ae7fb579b0
Model: fpadovani/eng-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
8079 lines
219 KiB
JSON
8079 lines
219 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.324499729583558,
|
|
"eval_steps": 1000,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.051201391220093,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.7284,
|
|
"mean_token_accuracy": 0.12397327944636345,
|
|
"num_tokens": 10948.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.085569572448731,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.6671,
|
|
"mean_token_accuracy": 0.1256408005952835,
|
|
"num_tokens": 21455.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.026061630249023,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.7001,
|
|
"mean_token_accuracy": 0.12955478951334953,
|
|
"num_tokens": 32377.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.060545969009399,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.6059,
|
|
"mean_token_accuracy": 0.12464395985007286,
|
|
"num_tokens": 42276.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.05833740234375,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.6772,
|
|
"mean_token_accuracy": 0.12779081240296364,
|
|
"num_tokens": 55049.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.0409410953521725,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.5771,
|
|
"mean_token_accuracy": 0.12367336973547935,
|
|
"num_tokens": 66792.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.047565269470215,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.6946,
|
|
"mean_token_accuracy": 0.1260973446071148,
|
|
"num_tokens": 78208.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.082484722137451,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.6216,
|
|
"mean_token_accuracy": 0.12194913551211357,
|
|
"num_tokens": 88871.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.0991229057312015,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.6677,
|
|
"mean_token_accuracy": 0.1267041429877281,
|
|
"num_tokens": 100118.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.066532564163208,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.6693,
|
|
"mean_token_accuracy": 0.12610288485884666,
|
|
"num_tokens": 113120.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.053462934494019,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.641,
|
|
"mean_token_accuracy": 0.12082915157079696,
|
|
"num_tokens": 123220.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.089922380447388,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.7837,
|
|
"mean_token_accuracy": 0.12217121347784995,
|
|
"num_tokens": 134479.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.060901641845703,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.6211,
|
|
"mean_token_accuracy": 0.12342652827501296,
|
|
"num_tokens": 145189.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.042802572250366,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.5941,
|
|
"mean_token_accuracy": 0.12326686978340148,
|
|
"num_tokens": 157354.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.083833312988281,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.7564,
|
|
"mean_token_accuracy": 0.11759938597679138,
|
|
"num_tokens": 168821.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.0386810302734375,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.5759,
|
|
"mean_token_accuracy": 0.12334224358201026,
|
|
"num_tokens": 180554.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.066742753982544,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.6724,
|
|
"mean_token_accuracy": 0.12350711151957512,
|
|
"num_tokens": 191229.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 6.984064149856567,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.5755,
|
|
"mean_token_accuracy": 0.12566502168774604,
|
|
"num_tokens": 201903.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.095264863967896,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.754,
|
|
"mean_token_accuracy": 0.1199577122926712,
|
|
"num_tokens": 214255.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.072626066207886,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.6307,
|
|
"mean_token_accuracy": 0.12475940734148025,
|
|
"num_tokens": 225423.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.049364042282105,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.7278,
|
|
"mean_token_accuracy": 0.12485888302326202,
|
|
"num_tokens": 237801.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 6.984118175506592,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.5268,
|
|
"mean_token_accuracy": 0.134263726323843,
|
|
"num_tokens": 249195.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 6.968926286697387,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.5154,
|
|
"mean_token_accuracy": 0.12820799723267556,
|
|
"num_tokens": 260129.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.053589344024658,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.7412,
|
|
"mean_token_accuracy": 0.11961034312844276,
|
|
"num_tokens": 271220.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.0658690452575685,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.5956,
|
|
"mean_token_accuracy": 0.12382429540157318,
|
|
"num_tokens": 281560.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 6.973921585083008,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.5231,
|
|
"mean_token_accuracy": 0.13096552416682244,
|
|
"num_tokens": 292518.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 6.9215654850006105,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.7008,
|
|
"mean_token_accuracy": 0.11911230981349945,
|
|
"num_tokens": 304865.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.06756010055542,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.5343,
|
|
"mean_token_accuracy": 0.12699158787727355,
|
|
"num_tokens": 315951.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.026167297363282,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.5997,
|
|
"mean_token_accuracy": 0.13356166407465936,
|
|
"num_tokens": 327094.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 6.972362804412842,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.6052,
|
|
"mean_token_accuracy": 0.12697447910904885,
|
|
"num_tokens": 337801.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 6.996054553985596,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.5938,
|
|
"mean_token_accuracy": 0.13034800440073013,
|
|
"num_tokens": 349039.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.017750644683838,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.6152,
|
|
"mean_token_accuracy": 0.12605011090636253,
|
|
"num_tokens": 360639.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.074881505966187,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.6717,
|
|
"mean_token_accuracy": 0.13049019277095794,
|
|
"num_tokens": 371005.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 6.990810108184815,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.5524,
|
|
"mean_token_accuracy": 0.12842826694250106,
|
|
"num_tokens": 383340.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 6.938203954696656,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.5105,
|
|
"mean_token_accuracy": 0.13346678763628006,
|
|
"num_tokens": 394682.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 6.959616565704346,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.6125,
|
|
"mean_token_accuracy": 0.12339805886149406,
|
|
"num_tokens": 405713.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 6.990169191360474,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.566,
|
|
"mean_token_accuracy": 0.12917021587491034,
|
|
"num_tokens": 416208.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 6.864281415939331,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.4592,
|
|
"mean_token_accuracy": 0.12425167560577392,
|
|
"num_tokens": 427595.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.1006927490234375,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.6387,
|
|
"mean_token_accuracy": 0.13051133900880812,
|
|
"num_tokens": 438273.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 6.979038572311401,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.5543,
|
|
"mean_token_accuracy": 0.12668330743908882,
|
|
"num_tokens": 448948.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 6.977887344360352,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.6267,
|
|
"mean_token_accuracy": 0.1252782464027405,
|
|
"num_tokens": 460838.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.0447746276855465,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.6526,
|
|
"mean_token_accuracy": 0.12487014755606651,
|
|
"num_tokens": 473555.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 6.9728395462036135,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.6119,
|
|
"mean_token_accuracy": 0.12403129488229751,
|
|
"num_tokens": 485221.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 6.938175296783447,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.4961,
|
|
"mean_token_accuracy": 0.125931017100811,
|
|
"num_tokens": 496552.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 6.918573665618896,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.5856,
|
|
"mean_token_accuracy": 0.12433982789516448,
|
|
"num_tokens": 509760.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 6.984656906127929,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.43,
|
|
"mean_token_accuracy": 0.1363843210041523,
|
|
"num_tokens": 521175.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 6.853548002243042,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.5608,
|
|
"mean_token_accuracy": 0.12902215570211412,
|
|
"num_tokens": 532597.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.019792795181274,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.6362,
|
|
"mean_token_accuracy": 0.12607059627771378,
|
|
"num_tokens": 545661.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 6.9751549243927,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.5991,
|
|
"mean_token_accuracy": 0.12282323241233825,
|
|
"num_tokens": 557336.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 6.929644775390625,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.4776,
|
|
"mean_token_accuracy": 0.13525281250476837,
|
|
"num_tokens": 567398.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 6.935058498382569,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.583,
|
|
"mean_token_accuracy": 0.12586332187056543,
|
|
"num_tokens": 579543.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 6.859152507781983,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.4077,
|
|
"mean_token_accuracy": 0.13932234793901443,
|
|
"num_tokens": 589218.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 6.889403057098389,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.4839,
|
|
"mean_token_accuracy": 0.13130961805582048,
|
|
"num_tokens": 600327.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 6.879350185394287,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.4478,
|
|
"mean_token_accuracy": 0.13568017333745958,
|
|
"num_tokens": 610923.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.02859468460083,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.5314,
|
|
"mean_token_accuracy": 0.12415590286254882,
|
|
"num_tokens": 621138.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 6.8203459739685055,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.477,
|
|
"mean_token_accuracy": 0.13310884237289428,
|
|
"num_tokens": 631440.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.021256637573242,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.5409,
|
|
"mean_token_accuracy": 0.13431001752614974,
|
|
"num_tokens": 642328.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 6.82467360496521,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.4725,
|
|
"mean_token_accuracy": 0.13090698570013046,
|
|
"num_tokens": 652875.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 6.948597621917725,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.4235,
|
|
"mean_token_accuracy": 0.13427285477519035,
|
|
"num_tokens": 663726.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 6.989661026000976,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.6491,
|
|
"mean_token_accuracy": 0.1237833745777607,
|
|
"num_tokens": 675638.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 6.91578369140625,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.4977,
|
|
"mean_token_accuracy": 0.1314319372177124,
|
|
"num_tokens": 687239.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 6.952743911743164,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.5642,
|
|
"mean_token_accuracy": 0.12674924060702325,
|
|
"num_tokens": 698905.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 6.899339485168457,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.5502,
|
|
"mean_token_accuracy": 0.12955581918358802,
|
|
"num_tokens": 710602.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 6.887494802474976,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.5107,
|
|
"mean_token_accuracy": 0.12705972641706467,
|
|
"num_tokens": 721326.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 6.9584733009338375,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.509,
|
|
"mean_token_accuracy": 0.1291506789624691,
|
|
"num_tokens": 732972.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 6.839654016494751,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.4595,
|
|
"mean_token_accuracy": 0.13303572162985802,
|
|
"num_tokens": 743624.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 6.992349910736084,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.5671,
|
|
"mean_token_accuracy": 0.1312260664999485,
|
|
"num_tokens": 756384.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 6.9219557762146,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.551,
|
|
"mean_token_accuracy": 0.13202789351344107,
|
|
"num_tokens": 767256.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 6.939928436279297,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.4653,
|
|
"mean_token_accuracy": 0.13179740980267524,
|
|
"num_tokens": 779059.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 6.9600786685943605,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.5818,
|
|
"mean_token_accuracy": 0.13102152496576308,
|
|
"num_tokens": 790291.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 6.866528797149658,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.4825,
|
|
"mean_token_accuracy": 0.13548264726996423,
|
|
"num_tokens": 801450.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 6.830200290679931,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.4039,
|
|
"mean_token_accuracy": 0.1368156909942627,
|
|
"num_tokens": 812428.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.793509531021118,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.4954,
|
|
"mean_token_accuracy": 0.1288583382964134,
|
|
"num_tokens": 823644.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.0165496349334715,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.5086,
|
|
"mean_token_accuracy": 0.1327413909137249,
|
|
"num_tokens": 834064.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 6.884370946884156,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.4397,
|
|
"mean_token_accuracy": 0.1367235191166401,
|
|
"num_tokens": 845207.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.871044635772705,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.5098,
|
|
"mean_token_accuracy": 0.12969856932759286,
|
|
"num_tokens": 855564.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.920437526702881,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.5511,
|
|
"mean_token_accuracy": 0.1326618030667305,
|
|
"num_tokens": 866959.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.824339103698731,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.4691,
|
|
"mean_token_accuracy": 0.1337393194437027,
|
|
"num_tokens": 877319.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.980741786956787,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.5581,
|
|
"mean_token_accuracy": 0.12818842977285386,
|
|
"num_tokens": 888037.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.727102470397949,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.3189,
|
|
"mean_token_accuracy": 0.13667411357164383,
|
|
"num_tokens": 898775.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.897746992111206,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.4451,
|
|
"mean_token_accuracy": 0.12954012975096701,
|
|
"num_tokens": 909513.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.793622636795044,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.5151,
|
|
"mean_token_accuracy": 0.12809686958789826,
|
|
"num_tokens": 920730.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.897607803344727,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.4482,
|
|
"mean_token_accuracy": 0.1362723909318447,
|
|
"num_tokens": 931457.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.811655569076538,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.4546,
|
|
"mean_token_accuracy": 0.13120315298438073,
|
|
"num_tokens": 942572.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.869682502746582,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.444,
|
|
"mean_token_accuracy": 0.1280343808233738,
|
|
"num_tokens": 953132.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.843258285522461,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.4848,
|
|
"mean_token_accuracy": 0.13226923644542693,
|
|
"num_tokens": 963551.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.876287460327148,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.4466,
|
|
"mean_token_accuracy": 0.13461819738149644,
|
|
"num_tokens": 975297.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.867691087722778,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.4438,
|
|
"mean_token_accuracy": 0.13997153118252753,
|
|
"num_tokens": 986910.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.878692197799682,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.5809,
|
|
"mean_token_accuracy": 0.13594842106103897,
|
|
"num_tokens": 997058.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.857804632186889,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.4646,
|
|
"mean_token_accuracy": 0.12713657170534134,
|
|
"num_tokens": 1008261.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.8630646705627445,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.381,
|
|
"mean_token_accuracy": 0.13709277138113976,
|
|
"num_tokens": 1019193.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.795991706848144,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.5169,
|
|
"mean_token_accuracy": 0.13212290033698082,
|
|
"num_tokens": 1031227.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.793545341491699,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.3292,
|
|
"mean_token_accuracy": 0.13823329731822015,
|
|
"num_tokens": 1042963.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.7479089260101315,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.406,
|
|
"mean_token_accuracy": 0.12962326854467393,
|
|
"num_tokens": 1053709.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.826336431503296,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.4424,
|
|
"mean_token_accuracy": 0.14208453595638276,
|
|
"num_tokens": 1064527.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.812590742111206,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.4849,
|
|
"mean_token_accuracy": 0.13389927595853807,
|
|
"num_tokens": 1075127.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.817338991165161,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.4905,
|
|
"mean_token_accuracy": 0.13158686384558677,
|
|
"num_tokens": 1086581.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.785533761978149,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.4204,
|
|
"mean_token_accuracy": 0.13986249789595603,
|
|
"num_tokens": 1097818.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.808701515197754,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.4129,
|
|
"mean_token_accuracy": 0.1334671624004841,
|
|
"num_tokens": 1108043.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.749666166305542,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.4721,
|
|
"mean_token_accuracy": 0.13080318421125411,
|
|
"num_tokens": 1119875.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.736193084716797,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.3796,
|
|
"mean_token_accuracy": 0.13133062347769736,
|
|
"num_tokens": 1132406.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.015969133377075,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.5432,
|
|
"mean_token_accuracy": 0.1301574267446995,
|
|
"num_tokens": 1143828.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.6993324756622314,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.3685,
|
|
"mean_token_accuracy": 0.13389391377568244,
|
|
"num_tokens": 1155482.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.733237409591675,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.3107,
|
|
"mean_token_accuracy": 0.14635325446724892,
|
|
"num_tokens": 1166847.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.804312419891358,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.398,
|
|
"mean_token_accuracy": 0.13656767308712006,
|
|
"num_tokens": 1177419.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.850254726409912,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.564,
|
|
"mean_token_accuracy": 0.12607400715351105,
|
|
"num_tokens": 1188975.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.744582986831665,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.3989,
|
|
"mean_token_accuracy": 0.13207888677716256,
|
|
"num_tokens": 1200252.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.745368576049804,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.4506,
|
|
"mean_token_accuracy": 0.14177516996860504,
