2046 lines
54 KiB
JSON
2046 lines
54 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0811249323958896,
|
||
|
|
"eval_steps": 1000,
|
||
|
|
"global_step": 1000,
|
||
|
|
"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
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"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": 520419248455680.0,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|