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.9455653667449955,
|
|
"epoch": 0.29029376736800316,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995870440746379,
|
|
"loss": 5.7778,
|
|
"mean_token_accuracy": 0.13875048831105233,
|
|
"num_tokens": 6435950.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.8589989185333256,
|
|
"epoch": 0.2907899960301707,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004995848956062333,
|
|
"loss": 5.7475,
|
|
"mean_token_accuracy": 0.13245701491832734,
|
|
"num_tokens": 6446336.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.905734395980835,
|
|
"epoch": 0.2912862246923382,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004995827415686071,
|
|
"loss": 5.8066,
|
|
"mean_token_accuracy": 0.12863307669758797,
|
|
"num_tokens": 6457655.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.950538587570191,
|
|
"epoch": 0.29178245335450576,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499580581961813,
|
|
"loss": 5.6973,
|
|
"mean_token_accuracy": 0.13707054033875465,
|
|
"num_tokens": 6469627.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.882530784606933,
|
|
"epoch": 0.2922786820166733,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004995784167859043,
|
|
"loss": 5.7915,
|
|
"mean_token_accuracy": 0.1303873836994171,
|
|
"num_tokens": 6480351.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.980832910537719,
|
|
"epoch": 0.2927749106788408,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004995762460409349,
|
|
"loss": 5.8253,
|
|
"mean_token_accuracy": 0.1353992909193039,
|
|
"num_tokens": 6491651.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.9034021377563475,
|
|
"epoch": 0.29327113934100835,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004995740697269584,
|
|
"loss": 5.6734,
|
|
"mean_token_accuracy": 0.13603452891111373,
|
|
"num_tokens": 6502531.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.818338775634766,
|
|
"epoch": 0.2937673680031759,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004995718878440289,
|
|
"loss": 5.6721,
|
|
"mean_token_accuracy": 0.13839882612228394,
|
|
"num_tokens": 6511938.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.965565156936646,
|
|
"epoch": 0.2942635966653434,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995697003922006,
|
|
"loss": 5.8283,
|
|
"mean_token_accuracy": 0.12819456234574317,
|
|
"num_tokens": 6522775.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.922029304504394,
|
|
"epoch": 0.29475982532751094,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004995675073715274,
|
|
"loss": 5.6627,
|
|
"mean_token_accuracy": 0.1319277822971344,
|
|
"num_tokens": 6534660.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.8830774307250975,
|
|
"epoch": 0.29525605398967847,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004995653087820641,
|
|
"loss": 5.7927,
|
|
"mean_token_accuracy": 0.13659714311361312,
|
|
"num_tokens": 6545673.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.924934196472168,
|
|
"epoch": 0.29575228265184594,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499563104623865,
|
|
"loss": 5.7547,
|
|
"mean_token_accuracy": 0.13512978926301003,
|
|
"num_tokens": 6556912.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.935703706741333,
|
|
"epoch": 0.2962485113140135,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004995608948969849,
|
|
"loss": 5.8517,
|
|
"mean_token_accuracy": 0.12844418361783028,
|
|
"num_tokens": 6567532.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.927107429504394,
|
|
"epoch": 0.296744739976181,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995586796014784,
|
|
"loss": 5.8246,
|
|
"mean_token_accuracy": 0.13212826922535897,
|
|
"num_tokens": 6577830.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.860514736175537,
|
|
"epoch": 0.29724096863834853,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004995564587374005,
|
|
"loss": 5.7143,
|
|
"mean_token_accuracy": 0.13365908712148666,
|
|
"num_tokens": 6587254.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.91803126335144,
|
|
"epoch": 0.29773719730051607,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995542323048063,
|
|
"loss": 5.9097,
|
|
"mean_token_accuracy": 0.1270786426961422,
|
|
"num_tokens": 6598294.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.29773719730051607,
|
|
"eval_entropy": 5.736310448491338,
|
|
"eval_loss": 5.833684921264648,
|
|
"eval_mean_token_accuracy": 0.1343776344935248,
|
|
"eval_num_tokens": 6598294.0,
|
|
"eval_runtime": 20.4598,
|
|
"eval_samples_per_second": 1645.621,
|
|
"eval_steps_per_second": 205.721,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 100750,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 3000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1645030401638400.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|