|
|
"num_tokens": 1211864.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.779013586044312,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.3265,
|
|
"mean_token_accuracy": 0.13602656349539757,
|
|
"num_tokens": 1222429.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.728393316268921,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.4471,
|
|
"mean_token_accuracy": 0.13411575108766555,
|
|
"num_tokens": 1233782.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.827554702758789,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.3854,
|
|
"mean_token_accuracy": 0.13783360570669173,
|
|
"num_tokens": 1244309.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.647427225112915,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.3455,
|
|
"mean_token_accuracy": 0.13646090924739837,
|
|
"num_tokens": 1256403.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.858051395416259,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4363,
|
|
"mean_token_accuracy": 0.13374576345086098,
|
|
"num_tokens": 1268758.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.639367294311524,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.3299,
|
|
"mean_token_accuracy": 0.13808067962527276,
|
|
"num_tokens": 1279355.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.747273302078247,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.3322,
|
|
"mean_token_accuracy": 0.13361302465200425,
|
|
"num_tokens": 1290619.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.703400897979736,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.2981,
|
|
"mean_token_accuracy": 0.13826739564538001,
|
|
"num_tokens": 1301851.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.771224403381348,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.4475,
|
|
"mean_token_accuracy": 0.13633046373724939,
|
|
"num_tokens": 1313260.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.688079690933227,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.4194,
|
|
"mean_token_accuracy": 0.13458380997180938,
|
|
"num_tokens": 1324189.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.8580217361450195,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.4634,
|
|
"mean_token_accuracy": 0.13595897927880288,
|
|
"num_tokens": 1335765.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.723800086975098,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.3969,
|
|
"mean_token_accuracy": 0.13711549192667008,
|
|
"num_tokens": 1347653.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.741200160980225,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3948,
|
|
"mean_token_accuracy": 0.13611036017537118,
|
|
"num_tokens": 1360201.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.662504434585571,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.2728,
|
|
"mean_token_accuracy": 0.13825351148843765,
|
|
"num_tokens": 1370974.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.655742025375366,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.3426,
|
|
"mean_token_accuracy": 0.1370379976928234,
|
|
"num_tokens": 1383855.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.6989470481872555,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.3126,
|
|
"mean_token_accuracy": 0.14310086369514466,
|
|
"num_tokens": 1395041.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.741576099395752,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4252,
|
|
"mean_token_accuracy": 0.13634751066565515,
|
|
"num_tokens": 1406102.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.622606801986694,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.236,
|
|
"mean_token_accuracy": 0.14010374397039413,
|
|
"num_tokens": 1417713.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.654081964492798,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.37,
|
|
"mean_token_accuracy": 0.14272874891757964,
|
|
"num_tokens": 1428526.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.707960319519043,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3237,
|
|
"mean_token_accuracy": 0.1384015902876854,
|
|
"num_tokens": 1438849.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.680639600753784,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4087,
|
|
"mean_token_accuracy": 0.13631037175655364,
|
|
"num_tokens": 1450776.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.651285934448242,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.2713,
|
|
"mean_token_accuracy": 0.13713359385728835,
|
|
"num_tokens": 1462552.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.584797048568726,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.2921,
|
|
"mean_token_accuracy": 0.13798285499215127,
|
|
"num_tokens": 1473867.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.530507802963257,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.2928,
|
|
"mean_token_accuracy": 0.1366693250834942,
|
|
"num_tokens": 1485233.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.7059835433959964,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.3492,
|
|
"mean_token_accuracy": 0.14286357387900353,
|
|
"num_tokens": 1496919.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.66720404624939,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.3022,
|
|
"mean_token_accuracy": 0.13693906292319297,
|
|
"num_tokens": 1508695.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.647221851348877,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.3508,
|
|
"mean_token_accuracy": 0.14086920469999314,
|
|
"num_tokens": 1519629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.619228792190552,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.2518,
|
|
"mean_token_accuracy": 0.13821905553340913,
|
|
"num_tokens": 1530144.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.519640779495239,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.2681,
|
|
"mean_token_accuracy": 0.1386340342462063,
|
|
"num_tokens": 1541013.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.7023632526397705,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3291,
|
|
"mean_token_accuracy": 0.141255185008049,
|
|
"num_tokens": 1553344.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.668040466308594,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.394,
|
|
"mean_token_accuracy": 0.13976067006587983,
|
|
"num_tokens": 1565298.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.601643800735474,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3218,
|
|
"mean_token_accuracy": 0.14546340629458426,
|
|
"num_tokens": 1577645.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.602117729187012,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.2693,
|
|
"mean_token_accuracy": 0.1462263584136963,
|
|
"num_tokens": 1588387.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.6649316310882565,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3715,
|
|
"mean_token_accuracy": 0.1455303281545639,
|
|
"num_tokens": 1599837.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.583027362823486,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.3233,
|
|
"mean_token_accuracy": 0.134693793207407,
|
|
"num_tokens": 1610401.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.567713499069214,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.2618,
|
|
"mean_token_accuracy": 0.14031144231557846,
|
|
"num_tokens": 1622688.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.690587854385376,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.13657473996281624,
|
|
"num_tokens": 1633594.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.569250011444092,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.2579,
|
|
"mean_token_accuracy": 0.13581339344382287,
|
|
"num_tokens": 1645861.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.466774034500122,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.1453,
|
|
"mean_token_accuracy": 0.1467580884695053,
|
|
"num_tokens": 1656903.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.587268447875976,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.2564,
|
|
"mean_token_accuracy": 0.1428845003247261,
|
|
"num_tokens": 1668130.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.389280652999878,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.1654,
|
|
"mean_token_accuracy": 0.14585966542363166,
|
|
"num_tokens": 1678228.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.6626745700836185,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3117,
|
|
"mean_token_accuracy": 0.14107734560966492,
|
|
"num_tokens": 1690339.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.601497745513916,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.323,
|
|
"mean_token_accuracy": 0.13589314743876457,
|
|
"num_tokens": 1701067.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.6254418849945065,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.269,
|
|
"mean_token_accuracy": 0.1457679569721222,
|
|
"num_tokens": 1712835.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.527891016006469,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.2239,
|
|
"mean_token_accuracy": 0.14019766300916672,
|
|
"num_tokens": 1723432.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.61613621711731,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.2652,
|
|
"mean_token_accuracy": 0.14233424216508866,
|
|
"num_tokens": 1734134.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.543462657928467,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.2148,
|
|
"mean_token_accuracy": 0.14095726907253264,
|
|
"num_tokens": 1744723.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.583271503448486,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3179,
|
|
"mean_token_accuracy": 0.14557504281401634,
|
|
"num_tokens": 1755269.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.61468505859375,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3479,
|
|
"mean_token_accuracy": 0.13776199892163277,
|
|
"num_tokens": 1766344.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.578250408172607,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3128,
|
|
"mean_token_accuracy": 0.13852658942341806,
|
|
"num_tokens": 1779052.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.43643536567688,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.164,
|
|
"mean_token_accuracy": 0.1452375516295433,
|
|
"num_tokens": 1789754.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.428613233566284,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.0883,
|
|
"mean_token_accuracy": 0.14618916660547257,
|
|
"num_tokens": 1801161.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.502646541595459,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2021,
|
|
"mean_token_accuracy": 0.13893655687570572,
|
|
"num_tokens": 1810892.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.446267414093017,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.268,
|
|
"mean_token_accuracy": 0.14093232452869414,
|
|
"num_tokens": 1821574.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.5819238185882565,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.298,
|
|
"mean_token_accuracy": 0.1420733690261841,
|
|
"num_tokens": 1833342.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.522675514221191,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.1798,
|
|
"mean_token_accuracy": 0.14461357444524764,
|
|
"num_tokens": 1843768.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.551895189285278,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2419,
|
|
"mean_token_accuracy": 0.14213603138923644,
|
|
"num_tokens": 1855569.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.540101480484009,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.2433,
|
|
"mean_token_accuracy": 0.13941342383623123,
|
|
"num_tokens": 1866684.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.516862154006958,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.1593,
|
|
"mean_token_accuracy": 0.14692240357398986,
|
|
"num_tokens": 1878700.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.46841516494751,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.2014,
|
|
"mean_token_accuracy": 0.14441732317209244,
|
|
"num_tokens": 1889970.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.395776987075806,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.0948,
|
|
"mean_token_accuracy": 0.14418264329433442,
|
|
"num_tokens": 1902008.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.538009738922119,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3405,
|
|
"mean_token_accuracy": 0.1436830170452595,
|
|
"num_tokens": 1913253.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.455433464050293,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.1778,
|
|
"mean_token_accuracy": 0.14896109327673912,
|
|
"num_tokens": 1923408.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.4827131748199465,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.2436,
|
|
"mean_token_accuracy": 0.14401304572820664,
|
|
"num_tokens": 1935315.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.42853045463562,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.0795,
|
|
"mean_token_accuracy": 0.1439765587449074,
|
|
"num_tokens": 1946524.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.469723701477051,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2223,
|
|
"mean_token_accuracy": 0.1448940485715866,
|
|
"num_tokens": 1957561.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.445930862426758,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.1569,
|
|
"mean_token_accuracy": 0.1493787460029125,
|
|
"num_tokens": 1969089.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.437417364120483,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.1798,
|
|
"mean_token_accuracy": 0.1484062746167183,
|
|
"num_tokens": 1980341.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.4072212219238285,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.1442,
|
|
"mean_token_accuracy": 0.1515571042895317,
|
|
"num_tokens": 1991220.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.483857440948486,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2265,
|
|
"mean_token_accuracy": 0.14315667673945426,
|
|
"num_tokens": 2002650.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.475426959991455,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.1463,
|
|
"mean_token_accuracy": 0.14812344163656235,
|
|
"num_tokens": 2012884.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.416254472732544,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2191,
|
|
"mean_token_accuracy": 0.14170301109552383,
|
|
"num_tokens": 2023543.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.5840905666351315,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2738,
|
|
"mean_token_accuracy": 0.1421659953892231,
|
|
"num_tokens": 2034789.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.4489476680755615,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.1754,
|
|
"mean_token_accuracy": 0.1458805352449417,
|
|
"num_tokens": 2046719.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.2773548603057865,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.1288,
|
|
"mean_token_accuracy": 0.14724763333797455,
|
|
"num_tokens": 2057016.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.437155485153198,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.1202,
|
|
"mean_token_accuracy": 0.14350012987852095,
|
|
"num_tokens": 2067898.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.385901186201307,
|
|
"epoch": 1.0,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.1454,
|
|
"mean_token_accuracy": 0.14857770336998832,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.408116817474365,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.0715,
|
|
"mean_token_accuracy": 0.13998154550790787,
|
|
"num_tokens": 2089228.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.37381329536438,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.0826,
|
|
"mean_token_accuracy": 0.15219226852059364,
|
|
"num_tokens": 2101098.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.405420303344727,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.0171,
|
|
"mean_token_accuracy": 0.15995113998651506,
|
|
"num_tokens": 2113081.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.384943532943725,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0034,
|
|
"mean_token_accuracy": 0.15575264096260072,
|
|
"num_tokens": 2123356.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.478774785995483,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.0969,
|
|
"mean_token_accuracy": 0.15050555020570755,
|
|
"num_tokens": 2134388.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.215549659729004,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 5.9639,
|
|
"mean_token_accuracy": 0.1580267071723938,
|
|
"num_tokens": 2145209.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.422420835494995,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.0031,
|
|
"mean_token_accuracy": 0.1567741423845291,
|
|
"num_tokens": 2157146.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.2772218704223635,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.0338,
|
|
"mean_token_accuracy": 0.14722552746534348,
|
|
"num_tokens": 2168721.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.415115833282471,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.0368,
|
|
"mean_token_accuracy": 0.1544208213686943,
|
|
"num_tokens": 2179213.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.353874444961548,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.0418,
|
|
"mean_token_accuracy": 0.1518670752644539,
|
|
"num_tokens": 2191224.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.365312385559082,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.1152,
|
|
"mean_token_accuracy": 0.15054795444011687,
|
|
"num_tokens": 2203665.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.40236873626709,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0167,
|
|
"mean_token_accuracy": 0.15078850239515304,
|
|
"num_tokens": 2215136.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.344824981689453,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.0651,
|
|
"mean_token_accuracy": 0.1549616940319538,
|
|
"num_tokens": 2226375.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.314645862579345,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 5.9136,
|
|
"mean_token_accuracy": 0.15960747003555298,
|
|
"num_tokens": 2236097.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.427794027328491,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.1535,
|
|
"mean_token_accuracy": 0.1500318855047226,
|
|
"num_tokens": 2247964.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.234298613938418,
|
|
"eval_loss": 6.301173686981201,
|
|
"eval_mean_token_accuracy": 0.14848606810820372,
|
|
"eval_num_tokens": 2247964.0,
|
|
"eval_runtime": 1.9522,
|
|
"eval_samples_per_second": 1800.502,
|
|
"eval_steps_per_second": 225.383,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.294188117980957,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.0967,
|
|
"mean_token_accuracy": 0.15498952567577362,
|
|
"num_tokens": 2260467.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.388325452804565,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 6.002,
|
|
"mean_token_accuracy": 0.1555919662117958,
|
|
"num_tokens": 2272176.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.356983327865601,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 6.0452,
|
|
"mean_token_accuracy": 0.15810664147138595,
|
|
"num_tokens": 2283491.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.31780686378479,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 5.9421,
|
|
"mean_token_accuracy": 0.15196095556020736,
|
|
"num_tokens": 2295131.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.276276922225952,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 5.9784,
|
|
"mean_token_accuracy": 0.15291586369276047,
|
|
"num_tokens": 2305832.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.208898210525513,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 5.976,
|
|
"mean_token_accuracy": 0.15655720829963685,
|
|
"num_tokens": 2315874.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.530113983154297,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.1517,
|
|
"mean_token_accuracy": 0.1527762308716774,
|
|
"num_tokens": 2327580.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.300721311569214,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 6.0105,
|
|
"mean_token_accuracy": 0.15481361150741577,
|
|
"num_tokens": 2340426.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.324107599258423,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 6.1329,
|
|
"mean_token_accuracy": 0.14513010084629058,
|
|
"num_tokens": 2353228.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.315619230270386,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 5.9384,
|
|
"mean_token_accuracy": 0.15674824565649031,
|
|
"num_tokens": 2364238.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.288443326950073,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.0783,
|
|
"mean_token_accuracy": 0.1514833614230156,
|
|
"num_tokens": 2375847.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.27244725227356,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 5.9658,
|
|
"mean_token_accuracy": 0.15787345468997954,
|
|
"num_tokens": 2386489.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.327007818222046,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 6.0072,
|
|
"mean_token_accuracy": 0.1553010106086731,
|
|
"num_tokens": 2398839.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.326588678359985,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 5.976,
|
|
"mean_token_accuracy": 0.1555561050772667,
|
|
"num_tokens": 2410016.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.278480052947998,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 6.0055,
|
|
"mean_token_accuracy": 0.15416339486837388,
|
|
"num_tokens": 2421202.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.35570878982544,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.0975,
|
|
"mean_token_accuracy": 0.1458723247051239,
|
|
"num_tokens": 2431943.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.250971841812134,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 5.9163,
|
|
"mean_token_accuracy": 0.15354986488819122,
|
|
"num_tokens": 2443454.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.320730972290039,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.1045,
|
|
"mean_token_accuracy": 0.14482270926237106,
|
|
"num_tokens": 2455909.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.229953575134277,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 5.9645,
|
|
"mean_token_accuracy": 0.159816737473011,
|
|
"num_tokens": 2467486.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.329736661911011,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 5.9201,
|
|
"mean_token_accuracy": 0.1614977702498436,
|
|
"num_tokens": 2477560.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.153037214279175,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 5.9692,
|
|
"mean_token_accuracy": 0.16537124514579774,
|
|
"num_tokens": 2487222.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.275479412078857,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 5.9952,
|
|
"mean_token_accuracy": 0.15594137758016585,
|
|
"num_tokens": 2497474.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.24587230682373,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.0581,
|
|
"mean_token_accuracy": 0.15599578768014907,
|
|
"num_tokens": 2508185.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.3197838306427006,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 6.002,
|
|
"mean_token_accuracy": 0.15079409927129744,
|
|
"num_tokens": 2518876.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.22892746925354,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 5.9758,
|
|
"mean_token_accuracy": 0.15842700153589248,
|
|
"num_tokens": 2530554.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.213609313964843,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 5.8791,
|
|
"mean_token_accuracy": 0.15403347164392472,
|
|
"num_tokens": 2541331.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.302558851242066,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 6.0825,
|
|
"mean_token_accuracy": 0.14472228586673735,
|
|
"num_tokens": 2554225.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.233757209777832,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 5.98,
|
|
"mean_token_accuracy": 0.1540392369031906,
|
|
"num_tokens": 2565050.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.367772436141967,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 6.0568,
|
|
"mean_token_accuracy": 0.1523862287402153,
|
|
"num_tokens": 2576926.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.316640138626099,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.127,
|
|
"mean_token_accuracy": 0.15167189091444017,
|
|
"num_tokens": 2588775.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.1811521530151365,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 5.9747,
|
|
"mean_token_accuracy": 0.15249996781349182,
|
|
"num_tokens": 2600534.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.1374688148498535,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 5.8141,
|
|
"mean_token_accuracy": 0.16029319763183594,
|
|
"num_tokens": 2612350.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.207683801651001,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 5.992,
|
|
"mean_token_accuracy": 0.15426600873470306,
|
|
"num_tokens": 2623871.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.20775728225708,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 5.9307,
|
|
"mean_token_accuracy": 0.15610288083553314,
|
|
"num_tokens": 2636371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.251627016067505,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 5.9776,
|
|
"mean_token_accuracy": 0.15636180639266967,
|
|
"num_tokens": 2646794.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.213901853561401,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 5.9745,
|
|
"mean_token_accuracy": 0.16118483394384384,
|
|
"num_tokens": 2657105.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.1841833114624025,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 5.9136,
|
|
"mean_token_accuracy": 0.15974865704774857,
|
|
"num_tokens": 2668297.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.15105357170105,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 5.778,
|
|
"mean_token_accuracy": 0.1612982451915741,
|
|
"num_tokens": 2678834.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.236361694335938,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 6.0237,
|
|
"mean_token_accuracy": 0.15312567353248596,
|
|
"num_tokens": 2690392.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.13387770652771,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 5.8177,
|
|
"mean_token_accuracy": 0.16519954651594163,
|
|
"num_tokens": 2701319.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.231759071350098,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 5.9873,
|
|
"mean_token_accuracy": 0.1522187188267708,
|
|
"num_tokens": 2713018.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.267031526565551,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 6.0829,
|
|
"mean_token_accuracy": 0.150100913643837,
|
|
"num_tokens": 2724995.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.227834606170655,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 5.9408,
|
|
"mean_token_accuracy": 0.15533591508865358,
|
|
"num_tokens": 2736733.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.10360312461853,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 5.9107,
|
|
"mean_token_accuracy": 0.16030677407979965,
|
|
"num_tokens": 2748436.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.18991494178772,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 5.9373,
|
|
"mean_token_accuracy": 0.15890274941921234,
|
|
"num_tokens": 2758924.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.213899087905884,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 5.9927,
|
|
"mean_token_accuracy": 0.15283182114362717,
|
|
"num_tokens": 2769773.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.344644165039062,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 6.1141,
|
|
"mean_token_accuracy": 0.15794118642807006,
|
|
"num_tokens": 2782414.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.064876842498779,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 5.8871,
|
|
"mean_token_accuracy": 0.15881086140871048,
|
|
"num_tokens": 2794325.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.073903369903564,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 5.7434,
|
|
"mean_token_accuracy": 0.1659662812948227,
|
|
"num_tokens": 2805391.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.082292318344116,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 5.8482,
|
|
"mean_token_accuracy": 0.16185838878154754,
|
|
"num_tokens": 2816633.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.214509344100952,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 5.9803,
|
|
"mean_token_accuracy": 0.16008955836296082,
|
|
"num_tokens": 2828395.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.2064978122711185,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 5.9607,
|
|
"mean_token_accuracy": 0.16311137825250627,
|
|
"num_tokens": 2839592.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.160184526443482,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 5.9266,
|
|
"mean_token_accuracy": 0.16404825299978257,
|
|
"num_tokens": 2849557.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.26066746711731,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 5.9809,
|
|
"mean_token_accuracy": 0.15878378599882126,
|
|
"num_tokens": 2862030.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.110366535186768,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 5.9034,
|
|
"mean_token_accuracy": 0.16052400171756745,
|
|
"num_tokens": 2872379.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.228089904785156,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 5.9781,
|
|
"mean_token_accuracy": 0.16122061163187026,
|
|
"num_tokens": 2883170.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.183463716506958,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 5.9202,
|
|
"mean_token_accuracy": 0.15599651485681534,
|
|
"num_tokens": 2894013.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.222345685958862,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 6.0014,
|
|
"mean_token_accuracy": 0.16009113192558289,
|
|
"num_tokens": 2905073.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.250334596633911,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 5.964,
|
|
"mean_token_accuracy": 0.15829345732927322,
|
|
"num_tokens": 2916579.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.089809846878052,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 5.8644,
|
|
"mean_token_accuracy": 0.16013842970132827,
|
|
"num_tokens": 2928522.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.225626420974732,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 5.9464,
|
|
"mean_token_accuracy": 0.16239900290966033,
|
|
"num_tokens": 2938754.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.968344354629517,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 5.751,
|
|
"mean_token_accuracy": 0.16453044712543488,
|
|
"num_tokens": 2949696.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.210727691650391,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 5.954,
|
|
"mean_token_accuracy": 0.1518259599804878,
|
|
"num_tokens": 2961144.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.193397474288941,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 5.9316,
|
|
"mean_token_accuracy": 0.15983943790197372,
|
|
"num_tokens": 2972368.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.107234382629395,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 5.876,
|
|
"mean_token_accuracy": 0.15900884866714476,
|
|
"num_tokens": 2984853.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.168393850326538,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 5.9241,
|
|
"mean_token_accuracy": 0.16157913506031035,
|
|
"num_tokens": 2995095.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.118945741653443,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 5.9752,
|
|
"mean_token_accuracy": 0.15459085404872894,
|
|
"num_tokens": 3005926.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.262211513519287,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 5.9399,
|
|
"mean_token_accuracy": 0.1581527978181839,
|
|
"num_tokens": 3017647.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.067259502410889,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 5.8572,
|
|
"mean_token_accuracy": 0.16498840600252151,
|
|
"num_tokens": 3028904.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.157345771789551,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 5.8862,
|
|
"mean_token_accuracy": 0.1650509089231491,
|
|
"num_tokens": 3039347.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.094135189056397,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 5.8615,
|
|
"mean_token_accuracy": 0.15922504812479019,
|
|
"num_tokens": 3049951.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.206204128265381,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 5.918,
|
|
"mean_token_accuracy": 0.15983717739582062,
|
|
"num_tokens": 3061894.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.080220317840576,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.16588944792747498,
|
|
"num_tokens": 3072393.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.125793933868408,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 5.9319,
|
|
"mean_token_accuracy": 0.15972879379987717,
|
|
"num_tokens": 3083541.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.093274116516113,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 5.7468,
|
|
"mean_token_accuracy": 0.1670151799917221,
|
|
"num_tokens": 3093851.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.995811128616333,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 5.8325,
|
|
"mean_token_accuracy": 0.16010409593582153,
|
|
"num_tokens": 3105703.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.227929496765137,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 5.9681,
|
|
"mean_token_accuracy": 0.15533651411533356,
|
|
"num_tokens": 3117063.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.109680032730102,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 5.9294,
|
|
"mean_token_accuracy": 0.1571633294224739,
|
|
"num_tokens": 3129020.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.144689559936523,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 5.9049,
|
|
"mean_token_accuracy": 0.16143753677606582,
|
|
"num_tokens": 3140627.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.128430318832398,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 5.9354,
|
|
"mean_token_accuracy": 0.16027805656194688,
|
|
"num_tokens": 3152294.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.164208173751831,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 5.9254,
|
|
"mean_token_accuracy": 0.15725037455558777,
|
|
"num_tokens": 3163770.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.165513277053833,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 5.8802,
|
|
"mean_token_accuracy": 0.1594867393374443,
|
|
"num_tokens": 3175804.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.190362501144409,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 5.9108,
|
|
"mean_token_accuracy": 0.16410526186227797,
|
|
"num_tokens": 3187233.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.075640535354614,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 5.8704,
|
|
"mean_token_accuracy": 0.1694168344140053,
|
|
"num_tokens": 3198884.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.037021780014038,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.8358,
|
|
"mean_token_accuracy": 0.161753848195076,
|
|
"num_tokens": 3209836.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.091703653335571,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 5.8595,
|
|
"mean_token_accuracy": 0.16449046581983567,
|
|
"num_tokens": 3221295.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.033438491821289,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 5.8029,
|
|
"mean_token_accuracy": 0.17073612213134765,
|
|
"num_tokens": 3231832.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.205273103713989,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 5.9025,
|
|
"mean_token_accuracy": 0.1606149658560753,
|
|
"num_tokens": 3243483.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.045125961303711,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 5.9294,
|
|
"mean_token_accuracy": 0.1597157448530197,
|
|
"num_tokens": 3255785.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.08787431716919,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.8257,
|
|
"mean_token_accuracy": 0.15949517041444777,
|
|
"num_tokens": 3267035.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.262358951568603,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 5.9089,
|
|
"mean_token_accuracy": 0.16256033331155778,
|
|
"num_tokens": 3277848.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.9939404964447025,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.824,
|
|
"mean_token_accuracy": 0.16070163995027542,
|
|
"num_tokens": 3289864.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.163877534866333,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 5.9514,
|
|
"mean_token_accuracy": 0.16202145665884018,
|
|
"num_tokens": 3301441.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.107748174667359,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 5.9025,
|
|
"mean_token_accuracy": 0.1600061386823654,
|
|
"num_tokens": 3313216.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.98473424911499,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 5.7308,
|
|
"mean_token_accuracy": 0.1621977671980858,
|
|
"num_tokens": 3323760.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.014540481567383,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 5.8028,
|
|
"mean_token_accuracy": 0.16561368405818938,
|
|
"num_tokens": 3334785.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.08791298866272,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 5.8686,
|
|
"mean_token_accuracy": 0.16399995535612105,
|
|
"num_tokens": 3345741.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.082849216461182,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 5.8223,
|
|
"mean_token_accuracy": 0.1676563873887062,
|
|
"num_tokens": 3356118.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.002215099334717,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.7512,
|
|
"mean_token_accuracy": 0.1649516150355339,
|
|
"num_tokens": 3368107.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.016508913040161,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 5.7753,
|
|
"mean_token_accuracy": 0.16459955722093583,
|
|
"num_tokens": 3379367.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.044958305358887,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.8976,
|
|
"mean_token_accuracy": 0.1571328893303871,
|
|
"num_tokens": 3390597.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.103057956695556,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 5.8201,
|
|
"mean_token_accuracy": 0.16496762484312058,
|
|
"num_tokens": 3402618.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.907073974609375,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.7279,
|
|
"mean_token_accuracy": 0.1686336636543274,
|
|
"num_tokens": 3413162.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.995378494262695,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.8067,
|
|
"mean_token_accuracy": 0.17164268642663955,
|
|
"num_tokens": 3424331.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.136151695251465,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 5.9157,
|
|
"mean_token_accuracy": 0.15557393282651902,
|
|
"num_tokens": 3434739.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.177920913696289,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 5.9784,
|
|
"mean_token_accuracy": 0.1551076963543892,
|
|
"num_tokens": 3445984.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.897915315628052,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.7168,
|
|
"mean_token_accuracy": 0.17252864986658095,
|
|
"num_tokens": 3456955.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.183189487457275,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 5.8691,
|
|
"mean_token_accuracy": 0.1613949254155159,
|
|
"num_tokens": 3468895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.896417140960693,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 5.7837,
|
|
"mean_token_accuracy": 0.16745151132345198,
|
|
"num_tokens": 3480083.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.138626670837402,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.9133,
|
|
"mean_token_accuracy": 0.15685956478118895,
|
|
"num_tokens": 3492242.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.0059911727905275,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.7348,
|
|
"mean_token_accuracy": 0.1668782651424408,
|
|
"num_tokens": 3501926.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.957527542114258,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 5.821,
|
|
"mean_token_accuracy": 0.16384195387363434,
|
|
"num_tokens": 3511421.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.090160799026489,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 5.7543,
|
|
"mean_token_accuracy": 0.16819013953208922,
|
|
"num_tokens": 3522033.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.992568016052246,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 5.8076,
|
|
"mean_token_accuracy": 0.16708549559116365,
|
|
"num_tokens": 3533232.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.113001489639283,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 5.8111,
|
|
"mean_token_accuracy": 0.16661059260368347,
|
|
"num_tokens": 3544904.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.079259061813355,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.8893,
|
|
"mean_token_accuracy": 0.163326995074749,
|
|
"num_tokens": 3556545.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.98661470413208,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.7771,
|
|
"mean_token_accuracy": 0.16965614557266234,
|
|
"num_tokens": 3567361.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.050508260726929,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.8019,
|
|
"mean_token_accuracy": 0.16412458717823028,
|
|
"num_tokens": 3578943.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.025697183609009,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.7871,
|
|
"mean_token_accuracy": 0.16324581503868102,
|
|
"num_tokens": 3589868.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.011065483093262,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 5.7753,
|
|
"mean_token_accuracy": 0.1689581260085106,
|
|
"num_tokens": 3601245.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.058443593978882,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 5.8314,
|
|
"mean_token_accuracy": 0.16541973650455474,
|
|
"num_tokens": 3612168.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.9666486263275145,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.7506,
|
|
"mean_token_accuracy": 0.16870851963758468,
|
|
"num_tokens": 3624699.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.0644035816192625,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 5.9058,
|
|
"mean_token_accuracy": 0.1674930915236473,
|
|
"num_tokens": 3635554.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.947196912765503,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.7838,
|
|
"mean_token_accuracy": 0.16523039489984512,
|
|
"num_tokens": 3646776.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.982953596115112,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.7377,
|
|
"mean_token_accuracy": 0.16649552434682846,
|
|
"num_tokens": 3657976.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.126852893829346,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.8588,
|
|
"mean_token_accuracy": 0.16035668104887008,
|
|
"num_tokens": 3668242.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.095741319656372,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.16134843677282334,
|
|
"num_tokens": 3679200.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.9892982006072994,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 5.7836,
|
|
"mean_token_accuracy": 0.16862665712833405,
|
|
"num_tokens": 3690227.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.064673614501953,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.8128,
|
|
"mean_token_accuracy": 0.16635770797729493,
|
|
"num_tokens": 3701821.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.893378639221192,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.7514,
|
|
"mean_token_accuracy": 0.16452899128198623,
|
|
"num_tokens": 3712420.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.063785886764526,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 5.8019,
|
|
"mean_token_accuracy": 0.16422065794467927,
|
|
"num_tokens": 3722211.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.015103960037232,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.779,
|
|
"mean_token_accuracy": 0.16545072942972183,
|
|
"num_tokens": 3732822.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.965624094009399,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 5.8486,
|
|
"mean_token_accuracy": 0.1628899648785591,
|
|
"num_tokens": 3744262.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.021666765213013,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.7515,
|
|
"mean_token_accuracy": 0.16657728254795073,
|
|
"num_tokens": 3755713.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.944289827346802,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.7656,
|
|
"mean_token_accuracy": 0.16776938140392303,
|
|
"num_tokens": 3767179.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.9336107730865475,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.7571,
|
|
"mean_token_accuracy": 0.16716961711645126,
|
|
"num_tokens": 3777714.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.115212440490723,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.7571,
|
|
"mean_token_accuracy": 0.1713837891817093,
|
|
"num_tokens": 3788092.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.912018632888794,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.7216,
|
|
"mean_token_accuracy": 0.16672396957874297,
|
|
"num_tokens": 3798080.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.000129365921021,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 5.7782,
|
|
"mean_token_accuracy": 0.167687226831913,
|
|
"num_tokens": 3809547.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.987514019012451,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.7574,
|
|
"mean_token_accuracy": 0.16955365985631943,
|
|
"num_tokens": 3820681.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.008990621566772,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 5.7507,
|
|
"mean_token_accuracy": 0.17382699102163315,
|
|
"num_tokens": 3831598.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.951951789855957,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.8059,
|
|
"mean_token_accuracy": 0.16858023703098296,
|
|
"num_tokens": 3843437.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.947147274017334,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.6827,
|
|
"mean_token_accuracy": 0.16696840077638625,
|
|
"num_tokens": 3855387.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.02001166343689,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.8419,
|
|
"mean_token_accuracy": 0.1597014658153057,
|
|
"num_tokens": 3866999.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.952745866775513,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.7682,
|
|
"mean_token_accuracy": 0.1655362516641617,
|
|
"num_tokens": 3877721.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.038488483428955,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.8515,
|
|
"mean_token_accuracy": 0.16018995195627211,
|
|
"num_tokens": 3888831.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.978685188293457,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.7837,
|
|
"mean_token_accuracy": 0.17195567339658738,
|
|
"num_tokens": 3900389.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.918451309204102,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 5.7529,
|
|
"mean_token_accuracy": 0.1685625821352005,
|
|
"num_tokens": 3911641.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.8622355461120605,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.6311,
|
|
"mean_token_accuracy": 0.17217615246772766,
|
|
"num_tokens": 3922877.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.977212810516358,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.7108,
|
|
"mean_token_accuracy": 0.169350203871727,
|
|
"num_tokens": 3933573.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.88216462135315,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.657,
|
|
"mean_token_accuracy": 0.17444958537817,
|
|
"num_tokens": 3944339.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.104385709762573,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.7627,
|
|
"mean_token_accuracy": 0.17519646883010864,
|
|
"num_tokens": 3955459.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.839329481124878,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.672,
|
|
"mean_token_accuracy": 0.17154742330312728,
|
|
"num_tokens": 3966782.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.943143749237061,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.8012,
|
|
"mean_token_accuracy": 0.16626969575881959,
|
|
"num_tokens": 3978794.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.079607820510864,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 5.8028,
|
|
"mean_token_accuracy": 0.16924634277820588,
|
|
"num_tokens": 3990838.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.992142820358277,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 5.8216,
|
|
"mean_token_accuracy": 0.1701921135187149,
|
|
"num_tokens": 4000630.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.011171674728393,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.7016,
|
|
"mean_token_accuracy": 0.17147723585367203,
|
|
"num_tokens": 4011433.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.9299561977386475,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.8566,
|
|
"mean_token_accuracy": 0.16703943461179732,
|
|
"num_tokens": 4023671.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.923189973831176,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.654,
|
|
"mean_token_accuracy": 0.17370906174182893,
|
|
"num_tokens": 4034686.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.99422173500061,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 5.7706,
|
|
"mean_token_accuracy": 0.1754398062825203,
|
|
"num_tokens": 4046387.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.921460151672363,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.6795,
|
|
"mean_token_accuracy": 0.17212869375944137,
|
|
"num_tokens": 4056103.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.91083083152771,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.6532,
|
|
"mean_token_accuracy": 0.17965114414691924,
|
|
"num_tokens": 4066940.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.897168970108032,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 5.7687,
|
|
"mean_token_accuracy": 0.1766805499792099,
|
|
"num_tokens": 4077232.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.966326427459717,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.8383,
|
|
"mean_token_accuracy": 0.16350417137145995,
|
|
"num_tokens": 4089322.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.950818872451782,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.6806,
|
|
"mean_token_accuracy": 0.17309372872114182,
|
|
"num_tokens": 4099737.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.905142450332642,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.6828,
|
|
"mean_token_accuracy": 0.1649150624871254,
|
|
"num_tokens": 4110793.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.941768169403076,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.7169,
|
|
"mean_token_accuracy": 0.16933204084634781,
|
|
"num_tokens": 4121481.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.868819999694824,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.6838,
|
|
"mean_token_accuracy": 0.1762157514691353,
|
|
"num_tokens": 4132701.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.963793468475342,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.7524,
|
|
"mean_token_accuracy": 0.16863830834627153,
|
|
"num_tokens": 4143764.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.94091002146403,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.7146,
|
|
"mean_token_accuracy": 0.1708045717742708,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.902195978164673,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.18596015870571136,
|
|
"num_tokens": 4165099.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.959309673309326,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.5603,
|
|
"mean_token_accuracy": 0.18278885036706924,
|
|
"num_tokens": 4178051.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.920917892456055,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.4658,
|
|
"mean_token_accuracy": 0.18581136763095857,
|
|
"num_tokens": 4189987.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.000860071182251,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.5055,
|
|
"mean_token_accuracy": 0.18382351845502853,
|
|
"num_tokens": 4200999.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.7818488597869875,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.4239,
|
|
"mean_token_accuracy": 0.18308175653219222,
|
|
"num_tokens": 4212122.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.977674198150635,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.496,
|
|
"mean_token_accuracy": 0.18817951381206513,
|
|
"num_tokens": 4222580.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.798975467681885,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.3531,
|
|
"mean_token_accuracy": 0.18991094529628755,
|
|
"num_tokens": 4233756.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.865762662887573,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.4645,
|
|
"mean_token_accuracy": 0.18212024122476578,
|
|
"num_tokens": 4245625.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.981123399734497,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.6667,
|
|
"mean_token_accuracy": 0.17891115546226502,
|
|
"num_tokens": 4257788.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.855894231796265,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.4335,
|
|
"mean_token_accuracy": 0.18747661262750626,
|
|
"num_tokens": 4269641.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.923849439620971,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.18903699219226838,
|
|
"num_tokens": 4280952.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.826491022109986,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.4345,
|
|
"mean_token_accuracy": 0.18995490819215774,
|
|
"num_tokens": 4294371.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.792111825942993,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.4359,
|
|
"mean_token_accuracy": 0.18065291047096252,
|
|
"num_tokens": 4307112.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.863767480850219,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.4123,
|
|
"mean_token_accuracy": 0.18984347432851792,
|
|
"num_tokens": 4317199.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.825750827789307,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.4273,
|
|
"mean_token_accuracy": 0.1858452305197716,
|
|
"num_tokens": 4328546.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.87761344909668,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.4795,
|
|
"mean_token_accuracy": 0.18624710738658906,
|
|
"num_tokens": 4338795.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.821884536743164,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.4619,
|
|
"mean_token_accuracy": 0.1902458444237709,
|
|
"num_tokens": 4350373.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.8365589618682865,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.4432,
|
|
"mean_token_accuracy": 0.1811908632516861,
|
|
"num_tokens": 4362801.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.937716817855835,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.5402,
|
|
"mean_token_accuracy": 0.18769277334213258,
|
|
"num_tokens": 4375782.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.759467935562133,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.4247,
|
|
"mean_token_accuracy": 0.18834268003702165,
|
|
"num_tokens": 4385580.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.843483209609985,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.4521,
|
|
"mean_token_accuracy": 0.18452123254537584,
|
|
"num_tokens": 4397434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.8548424243927,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.4941,
|
|
"mean_token_accuracy": 0.1868189662694931,
|
|
"num_tokens": 4407888.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.782220506668091,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.4332,
|
|
"mean_token_accuracy": 0.1842260479927063,
|
|
"num_tokens": 4420010.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.7892741680145265,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.4357,
|
|
"mean_token_accuracy": 0.182176573574543,
|
|
"num_tokens": 4430730.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.809188175201416,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.4291,
|
|
"mean_token_accuracy": 0.19012718945741652,
|
|
"num_tokens": 4442296.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.812354850769043,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.4681,
|
|
"mean_token_accuracy": 0.18536234945058822,
|
|
"num_tokens": 4453996.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.772347259521484,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.4549,
|
|
"mean_token_accuracy": 0.18334894627332687,
|
|
"num_tokens": 4465779.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.707958507537842,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.3975,
|
|
"mean_token_accuracy": 0.19363581985235215,
|
|
"num_tokens": 4477113.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.830227661132812,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.3528,
|
|
"mean_token_accuracy": 0.18912958949804307,
|
|
"num_tokens": 4487920.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.871522283554077,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.5485,
|
|
"mean_token_accuracy": 0.1786806032061577,
|
|
"num_tokens": 4499049.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.593202834779566,
|
|
"eval_loss": 5.982580661773682,
|
|
"eval_mean_token_accuracy": 0.16770411415702918,
|
|
"eval_num_tokens": 4499049.0,
|
|
"eval_runtime": 1.9326,
|
|
"eval_samples_per_second": 1818.765,
|
|
"eval_steps_per_second": 227.669,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.807149600982666,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.3833,
|
|
"mean_token_accuracy": 0.18815410435199736,
|
|
"num_tokens": 4508944.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.849584865570068,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.3963,
|
|
"mean_token_accuracy": 0.18354758322238923,
|
|
"num_tokens": 4518845.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.620027208328247,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.4429,
|
|
"mean_token_accuracy": 0.18136171400547027,
|
|
"num_tokens": 4530416.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.9211140155792235,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.4368,
|
|
"mean_token_accuracy": 0.1837288647890091,
|
|
"num_tokens": 4542630.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.747964763641358,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.4175,
|
|
"mean_token_accuracy": 0.18830150067806245,
|
|
"num_tokens": 4553270.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.626710844039917,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.2918,
|
|
"mean_token_accuracy": 0.19309426844120026,
|
|
"num_tokens": 4564647.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.865239524841309,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.4965,
|
|
"mean_token_accuracy": 0.17623504400253295,
|
|
"num_tokens": 4575507.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.753301572799683,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.4672,
|
|
"mean_token_accuracy": 0.1874419331550598,
|
|
"num_tokens": 4586761.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.77248854637146,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.3877,
|
|
"mean_token_accuracy": 0.1963614135980606,
|
|
"num_tokens": 4597689.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.771483516693115,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.4249,
|
|
"mean_token_accuracy": 0.1831813558936119,
|
|
"num_tokens": 4608886.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.830919599533081,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.4779,
|
|
"mean_token_accuracy": 0.17893050014972686,
|
|
"num_tokens": 4619989.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.708103322982788,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.3652,
|
|
"mean_token_accuracy": 0.1888834834098816,
|
|
"num_tokens": 4630218.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.786612176895142,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.448,
|
|
"mean_token_accuracy": 0.17653173208236694,
|
|
"num_tokens": 4642024.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.845461225509643,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.4587,
|
|
"mean_token_accuracy": 0.1910783141851425,
|
|
"num_tokens": 4653370.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.7815052509307865,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.18658977895975112,
|
|
"num_tokens": 4664522.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.804741907119751,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.5016,
|
|
"mean_token_accuracy": 0.18339928090572358,
|
|
"num_tokens": 4676542.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.738656234741211,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.4189,
|
|
"mean_token_accuracy": 0.18916883766651155,
|
|
"num_tokens": 4687810.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.784382581710815,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.417,
|
|
"mean_token_accuracy": 0.1858016088604927,
|
|
"num_tokens": 4697661.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.730100440979004,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.3396,
|
|
"mean_token_accuracy": 0.18939679116010666,
|
|
"num_tokens": 4708585.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.844074773788452,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.51,
|
|
"mean_token_accuracy": 0.1852951690554619,
|
|
"num_tokens": 4719848.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.826112365722656,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.4725,
|
|
"mean_token_accuracy": 0.17838488668203353,
|
|
"num_tokens": 4731474.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.732689571380615,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.4549,
|
|
"mean_token_accuracy": 0.18692175745964051,
|
|
"num_tokens": 4743081.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.816257524490356,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.4389,
|
|
"mean_token_accuracy": 0.1821078822016716,
|
|
"num_tokens": 4754829.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.737308597564697,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.3943,
|
|
"mean_token_accuracy": 0.18444081246852875,
|
|
"num_tokens": 4765522.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.768337392807007,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.3985,
|
|
"mean_token_accuracy": 0.18821057975292205,
|
|
"num_tokens": 4776246.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.705836391448974,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.4557,
|
|
"mean_token_accuracy": 0.18317651748657227,
|
|
"num_tokens": 4787595.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.891846323013306,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.5233,
|
|
"mean_token_accuracy": 0.17741688936948777,
|
|
"num_tokens": 4798877.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.736809396743775,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.5038,
|
|
"mean_token_accuracy": 0.18460205644369126,
|
|
"num_tokens": 4809542.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.892852926254273,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.6101,
|
|
"mean_token_accuracy": 0.17704683691263198,
|
|
"num_tokens": 4820982.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.706596040725708,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.4529,
|
|
"mean_token_accuracy": 0.18296151906251906,
|
|
"num_tokens": 4831543.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.988536262512207,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.5095,
|
|
"mean_token_accuracy": 0.1808498740196228,
|
|
"num_tokens": 4843221.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.697739982604981,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.369,
|
|
"mean_token_accuracy": 0.1832944557070732,
|
|
"num_tokens": 4855201.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.841998958587647,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.4623,
|
|
"mean_token_accuracy": 0.18599004447460174,
|
|
"num_tokens": 4866387.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.729000234603882,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.3832,
|
|
"mean_token_accuracy": 0.18703173249959945,
|
|
"num_tokens": 4877702.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.730487060546875,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.4508,
|
|
"mean_token_accuracy": 0.18570079952478408,
|
|
"num_tokens": 4888652.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.820274877548218,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.5562,
|
|
"mean_token_accuracy": 0.18157810121774673,
|
|
"num_tokens": 4900995.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.769424152374268,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.3209,
|
|
"mean_token_accuracy": 0.18976007997989655,
|
|
"num_tokens": 4912173.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.626969957351685,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.3509,
|
|
"mean_token_accuracy": 0.19335207492113113,
|
|
"num_tokens": 4923149.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.71943941116333,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.3625,
|
|
"mean_token_accuracy": 0.19323385059833526,
|
|
"num_tokens": 4934156.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.821151971817017,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.18162168562412262,
|
|
"num_tokens": 4946690.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.809539747238159,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.4732,
|
|
"mean_token_accuracy": 0.18971155434846879,
|
|
"num_tokens": 4958208.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.701359367370605,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.4163,
|
|
"mean_token_accuracy": 0.19193969070911407,
|
|
"num_tokens": 4968815.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.759443092346191,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.4028,
|
|
"mean_token_accuracy": 0.19135913997888565,
|
|
"num_tokens": 4979468.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.802725315093994,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.4822,
|
|
"mean_token_accuracy": 0.18243539780378343,
|
|
"num_tokens": 4990841.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.884067678451538,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.5346,
|
|
"mean_token_accuracy": 0.1835401624441147,
|
|
"num_tokens": 5001567.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.7102514743804935,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.392,
|
|
"mean_token_accuracy": 0.18596859723329545,
|
|
"num_tokens": 5013219.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.712259817123413,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.5171,
|
|
"mean_token_accuracy": 0.1820659190416336,
|
|
"num_tokens": 5025771.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.795153856277466,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.4776,
|
|
"mean_token_accuracy": 0.18499909341335297,
|
|
"num_tokens": 5038205.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.823522710800171,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.4215,
|
|
"mean_token_accuracy": 0.1871958017349243,
|
|
"num_tokens": 5048428.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.691017913818359,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.413,
|
|
"mean_token_accuracy": 0.18725276440382005,
|
|
"num_tokens": 5058391.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.664987564086914,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.4368,
|
|
"mean_token_accuracy": 0.1855875328183174,
|
|
"num_tokens": 5069832.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.771575784683227,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.4133,
|
|
"mean_token_accuracy": 0.1956903249025345,
|
|
"num_tokens": 5080198.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.753393507003784,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.4926,
|
|
"mean_token_accuracy": 0.18132777512073517,
|
|
"num_tokens": 5090817.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.689876937866211,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.3921,
|
|
"mean_token_accuracy": 0.18289187848567962,
|
|
"num_tokens": 5102699.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.706895446777343,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.3727,
|
|
"mean_token_accuracy": 0.1870929554104805,
|
|
"num_tokens": 5113922.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.764729595184326,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.4864,
|
|
"mean_token_accuracy": 0.18474530279636384,
|
|
"num_tokens": 5124920.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.7606587409973145,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.17456793487071992,
|
|
"num_tokens": 5136304.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.783349990844727,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.17929618954658508,
|
|
"num_tokens": 5148101.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.852411603927612,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.4543,
|
|
"mean_token_accuracy": 0.1895480766892433,
|
|
"num_tokens": 5159186.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.662082290649414,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.4406,
|
|
"mean_token_accuracy": 0.18616729378700256,
|
|
"num_tokens": 5170542.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.765320634841919,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.5319,
|
|
"mean_token_accuracy": 0.17854655236005784,
|
|
"num_tokens": 5182685.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.749820184707642,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.3484,
|
|
"mean_token_accuracy": 0.19427054077386857,
|
|
"num_tokens": 5192622.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.803789663314819,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.5325,
|
|
"mean_token_accuracy": 0.17829208076000214,
|
|
"num_tokens": 5205068.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.665061664581299,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.4057,
|
|
"mean_token_accuracy": 0.18939344435930253,
|
|
"num_tokens": 5216049.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.644347047805786,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.3981,
|
|
"mean_token_accuracy": 0.18718405961990356,
|
|
"num_tokens": 5227024.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.787604761123657,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.5062,
|
|
"mean_token_accuracy": 0.18521091043949128,
|
|
"num_tokens": 5238862.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.846492528915405,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.5446,
|
|
"mean_token_accuracy": 0.17633910179138185,
|
|
"num_tokens": 5250020.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.773733854293823,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.4604,
|
|
"mean_token_accuracy": 0.1801150918006897,
|
|
"num_tokens": 5260281.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.776580858230591,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.5551,
|
|
"mean_token_accuracy": 0.1811290755867958,
|
|
"num_tokens": 5272909.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.707731819152832,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.3911,
|
|
"mean_token_accuracy": 0.1896710976958275,
|
|
"num_tokens": 5285629.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.786190319061279,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.4558,
|
|
"mean_token_accuracy": 0.18662779033184052,
|
|
"num_tokens": 5296104.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.839002084732056,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.5815,
|
|
"mean_token_accuracy": 0.17700549364089965,
|
|
"num_tokens": 5307148.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.68599534034729,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.3993,
|
|
"mean_token_accuracy": 0.18886441141366958,
|
|
"num_tokens": 5317878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.829762172698975,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.4891,
|
|
"mean_token_accuracy": 0.18109112679958345,
|
|
"num_tokens": 5329603.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.703916358947754,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.4536,
|
|
"mean_token_accuracy": 0.18445155024528503,
|
|
"num_tokens": 5340611.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.6729371547698975,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.3872,
|
|
"mean_token_accuracy": 0.19100804924964904,
|
|
"num_tokens": 5352017.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.807322597503662,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.4433,
|
|
"mean_token_accuracy": 0.1862181842327118,
|
|
"num_tokens": 5362482.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.687590503692627,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.421,
|
|
"mean_token_accuracy": 0.1900341123342514,
|
|
"num_tokens": 5374074.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.6847916603088375,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.2442,
|
|
"mean_token_accuracy": 0.198955699801445,
|
|
"num_tokens": 5384650.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.629435348510742,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.456,
|
|
"mean_token_accuracy": 0.1850106418132782,
|
|
"num_tokens": 5395184.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.635414695739746,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.3054,
|
|
"mean_token_accuracy": 0.19298987686634064,
|
|
"num_tokens": 5405947.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.726533126831055,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.4058,
|
|
"mean_token_accuracy": 0.19271685779094697,
|
|
"num_tokens": 5418237.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.660487365722656,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.4673,
|
|
"mean_token_accuracy": 0.1789378523826599,
|
|
"num_tokens": 5429336.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.735605955123901,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.4248,
|
|
"mean_token_accuracy": 0.1831025019288063,
|
|
"num_tokens": 5440257.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.808689212799072,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.5412,
|
|
"mean_token_accuracy": 0.1831656202673912,
|
|
"num_tokens": 5453079.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.679704046249389,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.412,
|
|
"mean_token_accuracy": 0.1893279656767845,
|
|
"num_tokens": 5464551.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.773027610778809,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.4518,
|
|
"mean_token_accuracy": 0.1891171008348465,
|
|
"num_tokens": 5474797.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.711412668228149,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.4093,
|
|
"mean_token_accuracy": 0.1946184903383255,
|
|
"num_tokens": 5485760.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.702906465530395,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.3939,
|
|
"mean_token_accuracy": 0.1954643726348877,
|
|
"num_tokens": 5496182.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.684745454788208,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.4715,
|
|
"mean_token_accuracy": 0.1875578135251999,
|
|
"num_tokens": 5506848.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.800971651077271,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.5727,
|
|
"mean_token_accuracy": 0.18148866593837737,
|
|
"num_tokens": 5518430.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.767010307312011,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.4517,
|
|
"mean_token_accuracy": 0.18275612592697144,
|
|
"num_tokens": 5529741.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.653870058059693,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.453,
|
|
"mean_token_accuracy": 0.18839110136032106,
|
|
"num_tokens": 5541835.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.698930406570435,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.3497,
|
|
"mean_token_accuracy": 0.1893681764602661,
|
|
"num_tokens": 5552177.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.782176208496094,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.5527,
|
|
"mean_token_accuracy": 0.17554231435060502,
|
|
"num_tokens": 5564440.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.722240018844604,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.3684,
|
|
"mean_token_accuracy": 0.18853142857551575,
|
|
"num_tokens": 5574905.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.7167705535888675,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.4259,
|
|
"mean_token_accuracy": 0.1821403384208679,
|
|
"num_tokens": 5585583.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.622246122360229,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.4311,
|
|
"mean_token_accuracy": 0.18612712472677231,
|
|
"num_tokens": 5597077.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.691310882568359,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.2669,
|
|
"mean_token_accuracy": 0.1939096301794052,
|
|
"num_tokens": 5607904.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.702828073501587,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.4546,
|
|
"mean_token_accuracy": 0.18277173191308976,
|
|
"num_tokens": 5619611.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.726555967330933,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.4646,
|
|
"mean_token_accuracy": 0.1895301401615143,
|
|
"num_tokens": 5630365.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.739557886123658,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.4344,
|
|
"mean_token_accuracy": 0.18483877331018447,
|
|
"num_tokens": 5641008.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.720936632156372,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.4437,
|
|
"mean_token_accuracy": 0.18901977986097335,
|
|
"num_tokens": 5652633.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.682751607894898,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.4608,
|
|
"mean_token_accuracy": 0.175453220307827,
|
|
"num_tokens": 5664108.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.760388135910034,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.4193,
|
|
"mean_token_accuracy": 0.18853973895311354,
|
|
"num_tokens": 5675701.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.617059755325317,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.3587,
|
|
"mean_token_accuracy": 0.19339872449636458,
|
|
"num_tokens": 5687292.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.691484069824218,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.4061,
|
|
"mean_token_accuracy": 0.18431959301233292,
|
|
"num_tokens": 5698905.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.5797929763793945,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.2905,
|
|
"mean_token_accuracy": 0.1919816702604294,
|
|
"num_tokens": 5709678.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.739916181564331,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.41,
|
|
"mean_token_accuracy": 0.19399514198303222,
|
|
"num_tokens": 5720460.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.643140840530395,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.436,
|
|
"mean_token_accuracy": 0.18100445121526718,
|
|
"num_tokens": 5731701.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.666563558578491,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.4435,
|
|
"mean_token_accuracy": 0.18477470725774764,
|
|
"num_tokens": 5741512.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.678272676467896,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.3727,
|
|
"mean_token_accuracy": 0.1849587678909302,
|
|
"num_tokens": 5751882.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.667871332168579,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.18211861997842788,
|
|
"num_tokens": 5761989.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.688571977615356,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.3782,
|
|
"mean_token_accuracy": 0.1825553670525551,
|
|
"num_tokens": 5772532.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.748569631576538,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.4828,
|
|
"mean_token_accuracy": 0.18683700412511825,
|
|
"num_tokens": 5784152.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.711616277694702,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.4473,
|
|
"mean_token_accuracy": 0.18784596920013427,
|
|
"num_tokens": 5793748.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.6141791343688965,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.3989,
|
|
"mean_token_accuracy": 0.18610634952783583,
|
|
"num_tokens": 5804371.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.723416376113891,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.3997,
|
|
"mean_token_accuracy": 0.1904323250055313,
|
|
"num_tokens": 5815470.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.7119588375091555,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.4692,
|
|
"mean_token_accuracy": 0.1842414289712906,
|
|
"num_tokens": 5826037.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.692988729476928,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.349,
|
|
"mean_token_accuracy": 0.19438337683677673,
|
|
"num_tokens": 5836468.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.609593677520752,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.2803,
|
|
"mean_token_accuracy": 0.1908375695347786,
|
|
"num_tokens": 5846371.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.528558444976807,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.3481,
|
|
"mean_token_accuracy": 0.1925843819975853,
|
|
"num_tokens": 5856890.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.703276491165161,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.4199,
|
|
"mean_token_accuracy": 0.18860861361026765,
|
|
"num_tokens": 5868075.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.727088546752929,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.4886,
|
|
"mean_token_accuracy": 0.18812754899263381,
|
|
"num_tokens": 5879579.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.617440986633301,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.3489,
|
|
"mean_token_accuracy": 0.18796394467353822,
|
|
"num_tokens": 5890648.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.6158476829528805,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.3994,
|
|
"mean_token_accuracy": 0.18484208583831788,
|
|
"num_tokens": 5902914.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.7284759998321535,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.4298,
|
|
"mean_token_accuracy": 0.1767915442585945,
|
|
"num_tokens": 5915801.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.762837600708008,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.4921,
|
|
"mean_token_accuracy": 0.1865209236741066,
|
|
"num_tokens": 5928973.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.699053049087524,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.506,
|
|
"mean_token_accuracy": 0.18225849717855452,
|
|
"num_tokens": 5939720.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.7778150081634525,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.4624,
|
|
"mean_token_accuracy": 0.18260607868433,
|
|
"num_tokens": 5951956.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.763799476623535,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.5963,
|
|
"mean_token_accuracy": 0.1797615483403206,
|
|
"num_tokens": 5963735.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.737559413909912,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.352,
|
|
"mean_token_accuracy": 0.19273364096879958,
|
|
"num_tokens": 5974567.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.677313995361328,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.1872257634997368,
|
|
"num_tokens": 5985800.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.641250658035278,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.3247,
|
|
"mean_token_accuracy": 0.19086155891418458,
|
|
"num_tokens": 5997388.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.721637773513794,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.4343,
|
|
"mean_token_accuracy": 0.1899944320321083,
|
|
"num_tokens": 6008739.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.57652177810669,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.3053,
|
|
"mean_token_accuracy": 0.19162104576826094,
|
|
"num_tokens": 6019490.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.621979236602783,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.4308,
|
|
"mean_token_accuracy": 0.18952557146549226,
|
|
"num_tokens": 6030712.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.664051485061646,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.3424,
|
|
"mean_token_accuracy": 0.19479856938123702,
|
|
"num_tokens": 6042161.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.680166864395142,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.3694,
|
|
"mean_token_accuracy": 0.18398038446903228,
|
|
"num_tokens": 6054107.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.619544601440429,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.4503,
|
|
"mean_token_accuracy": 0.19078068137168885,
|
|
"num_tokens": 6065232.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.678529405593872,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.3921,
|
|
"mean_token_accuracy": 0.18789276629686355,
|
|
"num_tokens": 6076294.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.559504795074463,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.3123,
|
|
"mean_token_accuracy": 0.20101085007190705,
|
|
"num_tokens": 6087700.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.634823513031006,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.3865,
|
|
"mean_token_accuracy": 0.19205806851387025,
|
|
"num_tokens": 6098561.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.7596086978912355,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.4152,
|
|
"mean_token_accuracy": 0.18975189030170442,
|
|
"num_tokens": 6109729.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.633274078369141,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.3896,
|
|
"mean_token_accuracy": 0.1907068282365799,
|
|
"num_tokens": 6120468.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.57431116104126,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.3256,
|
|
"mean_token_accuracy": 0.1923707664012909,
|
|
"num_tokens": 6132457.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.632637548446655,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.3736,
|
|
"mean_token_accuracy": 0.18870125859975814,
|
|
"num_tokens": 6142289.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.683001518249512,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.3928,
|
|
"mean_token_accuracy": 0.18968709409236909,
|
|
"num_tokens": 6153786.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.679001092910767,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.3704,
|
|
"mean_token_accuracy": 0.1918751373887062,
|
|
"num_tokens": 6163988.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.61111969947815,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.4718,
|
|
"mean_token_accuracy": 0.1874196857213974,
|
|
"num_tokens": 6175859.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.908012390136719,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.6334,
|
|
"mean_token_accuracy": 0.17676504254341124,
|
|
"num_tokens": 6188116.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.667236995697022,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.432,
|
|
"mean_token_accuracy": 0.19435504376888274,
|
|
"num_tokens": 6198606.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.7496757984161375,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.4775,
|
|
"mean_token_accuracy": 0.18836485743522643,
|
|
"num_tokens": 6210366.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.600676155090332,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.3599,
|
|
"mean_token_accuracy": 0.1886156052350998,
|
|
"num_tokens": 6221376.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.705629666646321,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.3522,
|
|
"mean_token_accuracy": 0.1897873795694775,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.614698266983032,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 5.0476,
|
|
"mean_token_accuracy": 0.21085548996925355,
|
|
"num_tokens": 6241158.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.682580614089966,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.1309,
|
|
"mean_token_accuracy": 0.19781079292297363,
|
|
"num_tokens": 6253811.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.656336975097656,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 5.1236,
|
|
"mean_token_accuracy": 0.20051462352275848,
|
|
"num_tokens": 6265066.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.566812038421631,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.0815,
|
|
"mean_token_accuracy": 0.2031268909573555,
|
|
"num_tokens": 6275889.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.598412704467774,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 5.0551,
|
|
"mean_token_accuracy": 0.2089492678642273,
|
|
"num_tokens": 6287576.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.617827415466309,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 5.1169,
|
|
"mean_token_accuracy": 0.20195962637662887,
|
|
"num_tokens": 6299419.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.62644476890564,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.1006,
|
|
"mean_token_accuracy": 0.21256751716136932,
|
|
"num_tokens": 6310464.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.604681015014648,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.0809,
|
|
"mean_token_accuracy": 0.21031670421361923,
|
|
"num_tokens": 6321442.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.624624633789063,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.1431,
|
|
"mean_token_accuracy": 0.2053651437163353,
|
|
"num_tokens": 6332868.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.713912725448608,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.1459,
|
|
"mean_token_accuracy": 0.2046747013926506,
|
|
"num_tokens": 6345232.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.643095970153809,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.1785,
|
|
"mean_token_accuracy": 0.19962910562753677,
|
|
"num_tokens": 6356524.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.668594741821289,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.1229,
|
|
"mean_token_accuracy": 0.20840518325567245,
|
|
"num_tokens": 6367590.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.6138914108276365,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 5.1352,
|
|
"mean_token_accuracy": 0.19957486540079117,
|
|
"num_tokens": 6378211.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.545917272567749,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.1611,
|
|
"mean_token_accuracy": 0.21044072061777114,
|
|
"num_tokens": 6389595.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.627330112457275,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.1457,
|
|
"mean_token_accuracy": 0.1995941311120987,
|
|
"num_tokens": 6401286.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.63323745727539,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 5.0891,
|
|
"mean_token_accuracy": 0.21249493956565857,
|
|
"num_tokens": 6411946.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.597933721542359,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 5.059,
|
|
"mean_token_accuracy": 0.2097697600722313,
|
|
"num_tokens": 6422463.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.4642339706420895,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.0497,
|
|
"mean_token_accuracy": 0.20787030160427095,
|
|
"num_tokens": 6433576.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.555215692520141,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 5.0154,
|
|
"mean_token_accuracy": 0.213714163005352,
|
|
"num_tokens": 6444058.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.643922615051269,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.2238,
|
|
"mean_token_accuracy": 0.20050896108150482,
|
|
"num_tokens": 6453872.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.547750759124756,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 5.0981,
|
|
"mean_token_accuracy": 0.2021889641880989,
|
|
"num_tokens": 6465367.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.573264932632446,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 5.0483,
|
|
"mean_token_accuracy": 0.2199981227517128,
|
|
"num_tokens": 6476655.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.537868309020996,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.1326,
|
|
"mean_token_accuracy": 0.2010281875729561,
|
|
"num_tokens": 6488348.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.604411268234253,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.1372,
|
|
"mean_token_accuracy": 0.20590841472148896,
|
|
"num_tokens": 6500127.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.507631635665893,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 5.0962,
|
|
"mean_token_accuracy": 0.20493392795324325,
|
|
"num_tokens": 6512772.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.633540773391724,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.1172,
|
|
"mean_token_accuracy": 0.20897855907678603,
|
|
"num_tokens": 6523359.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.528751182556152,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.0883,
|
|
"mean_token_accuracy": 0.20566013008356093,
|
|
"num_tokens": 6533679.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.543372440338135,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 5.0461,
|
|
"mean_token_accuracy": 0.20662636458873748,
|
|
"num_tokens": 6544694.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.599698448181153,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.0897,
|
|
"mean_token_accuracy": 0.20754512399435043,
|
|
"num_tokens": 6555858.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.5275777816772464,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.1295,
|
|
"mean_token_accuracy": 0.20758441537618638,
|
|
"num_tokens": 6568214.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.499366903305054,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.0788,
|
|
"mean_token_accuracy": 0.2093128427863121,
|
|
"num_tokens": 6579646.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.55049877166748,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 5.0447,
|
|
"mean_token_accuracy": 0.2128044471144676,
|
|
"num_tokens": 6590683.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.634090852737427,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.1678,
|
|
"mean_token_accuracy": 0.20399952083826065,
|
|
"num_tokens": 6601197.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.5481596946716305,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.2111,
|
|
"mean_token_accuracy": 0.19378461986780166,
|
|
"num_tokens": 6613564.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.620682859420777,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.0836,
|
|
"mean_token_accuracy": 0.20560076981782913,
|
|
"num_tokens": 6624696.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.571142387390137,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.1002,
|
|
"mean_token_accuracy": 0.21549748331308366,
|
|
"num_tokens": 6636302.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.614317941665649,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.1348,
|
|
"mean_token_accuracy": 0.20918277353048326,
|
|
"num_tokens": 6646517.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.49672679901123,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.1052,
|
|
"mean_token_accuracy": 0.20810072124004364,
|
|
"num_tokens": 6657768.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.468635559082031,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 5.0278,
|
|
"mean_token_accuracy": 0.21080595403909683,
|
|
"num_tokens": 6669110.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.526268577575683,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.1146,
|
|
"mean_token_accuracy": 0.20088636577129365,
|
|
"num_tokens": 6680925.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.621420097351074,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.1968,
|
|
"mean_token_accuracy": 0.20468207746744155,
|
|
"num_tokens": 6692923.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.612257051467895,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.0758,
|
|
"mean_token_accuracy": 0.21096018105745315,
|
|
"num_tokens": 6702692.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.564406156539917,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 5.1453,
|
|
"mean_token_accuracy": 0.20293173789978028,
|
|
"num_tokens": 6713830.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.551304197311401,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.1366,
|
|
"mean_token_accuracy": 0.20120118111371993,
|
|
"num_tokens": 6725564.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.480319499969482,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.0978,
|
|
"mean_token_accuracy": 0.20480770021677017,
|
|
"num_tokens": 6736563.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.36486209414222,
|
|
"eval_loss": 5.875669956207275,
|
|
"eval_mean_token_accuracy": 0.1756818814541806,
|
|
"eval_num_tokens": 6736563.0,
|
|
"eval_runtime": 1.9319,
|
|
"eval_samples_per_second": 1819.489,
|
|
"eval_steps_per_second": 227.76,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.477034616470337,
|
|
"epoch": 3.248783126014062,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004374593708595049,
|
|
"loss": 5.0778,
|
|
"mean_token_accuracy": 0.2087254583835602,
|
|
"num_tokens": 6745845.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.477921485900879,
|
|
"epoch": 3.2541914548404542,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043716232848439555,
|
|
"loss": 5.0238,
|
|
"mean_token_accuracy": 0.214997561275959,
|
|
"num_tokens": 6757450.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.623067045211792,
|
|
"epoch": 3.259599783666847,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043686469681100985,
|
|
"loss": 5.18,
|
|
"mean_token_accuracy": 0.20380834639072418,
|
|
"num_tokens": 6768542.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.534489154815674,
|
|
"epoch": 3.2650081124932395,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043656647692094186,
|
|
"loss": 5.0934,
|
|
"mean_token_accuracy": 0.22307817190885543,
|
|
"num_tokens": 6779873.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.529124355316162,
|
|
"epoch": 3.2704164413196324,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004362676698979233,
|
|
"loss": 5.1603,
|
|
"mean_token_accuracy": 0.20711462795734406,
|
|
"num_tokens": 6790898.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.569097089767456,
|
|
"epoch": 3.275824770146025,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043596827682781974,
|
|
"loss": 5.1182,
|
|
"mean_token_accuracy": 0.2055336758494377,
|
|
"num_tokens": 6802175.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.648894357681274,
|
|
"epoch": 3.2812330989724177,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004356682987986262,
|
|
"loss": 5.2412,
|
|
"mean_token_accuracy": 0.20254767835140228,
|
|
"num_tokens": 6813675.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.439459562301636,
|
|
"epoch": 3.28664142779881,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004353677369004635,
|
|
"loss": 5.0957,
|
|
"mean_token_accuracy": 0.2072897344827652,
|
|
"num_tokens": 6825020.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.507056760787964,
|
|
"epoch": 3.292049756625203,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043506659222557426,
|
|
"loss": 5.0006,
|
|
"mean_token_accuracy": 0.22333719283342363,
|
|
"num_tokens": 6834954.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.620904874801636,
|
|
"epoch": 3.2974580854515954,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000434764865868319,
|
|
"loss": 5.2454,
|
|
"mean_token_accuracy": 0.1947852447628975,
|
|
"num_tokens": 6847958.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.567670154571533,
|
|
"epoch": 3.302866414277988,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000434462558925172,
|
|
"loss": 5.0716,
|
|
"mean_token_accuracy": 0.21577776074409485,
|
|
"num_tokens": 6858782.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.398960065841675,
|
|
"epoch": 3.3082747431043806,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043415967249471734,
|
|
"loss": 5.0752,
|
|
"mean_token_accuracy": 0.20665508806705474,
|
|
"num_tokens": 6869807.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.5898058891296385,
|
|
"epoch": 3.3136830719307735,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000433856207677645,
|
|
"loss": 5.1167,
|
|
"mean_token_accuracy": 0.2124245509505272,
|
|
"num_tokens": 6880642.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.593895387649536,
|
|
"epoch": 3.319091400757166,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000433552165576747,
|
|
"loss": 5.2227,
|
|
"mean_token_accuracy": 0.1903145879507065,
|
|
"num_tokens": 6891677.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.595327711105346,
|
|
"epoch": 3.3244997295835588,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043324754729691275,
|
|
"loss": 5.108,
|
|
"mean_token_accuracy": 0.20558954626321793,
|
|
"num_tokens": 6903383.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.595440340042114,
|
|
"epoch": 3.329908058409951,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043294235394512594,
|
|
"loss": 5.1616,
|
|
"mean_token_accuracy": 0.2116692215204239,
|
|
"num_tokens": 6915874.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.490843057632446,
|
|
"epoch": 3.335316387236344,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004326365866304599,
|
|
"loss": 4.9888,
|
|
"mean_token_accuracy": 0.2135483980178833,
|
|
"num_tokens": 6926356.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.554906368255615,
|
|
"epoch": 3.3407247160627365,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004323302464640738,
|
|
"loss": 5.1311,
|
|
"mean_token_accuracy": 0.2131600260734558,
|
|
"num_tokens": 6937678.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.582224798202515,
|
|
"epoch": 3.3461330448891293,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043202333455920843,
|
|
"loss": 5.1219,
|
|
"mean_token_accuracy": 0.20725697576999663,
|
|
"num_tokens": 6949447.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.509426879882812,
|
|
"epoch": 3.3515413737155217,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004317158520311824,
|
|
"loss": 5.0435,
|
|
"mean_token_accuracy": 0.20771547257900239,
|
|
"num_tokens": 6960347.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.431267499923706,
|
|
"epoch": 3.3569497025419146,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004314077999973879,
|
|
"loss": 5.0497,
|
|
"mean_token_accuracy": 0.20607315450906755,
|
|
"num_tokens": 6971353.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.575613403320313,
|
|
"epoch": 3.362358031368307,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004310991795772869,
|
|
"loss": 5.1946,
|
|
"mean_token_accuracy": 0.19296876043081285,
|
|
"num_tokens": 6982519.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.4554633617401125,
|
|
"epoch": 3.3677663601947,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004307899918924065,
|
|
"loss": 5.0344,
|
|
"mean_token_accuracy": 0.21482468247413636,
|
|
"num_tokens": 6994021.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.5181718349456785,
|
|
"epoch": 3.3731746890210923,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004304802380663356,
|
|
"loss": 5.1398,
|
|
"mean_token_accuracy": 0.2069343864917755,
|
|
"num_tokens": 7005505.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.499838495254517,
|
|
"epoch": 3.378583017847485,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043016991922472015,
|
|
"loss": 5.1548,
|
|
"mean_token_accuracy": 0.20692490488290788,
|
|
"num_tokens": 7016123.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.614477443695068,
|
|
"epoch": 3.383991346673878,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00042985903649525977,
|
|
"loss": 5.106,
|
|
"mean_token_accuracy": 0.20939873307943344,
|
|
"num_tokens": 7027083.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.56683382987976,
|
|
"epoch": 3.3893996755002704,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042954759100770275,
|
|
"loss": 5.2087,
|
|
"mean_token_accuracy": 0.19724143743515016,
|
|
"num_tokens": 7037892.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.547694826126099,
|
|
"epoch": 3.394808004326663,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042923558389384303,
|
|
"loss": 5.1748,
|
|
"mean_token_accuracy": 0.20102300941944123,
|
|
"num_tokens": 7048996.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.629776382446289,
|
|
"epoch": 3.4002163331530557,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004289230162875149,
|
|
"loss": 5.1915,
|
|
"mean_token_accuracy": 0.19786946177482606,
|
|
"num_tokens": 7060884.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.5279693603515625,
|
|
"epoch": 3.4056246619794486,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042860988932458987,
|
|
"loss": 5.1936,
|
|
"mean_token_accuracy": 0.2041422590613365,
|
|
"num_tokens": 7072550.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.541787385940552,
|
|
"epoch": 3.411032990805841,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004282962041429721,
|
|
"loss": 5.1203,
|
|
"mean_token_accuracy": 0.21275101900100707,
|
|
"num_tokens": 7085325.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.486838388442993,
|
|
"epoch": 3.4164413196322334,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00042798196188259427,
|
|
"loss": 4.9935,
|
|
"mean_token_accuracy": 0.2071891874074936,
|
|
"num_tokens": 7095691.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.425812482833862,
|
|
"epoch": 3.4218496484586263,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004276671636854133,
|
|
"loss": 4.974,
|
|
"mean_token_accuracy": 0.22325226217508315,
|
|
"num_tokens": 7106552.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.4184832096099855,
|
|
"epoch": 3.427257977285019,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042735181069540677,
|
|
"loss": 5.1161,
|
|
"mean_token_accuracy": 0.2023777574300766,
|
|
"num_tokens": 7119708.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.538694620132446,
|
|
"epoch": 3.4326663061114115,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004270359040585681,
|
|
"loss": 5.1114,
|
|
"mean_token_accuracy": 0.2060562402009964,
|
|
"num_tokens": 7130663.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.410373973846435,
|
|
"epoch": 3.4380746349378044,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00042671944492290277,
|
|
"loss": 5.0489,
|
|
"mean_token_accuracy": 0.20108197778463363,
|
|
"num_tokens": 7141513.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.474251174926758,
|
|
"epoch": 3.443482963764197,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004264024344384239,
|
|
"loss": 5.105,
|
|
"mean_token_accuracy": 0.20920044034719468,
|
|
"num_tokens": 7153387.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.584701156616211,
|
|
"epoch": 3.4488912925905897,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00042608487375714847,
|
|
"loss": 5.1117,
|
|
"mean_token_accuracy": 0.20869659632444382,
|
|
"num_tokens": 7164905.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.489509057998657,
|
|
"epoch": 3.454299621416982,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00042576676403309265,
|
|
"loss": 5.1315,
|
|
"mean_token_accuracy": 0.2072004646062851,
|
|
"num_tokens": 7175636.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.580786657333374,
|
|
"epoch": 3.459707950243375,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004254481064222679,
|
|
"loss": 5.1376,
|
|
"mean_token_accuracy": 0.2005324363708496,
|
|
"num_tokens": 7185903.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.4354987144470215,
|
|
"epoch": 3.4651162790697674,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042512890208267677,
|
|
"loss": 5.0636,
|
|
"mean_token_accuracy": 0.20085968673229218,
|
|
"num_tokens": 7198257.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.548749876022339,
|
|
"epoch": 3.4705246078961602,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042480915217430853,
|
|
"loss": 5.0454,
|
|
"mean_token_accuracy": 0.2147142469882965,
|
|
"num_tokens": 7208839.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.5313129901885985,
|
|
"epoch": 3.4759329367225527,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042448885785913507,
|
|
"loss": 5.1501,
|
|
"mean_token_accuracy": 0.20367118418216706,
|
|
"num_tokens": 7221078.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.510398101806641,
|
|
"epoch": 3.4813412655489455,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042416802030110677,
|
|
"loss": 5.0621,
|
|
"mean_token_accuracy": 0.2134926751255989,
|
|
"num_tokens": 7230287.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.449747896194458,
|
|
"epoch": 3.486749594375338,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042384664066614787,
|
|
"loss": 5.1444,
|
|
"mean_token_accuracy": 0.20988204777240754,
|
|
"num_tokens": 7242253.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.550597238540649,
|
|
"epoch": 3.492157923201731,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004235247201221528,
|
|
"loss": 5.1173,
|
|
"mean_token_accuracy": 0.2129192590713501,
|
|
"num_tokens": 7254098.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.384794902801514,
|
|
"epoch": 3.497566252028123,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004232022598389816,
|
|
"loss": 4.9997,
|
|
"mean_token_accuracy": 0.2175978496670723,
|
|
"num_tokens": 7264153.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.4585036277771,
|
|
"epoch": 3.502974580854516,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004228792609884557,
|
|
"loss": 5.129,
|
|
"mean_token_accuracy": 0.19972793757915497,
|
|
"num_tokens": 7276431.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.504396390914917,
|
|
"epoch": 3.5083829096809085,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042255572474435364,
|
|
"loss": 5.0661,
|
|
"mean_token_accuracy": 0.20680945366621017,
|
|
"num_tokens": 7287882.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.483049917221069,
|
|
"epoch": 3.5137912385073014,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004222316522824069,
|
|
"loss": 5.1973,
|
|
"mean_token_accuracy": 0.1974351167678833,
|
|
"num_tokens": 7298846.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.493984317779541,
|
|
"epoch": 3.5191995673336938,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004219070447802955,
|
|
"loss": 5.0907,
|
|
"mean_token_accuracy": 0.2046854630112648,
|
|
"num_tokens": 7310372.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.606421995162964,
|
|
"epoch": 3.5246078961600866,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042158190341764384,
|
|
"loss": 5.1775,
|
|
"mean_token_accuracy": 0.2005749002099037,
|
|
"num_tokens": 7322200.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.522876214981079,
|
|
"epoch": 3.530016224986479,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042125622937601645,
|
|
"loss": 5.0589,
|
|
"mean_token_accuracy": 0.21006799787282943,
|
|
"num_tokens": 7331824.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.460476970672607,
|
|
"epoch": 3.535424553812872,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004209300238389135,
|
|
"loss": 5.0561,
|
|
"mean_token_accuracy": 0.2086014285683632,
|
|
"num_tokens": 7341522.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.495606231689453,
|
|
"epoch": 3.5408328826392643,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004206032879917667,
|
|
"loss": 5.1334,
|
|
"mean_token_accuracy": 0.21335540115833282,
|
|
"num_tokens": 7353312.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.553105783462525,
|
|
"epoch": 3.546241211465657,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004202760230219348,
|
|
"loss": 5.1341,
|
|
"mean_token_accuracy": 0.21407779902219773,
|
|
"num_tokens": 7363985.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.539524984359741,
|
|
"epoch": 3.5516495402920496,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041994823011869955,
|
|
"loss": 5.1154,
|
|
"mean_token_accuracy": 0.20620542466640474,
|
|
"num_tokens": 7373872.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.558499526977539,
|
|
"epoch": 3.5570578691184425,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00041961991047326116,
|
|
"loss": 5.1424,
|
|
"mean_token_accuracy": 0.21372735798358916,
|
|
"num_tokens": 7386759.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.444520282745361,
|
|
"epoch": 3.5624661979448353,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004192910652787338,
|
|
"loss": 5.0744,
|
|
"mean_token_accuracy": 0.20959434062242507,
|
|
"num_tokens": 7398001.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.549804401397705,
|
|
"epoch": 3.5678745267712277,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004189616957301419,
|
|
"loss": 5.1322,
|
|
"mean_token_accuracy": 0.204277004301548,
|
|
"num_tokens": 7409159.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.441422653198242,
|
|
"epoch": 3.57328285559762,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004186318030244149,
|
|
"loss": 5.0597,
|
|
"mean_token_accuracy": 0.2096458375453949,
|
|
"num_tokens": 7421028.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.474045181274414,
|
|
"epoch": 3.578691184424013,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041830138836038383,
|
|
"loss": 5.1511,
|
|
"mean_token_accuracy": 0.20500203222036362,
|
|
"num_tokens": 7432321.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.59620885848999,
|
|
"epoch": 3.584099513250406,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004179704529387763,
|
|
"loss": 5.1694,
|
|
"mean_token_accuracy": 0.20975432842969893,
|
|
"num_tokens": 7443607.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.479234647750855,
|
|
"epoch": 3.5895078420767983,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004176389979622125,
|
|
"loss": 5.0564,
|
|
"mean_token_accuracy": 0.20859690755605698,
|
|
"num_tokens": 7455222.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.56280746459961,
|
|
"epoch": 3.5949161709031907,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004173070246352003,
|
|
"loss": 5.212,
|
|
"mean_token_accuracy": 0.20601428300142288,
|
|
"num_tokens": 7468511.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.503955841064453,
|
|
"epoch": 3.6003244997295836,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041697453416413184,
|
|
"loss": 5.1903,
|
|
"mean_token_accuracy": 0.21073466092348098,
|
|
"num_tokens": 7479810.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.476354122161865,
|
|
"epoch": 3.6057328285559764,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000416641527757278,
|
|
"loss": 5.0218,
|
|
"mean_token_accuracy": 0.20666715502738953,
|
|
"num_tokens": 7491405.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.5011640071868895,
|
|
"epoch": 3.611141157382369,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041630800662478514,
|
|
"loss": 5.1532,
|
|
"mean_token_accuracy": 0.1990375906229019,
|
|
"num_tokens": 7501820.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.528839254379273,
|
|
"epoch": 3.6165494862087613,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00041597397197866957,
|
|
"loss": 5.0976,
|
|
"mean_token_accuracy": 0.20671522468328477,
|
|
"num_tokens": 7512165.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.451577520370483,
|
|
"epoch": 3.621957815035154,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004156394250328141,
|
|
"loss": 5.0915,
|
|
"mean_token_accuracy": 0.2108155071735382,
|
|
"num_tokens": 7523755.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.434593868255615,
|
|
"epoch": 3.627366143861547,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000415304367002963,
|
|
"loss": 5.0981,
|
|
"mean_token_accuracy": 0.2049677386879921,
|
|
"num_tokens": 7535674.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.4800276279449465,
|
|
"epoch": 3.6327744726879394,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041496879910671796,
|
|
"loss": 5.0727,
|
|
"mean_token_accuracy": 0.2093690127134323,
|
|
"num_tokens": 7546885.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.482887125015258,
|
|
"epoch": 3.638182801514332,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004146327225635336,
|
|
"loss": 5.0632,
|
|
"mean_token_accuracy": 0.21030590832233428,
|
|
"num_tokens": 7557718.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.451445770263672,
|
|
"epoch": 3.6435911303407247,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004142961385947127,
|
|
"loss": 5.0331,
|
|
"mean_token_accuracy": 0.2030412808060646,
|
|
"num_tokens": 7569472.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.5427018165588375,
|
|
"epoch": 3.6489994591671175,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041395904842340215,
|
|
"loss": 5.0808,
|
|
"mean_token_accuracy": 0.20053068101406096,
|
|
"num_tokens": 7579954.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.444069147109985,
|
|
"epoch": 3.65440778799351,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004136214532745887,
|
|
"loss": 5.0231,
|
|
"mean_token_accuracy": 0.2148769363760948,
|
|
"num_tokens": 7590717.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.544438886642456,
|
|
"epoch": 3.6598161168199024,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004132833543750936,
|
|
"loss": 5.3422,
|
|
"mean_token_accuracy": 0.19506975263357162,
|
|
"num_tokens": 7602802.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.663115406036377,
|
|
"epoch": 3.6652244456462952,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041294475295356917,
|
|
"loss": 5.2622,
|
|
"mean_token_accuracy": 0.19920933544635772,
|
|
"num_tokens": 7614944.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.549839067459106,
|
|
"epoch": 3.670632774472688,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004126056502404937,
|
|
"loss": 5.1632,
|
|
"mean_token_accuracy": 0.20209028720855712,
|
|
"num_tokens": 7626449.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.4694384098052975,
|
|
"epoch": 3.6760411032990805,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004122660474681672,
|
|
"loss": 5.0205,
|
|
"mean_token_accuracy": 0.21435775905847548,
|
|
"num_tokens": 7637878.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.557307624816895,
|
|
"epoch": 3.6814494321254734,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041192594587070706,
|
|
"loss": 5.2475,
|
|
"mean_token_accuracy": 0.19705326855182648,
|
|
"num_tokens": 7649428.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.598140048980713,
|
|
"epoch": 3.686857760951866,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041158534668404325,
|
|
"loss": 5.1821,
|
|
"mean_token_accuracy": 0.1988816738128662,
|
|
"num_tokens": 7660527.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.436961364746094,
|
|
"epoch": 3.6922660897782587,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00041124425114591386,
|
|
"loss": 5.1057,
|
|
"mean_token_accuracy": 0.21171297878026962,
|
|
"num_tokens": 7671249.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.538482189178467,
|
|
"epoch": 3.697674418604651,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00041090266049586107,
|
|
"loss": 5.0939,
|
|
"mean_token_accuracy": 0.20485997796058655,
|
|
"num_tokens": 7681824.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.486778497695923,
|
|
"epoch": 3.703082747431044,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004105605759752258,
|
|
"loss": 5.1454,
|
|
"mean_token_accuracy": 0.2030759871006012,
|
|
"num_tokens": 7691916.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.534282684326172,
|
|
"epoch": 3.7084910762574363,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00041021799882714436,
|
|
"loss": 5.1964,
|
|
"mean_token_accuracy": 0.20107840299606322,
|
|
"num_tokens": 7703355.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.449303102493286,
|
|
"epoch": 3.713899405083829,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004098749302965426,
|
|
"loss": 5.117,
|
|
"mean_token_accuracy": 0.20992861241102218,
|
|
"num_tokens": 7714731.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.494414329528809,
|
|
"epoch": 3.7193077339102216,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004095313716301324,
|
|
"loss": 5.0761,
|
|
"mean_token_accuracy": 0.21541002839803697,
|
|
"num_tokens": 7725696.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.478948307037354,
|
|
"epoch": 3.7247160627366145,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004091873240764069,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.2076568454504013,
|
|
"num_tokens": 7737121.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.559846687316894,
|
|
"epoch": 3.730124391563007,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00040884278888563566,
|
|
"loss": 5.1699,
|
|
"mean_token_accuracy": 0.2058560162782669,
|
|
"num_tokens": 7749088.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.620992994308471,
|
|
"epoch": 3.7355327203893998,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040849776730986014,
|
|
"loss": 5.255,
|
|
"mean_token_accuracy": 0.20271677225828172,
|
|
"num_tokens": 7760676.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.54010043144226,
|
|
"epoch": 3.740941049215792,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004081522606028899,
|
|
"loss": 5.1728,
|
|
"mean_token_accuracy": 0.1997041642665863,
|
|
"num_tokens": 7772560.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.531512641906739,
|
|
"epoch": 3.746349378042185,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040780627002029674,
|
|
"loss": 5.0796,
|
|
"mean_token_accuracy": 0.21351829022169114,
|
|
"num_tokens": 7782206.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.494977807998657,
|
|
"epoch": 3.7517577068685775,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004074597968194116,
|
|
"loss": 5.2073,
|
|
"mean_token_accuracy": 0.19656713753938676,
|
|
"num_tokens": 7793862.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.56919903755188,
|
|
"epoch": 3.7571660356949703,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004071128422593186,
|
|
"loss": 5.2218,
|
|
"mean_token_accuracy": 0.20098826736211778,
|
|
"num_tokens": 7805291.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.581046009063721,
|
|
"epoch": 3.7625743645213627,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004067654076008515,
|
|
"loss": 5.1824,
|
|
"mean_token_accuracy": 0.20532560646533965,
|
|
"num_tokens": 7816141.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.4656984329223635,
|
|
"epoch": 3.7679826933477556,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040641749410658877,
|
|
"loss": 5.1549,
|
|
"mean_token_accuracy": 0.19738420099020004,
|
|
"num_tokens": 7826209.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.547751474380493,
|
|
"epoch": 3.773391022174148,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004060691030408487,
|
|
"loss": 5.1185,
|
|
"mean_token_accuracy": 0.19970337599515914,
|
|
"num_tokens": 7836734.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.543689632415772,
|
|
"epoch": 3.778799351000541,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040572023566968536,
|
|
"loss": 5.1897,
|
|
"mean_token_accuracy": 0.19948069304227828,
|
|
"num_tokens": 7848283.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.521392440795898,
|
|
"epoch": 3.7842076798269337,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040537089326088345,
|
|
"loss": 5.0821,
|
|
"mean_token_accuracy": 0.1997026115655899,
|
|
"num_tokens": 7861226.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.401424074172974,
|
|
"epoch": 3.789616008653326,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004050210770839541,
|
|
"loss": 5.0608,
|
|
"mean_token_accuracy": 0.21395482420921325,
|
|
"num_tokens": 7872426.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.588377380371094,
|
|
"epoch": 3.7950243374797186,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004046707884101303,
|
|
"loss": 5.1394,
|
|
"mean_token_accuracy": 0.2114765614271164,
|
|
"num_tokens": 7883884.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.4410529136657715,
|
|
"epoch": 3.8004326663061114,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040432002851236164,
|
|
"loss": 5.0823,
|
|
"mean_token_accuracy": 0.21134976893663407,
|
|
"num_tokens": 7895415.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.455625724792481,
|
|
"epoch": 3.8058409951325043,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004039687986653106,
|
|
"loss": 5.1752,
|
|
"mean_token_accuracy": 0.20156481862068176,
|
|
"num_tokens": 7907444.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.597575378417969,
|
|
"epoch": 3.8112493239588967,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040361710014534714,
|
|
"loss": 5.1483,
|
|
"mean_token_accuracy": 0.2022449016571045,
|
|
"num_tokens": 7918811.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.566057825088501,
|
|
"epoch": 3.816657652785289,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004032649342305445,
|
|
"loss": 5.2057,
|
|
"mean_token_accuracy": 0.2016289085149765,
|
|
"num_tokens": 7930657.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.374592065811157,
|
|
"epoch": 3.822065981611682,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004029123022006745,
|
|
"loss": 5.0479,
|
|
"mean_token_accuracy": 0.20045577138662338,
|
|
"num_tokens": 7942469.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.553105974197388,
|
|
"epoch": 3.827474310438075,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004025592053372027,
|
|
"loss": 5.2169,
|
|
"mean_token_accuracy": 0.20230122655630112,
|
|
"num_tokens": 7954034.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.508181428909301,
|
|
"epoch": 3.8328826392644673,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004022056449232839,
|
|
"loss": 5.1221,
|
|
"mean_token_accuracy": 0.20578389316797258,
|
|
"num_tokens": 7966079.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.400002288818359,
|
|
"epoch": 3.8382909680908597,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004018516222437575,
|
|
"loss": 5.0816,
|
|
"mean_token_accuracy": 0.21338423639535903,
|
|
"num_tokens": 7977785.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.565724468231201,
|
|
"epoch": 3.8436992969172525,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004014971385851428,
|
|
"loss": 5.2487,
|
|
"mean_token_accuracy": 0.20151972621679307,
|
|
"num_tokens": 7988874.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.489980268478393,
|
|
"epoch": 3.8491076257436454,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000401142195235634,
|
|
"loss": 5.0433,
|
|
"mean_token_accuracy": 0.21224800497293472,
|
|
"num_tokens": 7999438.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.492645788192749,
|
|
"epoch": 3.854515954570038,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004007867934850963,
|
|
"loss": 5.116,
|
|
"mean_token_accuracy": 0.20708084255456924,
|
|
"num_tokens": 8009658.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.524990892410278,
|
|
"epoch": 3.8599242833964302,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004004309346250603,
|
|
"loss": 5.0948,
|
|
"mean_token_accuracy": 0.20436943620443343,
|
|
"num_tokens": 8020988.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.38110032081604,
|
|
"epoch": 3.865332612222823,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004000746199487181,
|
|
"loss": 4.9885,
|
|
"mean_token_accuracy": 0.20869673937559127,
|
|
"num_tokens": 8031329.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.438455724716187,
|
|
"epoch": 3.870740941049216,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000399717850750918,
|
|
"loss": 5.0629,
|
|
"mean_token_accuracy": 0.2046133756637573,
|
|
"num_tokens": 8041693.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.51210126876831,
|
|
"epoch": 3.8761492698756084,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039936062832816,
|
|
"loss": 5.1454,
|
|
"mean_token_accuracy": 0.207726913690567,
|
|
"num_tokens": 8053123.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.5538458824157715,
|
|
"epoch": 3.8815575987020012,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003990029539785913,
|
|
"loss": 5.1341,
|
|
"mean_token_accuracy": 0.20657602101564407,
|
|
"num_tokens": 8064270.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.454948377609253,
|
|
"epoch": 3.8869659275283936,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039864482900200135,
|
|
"loss": 5.0618,
|
|
"mean_token_accuracy": 0.21040822863578795,
|
|
"num_tokens": 8074631.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.459717416763306,
|
|
"epoch": 3.8923742563547865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003982862546998169,
|
|
"loss": 5.1479,
|
|
"mean_token_accuracy": 0.20727614909410477,
|
|
"num_tokens": 8086794.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.524269247055054,
|
|
"epoch": 3.897782585181179,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003979272323750981,
|
|
"loss": 5.1299,
|
|
"mean_token_accuracy": 0.20729521214962005,
|
|
"num_tokens": 8097595.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.492750549316407,
|
|
"epoch": 3.903190914007572,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003975677633325327,
|
|
"loss": 5.0834,
|
|
"mean_token_accuracy": 0.21169006675481797,
|
|
"num_tokens": 8107174.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.5097087860107425,
|
|
"epoch": 3.908599242833964,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003972078488784321,
|
|
"loss": 5.1286,
|
|
"mean_token_accuracy": 0.21394529640674592,
|
|
"num_tokens": 8118065.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.412583351135254,
|
|
"epoch": 3.914007571660357,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003968474903207263,
|
|
"loss": 5.0914,
|
|
"mean_token_accuracy": 0.20236337631940843,
|
|
"num_tokens": 8130604.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.550129985809326,
|
|
"epoch": 3.9194159004867495,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003964866889689592,
|
|
"loss": 5.15,
|
|
"mean_token_accuracy": 0.20828114300966263,
|
|
"num_tokens": 8141808.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.558591413497925,
|
|
"epoch": 3.9248242293131423,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003961254461342834,
|
|
"loss": 5.1224,
|
|
"mean_token_accuracy": 0.201833076775074,
|
|
"num_tokens": 8152101.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.509577751159668,
|
|
"epoch": 3.9302325581395348,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003957637631294566,
|
|
"loss": 5.154,
|
|
"mean_token_accuracy": 0.20369689911603928,
|
|
"num_tokens": 8163963.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.471685552597046,
|
|
"epoch": 3.9356408869659276,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039540164126883554,
|
|
"loss": 5.0895,
|
|
"mean_token_accuracy": 0.20885697603225709,
|
|
"num_tokens": 8174063.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.375539350509643,
|
|
"epoch": 3.94104921579232,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003950390818683718,
|
|
"loss": 5.079,
|
|
"mean_token_accuracy": 0.20944282710552214,
|
|
"num_tokens": 8186800.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.480473375320434,
|
|
"epoch": 3.946457544618713,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003946760862456071,
|
|
"loss": 5.1539,
|
|
"mean_token_accuracy": 0.2049423038959503,
|
|
"num_tokens": 8198420.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.59925389289856,
|
|
"epoch": 3.9518658734451053,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003943126557196685,
|
|
"loss": 5.1739,
|
|
"mean_token_accuracy": 0.19827647656202316,
|
|
"num_tokens": 8209384.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.477585172653198,
|
|
"epoch": 3.957274202271498,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039394879161126333,
|
|
"loss": 5.1374,
|
|
"mean_token_accuracy": 0.2023727759718895,
|
|
"num_tokens": 8220356.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.443531703948975,
|
|
"epoch": 3.9626825310978906,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003935844952426745,
|
|
"loss": 5.0697,
|
|
"mean_token_accuracy": 0.2037736490368843,
|
|
"num_tokens": 8231597.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.430696105957031,
|
|
"epoch": 3.9680908599242835,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039321976793775586,
|
|
"loss": 5.0421,
|
|
"mean_token_accuracy": 0.21577501595020293,
|
|
"num_tokens": 8242554.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.5594230651855465,
|
|
"epoch": 3.973499188750676,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003928546110219275,
|
|
"loss": 5.277,
|
|
"mean_token_accuracy": 0.1999075621366501,
|
|
"num_tokens": 8253396.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.48715500831604,
|
|
"epoch": 3.9789075175770687,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039248902582217024,
|
|
"loss": 5.0712,
|
|
"mean_token_accuracy": 0.21160622984170913,
|
|
"num_tokens": 8265193.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.442157030105591,
|
|
"epoch": 3.9843158464034616,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039212301366702153,
|
|
"loss": 5.1622,
|
|
"mean_token_accuracy": 0.2020845428109169,
|
|
"num_tokens": 8276343.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.522835111618042,
|
|
"epoch": 3.989724175229854,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039175657588657063,
|
|
"loss": 5.1435,
|
|
"mean_token_accuracy": 0.20641487389802932,
|
|
"num_tokens": 8288069.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.474055528640747,
|
|
"epoch": 3.9951325040562464,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003913897138124531,
|
|
"loss": 5.115,
|
|
"mean_token_accuracy": 0.20332870483398438,
|
|
"num_tokens": 8299382.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.483558548821343,
|
|
"epoch": 4.0,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 0.0003910224287778466,
|
|
"loss": 5.0273,
|
|
"mean_token_accuracy": 0.21283315287695992,
|
|
"num_tokens": 8307556.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.454630136489868,
|
|
"epoch": 4.005408328826393,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00039065472211746586,
|
|
"loss": 4.8721,
|
|
"mean_token_accuracy": 0.2291209504008293,
|
|
"num_tokens": 8317851.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.36405029296875,
|
|
"epoch": 4.010816657652786,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003902865951675576,
|
|
"loss": 4.769,
|
|
"mean_token_accuracy": 0.23732401728630065,
|
|
"num_tokens": 8328485.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.472476053237915,
|
|
"epoch": 4.016224986479178,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038991804926589616,
|
|
"loss": 4.7499,
|
|
"mean_token_accuracy": 0.2332100212574005,
|
|
"num_tokens": 8339625.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.460033845901489,
|
|
"epoch": 4.0216333153055706,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000389549085751778,
|
|
"loss": 4.7879,
|
|
"mean_token_accuracy": 0.23247666358947755,
|
|
"num_tokens": 8350667.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.474619913101196,
|
|
"epoch": 4.027041644131963,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038917970596601765,
|
|
"loss": 4.8838,
|
|
"mean_token_accuracy": 0.22240282595157623,
|
|
"num_tokens": 8362521.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.399797773361206,
|
|
"epoch": 4.032449972958356,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003888099112509419,
|
|
"loss": 4.903,
|
|
"mean_token_accuracy": 0.22196126580238343,
|
|
"num_tokens": 8372703.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.484219646453857,
|
|
"epoch": 4.037858301784748,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038843970295038564,
|
|
"loss": 4.8414,
|
|
"mean_token_accuracy": 0.2308829605579376,
|
|
"num_tokens": 8383345.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.441246795654297,
|
|
"epoch": 4.043266630611141,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00038806908240968666,
|
|
"loss": 4.7829,
|
|
"mean_token_accuracy": 0.2342671424150467,
|
|
"num_tokens": 8394056.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.3632416248321535,
|
|
"epoch": 4.048674959437534,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000387698050975681,
|
|
"loss": 4.821,
|
|
"mean_token_accuracy": 0.22387926280498505,
|
|
"num_tokens": 8403889.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.425527095794678,
|
|
"epoch": 4.054083288263927,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003873266099966976,
|
|
"loss": 4.8233,
|
|
"mean_token_accuracy": 0.22568415552377702,
|
|
"num_tokens": 8415169.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.4409631252288815,
|
|
"epoch": 4.059491617090319,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038695476082255395,
|
|
"loss": 4.8307,
|
|
"mean_token_accuracy": 0.22165250778198242,
|
|
"num_tokens": 8427056.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.383736801147461,
|
|
"epoch": 4.064899945916712,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003865825048045507,
|
|
"loss": 4.822,
|
|
"mean_token_accuracy": 0.22757732272148132,
|
|
"num_tokens": 8438900.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.409762287139893,
|
|
"epoch": 4.0703082747431045,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003862098432954672,
|
|
"loss": 4.7956,
|
|
"mean_token_accuracy": 0.22786156833171844,
|
|
"num_tokens": 8449037.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.359689807891845,
|
|
"epoch": 4.075716603569497,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003858367776495561,
|
|
"loss": 4.7775,
|
|
"mean_token_accuracy": 0.22574756145477295,
|
|
"num_tokens": 8459409.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.435139322280884,
|
|
"epoch": 4.081124932395889,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038546330922253886,
|
|
"loss": 4.8461,
|
|
"mean_token_accuracy": 0.22603178322315215,
|
|
"num_tokens": 8470190.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.420550966262818,
|
|
"epoch": 4.086533261222282,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003850894393716007,
|
|
"loss": 4.8761,
|
|
"mean_token_accuracy": 0.22616974711418153,
|
|
"num_tokens": 8481559.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.438857793807983,
|
|
"epoch": 4.091941590048675,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003847151694553855,
|
|
"loss": 4.876,
|
|
"mean_token_accuracy": 0.22165482491254807,
|
|
"num_tokens": 8492607.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.486331605911255,
|
|
"epoch": 4.097349918875068,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003843405008339908,
|
|
"loss": 4.9794,
|
|
"mean_token_accuracy": 0.21736517250537873,
|
|
"num_tokens": 8504954.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.5537192821502686,
|
|
"epoch": 4.10275824770146,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00038396543486896354,
|
|
"loss": 4.8743,
|
|
"mean_token_accuracy": 0.21999874264001845,
|
|
"num_tokens": 8516250.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.389536809921265,
|
|
"epoch": 4.108166576527853,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038358997292329406,
|
|
"loss": 4.8185,
|
|
"mean_token_accuracy": 0.2275618776679039,
|
|
"num_tokens": 8527220.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.4033979892730715,
|
|
"epoch": 4.113574905354246,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038321411636141214,
|
|
"loss": 4.9116,
|
|
"mean_token_accuracy": 0.22108134478330613,
|
|
"num_tokens": 8538037.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.460401010513306,
|
|
"epoch": 4.1189832341806385,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003828378665491811,
|
|
"loss": 4.8824,
|
|
"mean_token_accuracy": 0.22510204315185547,
|
|
"num_tokens": 8548684.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.40023512840271,
|
|
"epoch": 4.1243915630070305,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000382461224853894,
|
|
"loss": 4.8054,
|
|
"mean_token_accuracy": 0.22698983401060105,
|
|
"num_tokens": 8559756.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.3671165943145756,
|
|
"epoch": 4.129799891833423,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003820841926442674,
|
|
"loss": 4.823,
|
|
"mean_token_accuracy": 0.2331912711262703,
|
|
"num_tokens": 8570640.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.450835752487182,
|
|
"epoch": 4.135208220659816,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003817067712904372,
|
|
"loss": 4.8697,
|
|
"mean_token_accuracy": 0.22080679833889008,
|
|
"num_tokens": 8583127.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.3416649341583256,
|
|
"epoch": 4.140616549486209,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038132896216395355,
|
|
"loss": 4.8379,
|
|
"mean_token_accuracy": 0.22713606655597687,
|
|
"num_tokens": 8595213.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.432481861114502,
|
|
"epoch": 4.146024878312601,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003809507666377758,
|
|
"loss": 4.832,
|
|
"mean_token_accuracy": 0.2332100972533226,
|
|
"num_tokens": 8606057.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.484424257278443,
|
|
"epoch": 4.151433207138994,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038057218608626714,
|
|
"loss": 4.8935,
|
|
"mean_token_accuracy": 0.23745568245649337,
|
|
"num_tokens": 8617660.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.344848346710205,
|
|
"epoch": 4.156841535965387,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003801932218851903,
|
|
"loss": 4.8241,
|
|
"mean_token_accuracy": 0.23041671961545945,
|
|
"num_tokens": 8628595.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.462051725387573,
|
|
"epoch": 4.16224986479178,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000379813875411702,
|
|
"loss": 4.8818,
|
|
"mean_token_accuracy": 0.2281558856368065,
|
|
"num_tokens": 8639831.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.386359024047851,
|
|
"epoch": 4.167658193618172,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037943414804434826,
|
|
"loss": 4.825,
|
|
"mean_token_accuracy": 0.22668364346027375,
|
|
"num_tokens": 8650470.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.427944087982178,
|
|
"epoch": 4.173066522444564,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003790540411630592,
|
|
"loss": 4.9267,
|
|
"mean_token_accuracy": 0.22414447665214537,
|
|
"num_tokens": 8660969.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.379233932495117,
|
|
"epoch": 4.178474851270957,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037867355614914404,
|
|
"loss": 4.8006,
|
|
"mean_token_accuracy": 0.23358393609523773,
|
|
"num_tokens": 8670913.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.390183019638061,
|
|
"epoch": 4.18388318009735,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003782926943852864,
|
|
"loss": 4.9336,
|
|
"mean_token_accuracy": 0.21998182982206343,
|
|
"num_tokens": 8683028.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.519228839874268,
|
|
"epoch": 4.189291508923742,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037791145725553853,
|
|
"loss": 5.0252,
|
|
"mean_token_accuracy": 0.21582168936729432,
|
|
"num_tokens": 8696454.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.49918122291565,
|
|
"epoch": 4.194699837750135,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003775298461453172,
|
|
"loss": 4.8048,
|
|
"mean_token_accuracy": 0.23216895312070845,
|
|
"num_tokens": 8707112.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.343139934539795,
|
|
"epoch": 4.200108166576528,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000377147862441398,
|
|
"loss": 4.8195,
|
|
"mean_token_accuracy": 0.23519303649663925,
|
|
"num_tokens": 8717696.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.37954273223877,
|
|
"epoch": 4.205516495402921,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037676550753191075,
|
|
"loss": 4.7823,
|
|
"mean_token_accuracy": 0.23588618636131287,
|
|
"num_tokens": 8728571.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.461633634567261,
|
|
"epoch": 4.210924824229313,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003763827828063339,
|
|
"loss": 4.8645,
|
|
"mean_token_accuracy": 0.22649067640304565,
|
|
"num_tokens": 8739582.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.408328342437744,
|
|
"epoch": 4.2163331530557056,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037599968965549,
|
|
"loss": 4.8289,
|
|
"mean_token_accuracy": 0.2264792263507843,
|
|
"num_tokens": 8750527.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.346494817733765,
|
|
"epoch": 4.221741481882098,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003756162294715405,
|
|
"loss": 4.8416,
|
|
"mean_token_accuracy": 0.22670903354883193,
|
|
"num_tokens": 8763140.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.35790376663208,
|
|
"epoch": 4.227149810708491,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037523240364798063,
|
|
"loss": 4.8603,
|
|
"mean_token_accuracy": 0.2236696034669876,
|
|
"num_tokens": 8774328.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.428726053237915,
|
|
"epoch": 4.232558139534884,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037484821357963423,
|
|
"loss": 4.8852,
|
|
"mean_token_accuracy": 0.2263640746474266,
|
|
"num_tokens": 8785119.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.370996904373169,
|
|
"epoch": 4.237966468361276,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000374463660662649,
|
|
"loss": 4.8484,
|
|
"mean_token_accuracy": 0.21960169672966004,
|
|
"num_tokens": 8795833.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.448162269592285,
|
|
"epoch": 4.243374797187669,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000374078746294491,
|
|
"loss": 4.9381,
|
|
"mean_token_accuracy": 0.22060806900262833,
|
|
"num_tokens": 8807098.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.4237700462341305,
|
|
"epoch": 4.248783126014062,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00037369347187393996,
|
|
"loss": 4.7655,
|
|
"mean_token_accuracy": 0.23257531076669694,
|
|
"num_tokens": 8818096.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.4354085445404055,
|
|
"epoch": 4.254191454840455,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003733078388010842,
|
|
"loss": 4.9351,
|
|
"mean_token_accuracy": 0.2113041177392006,
|
|
"num_tokens": 8829451.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.375819683074951,
|
|
"epoch": 4.259599783666847,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003729218484773149,
|
|
"loss": 4.8811,
|
|
"mean_token_accuracy": 0.2286873310804367,
|
|
"num_tokens": 8840099.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.375582981109619,
|
|
"epoch": 4.2650081124932395,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003725355023053221,
|
|
"loss": 4.7751,
|
|
"mean_token_accuracy": 0.2340745896100998,
|
|
"num_tokens": 8851269.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.375064754486084,
|
|
"epoch": 4.270416441319632,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003721488016890884,
|
|
"loss": 4.809,
|
|
"mean_token_accuracy": 0.23112039119005204,
|
|
"num_tokens": 8862871.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.411102676391602,
|
|
"epoch": 4.275824770146025,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037176174803388496,
|
|
"loss": 4.8481,
|
|
"mean_token_accuracy": 0.23287083804607392,
|
|
"num_tokens": 8874328.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.40824704170227,
|
|
"epoch": 4.281233098972417,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037137434274626545,
|
|
"loss": 4.8773,
|
|
"mean_token_accuracy": 0.22311595231294631,
|
|
"num_tokens": 8885133.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.373760604858399,
|
|
"epoch": 4.28664142779881,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003709865872340618,
|
|
"loss": 4.8717,
|
|
"mean_token_accuracy": 0.2259731188416481,
|
|
"num_tokens": 8896479.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.4292055606842045,
|
|
"epoch": 4.292049756625203,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003705984829063782,
|
|
"loss": 4.7924,
|
|
"mean_token_accuracy": 0.22853993624448776,
|
|
"num_tokens": 8907966.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.341626214981079,
|
|
"epoch": 4.297458085451596,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003702100311735869,
|
|
"loss": 4.847,
|
|
"mean_token_accuracy": 0.22714053839445114,
|
|
"num_tokens": 8919042.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.356872034072876,
|
|
"epoch": 4.302866414277988,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036982123344732223,
|
|
"loss": 4.8559,
|
|
"mean_token_accuracy": 0.22222124338150023,
|
|
"num_tokens": 8930456.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.407025480270386,
|
|
"epoch": 4.308274743104381,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00036943209114047613,
|
|
"loss": 4.9087,
|
|
"mean_token_accuracy": 0.22985780835151673,
|
|
"num_tokens": 8941473.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.383913326263428,
|
|
"epoch": 4.3136830719307735,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003690426056671926,
|
|
"loss": 4.9478,
|
|
"mean_token_accuracy": 0.21982176452875138,
|
|
"num_tokens": 8952927.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.414112615585327,
|
|
"epoch": 4.319091400757166,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003686527784428627,
|
|
"loss": 4.8944,
|
|
"mean_token_accuracy": 0.22251450568437575,
|
|
"num_tokens": 8963831.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.426124906539917,
|
|
"epoch": 4.324499729583558,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036826261088411955,
|
|
"loss": 4.8524,
|
|
"mean_token_accuracy": 0.21882633119821548,
|
|
"num_tokens": 8977147.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.324499729583558,
|
|
"eval_entropy": 5.2865768519314855,
|
|
"eval_loss": 5.837026119232178,
|
|
"eval_mean_token_accuracy": 0.1801806160502813,
|
|
"eval_num_tokens": 8977147.0,
|
|
"eval_runtime": 2.1389,
|
|
"eval_samples_per_second": 1643.395,
|
|
"eval_steps_per_second": 205.717,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2079694613790720.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|