4035 lines
109 KiB
JSON
4035 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.046128652812694605,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742666625976563,
|
|
"epoch": 0.0001153216320317365,
|
|
"grad_norm": 7.125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.4073,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 3293.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742696571350098,
|
|
"epoch": 0.000230643264063473,
|
|
"grad_norm": 7.75,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.3973,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 6743.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742678165435791,
|
|
"epoch": 0.0003459648960952095,
|
|
"grad_norm": 6.6875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.399,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10261.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742677593231202,
|
|
"epoch": 0.000461286528126946,
|
|
"grad_norm": 7.15625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.367,
|
|
"mean_token_accuracy": 0.0006192009896039963,
|
|
"num_tokens": 14153.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742670631408691,
|
|
"epoch": 0.0005766081601586826,
|
|
"grad_norm": 6.25,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.2426,
|
|
"mean_token_accuracy": 0.011197958327829837,
|
|
"num_tokens": 17774.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742291259765626,
|
|
"epoch": 0.000691929792190419,
|
|
"grad_norm": 5.28125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.1259,
|
|
"mean_token_accuracy": 0.0403778588399291,
|
|
"num_tokens": 21521.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741307258605957,
|
|
"epoch": 0.0008072514242221556,
|
|
"grad_norm": 4.71875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.9355,
|
|
"mean_token_accuracy": 0.05537886433303356,
|
|
"num_tokens": 24880.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.738965320587159,
|
|
"epoch": 0.000922573056253892,
|
|
"grad_norm": 3.9375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.7418,
|
|
"mean_token_accuracy": 0.05559282712638378,
|
|
"num_tokens": 28167.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.73506326675415,
|
|
"epoch": 0.0010378946882856287,
|
|
"grad_norm": 3.03125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6835,
|
|
"mean_token_accuracy": 0.055228251963853836,
|
|
"num_tokens": 31745.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.731123352050782,
|
|
"epoch": 0.0011532163203173652,
|
|
"grad_norm": 3.21875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5955,
|
|
"mean_token_accuracy": 0.06102724336087704,
|
|
"num_tokens": 35270.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.729140281677246,
|
|
"epoch": 0.0012685379523491016,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5451,
|
|
"mean_token_accuracy": 0.05829254724085331,
|
|
"num_tokens": 38943.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.726730442047119,
|
|
"epoch": 0.001383859584380838,
|
|
"grad_norm": 2.765625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.3612,
|
|
"mean_token_accuracy": 0.07044943682849407,
|
|
"num_tokens": 42299.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.723068046569825,
|
|
"epoch": 0.0014991812164125747,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3629,
|
|
"mean_token_accuracy": 0.06617457047104836,
|
|
"num_tokens": 45661.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.719834232330323,
|
|
"epoch": 0.0016145028484443112,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.4436,
|
|
"mean_token_accuracy": 0.05358973033726215,
|
|
"num_tokens": 49683.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.716299343109132,
|
|
"epoch": 0.0017298244804760477,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2066,
|
|
"mean_token_accuracy": 0.07446158789098263,
|
|
"num_tokens": 52752.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.702353477478027,
|
|
"epoch": 0.001845146112507784,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1991,
|
|
"mean_token_accuracy": 0.06319843120872974,
|
|
"num_tokens": 56367.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.687633800506593,
|
|
"epoch": 0.0019604677445395208,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1229,
|
|
"mean_token_accuracy": 0.06458256356418132,
|
|
"num_tokens": 59967.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.677945995330811,
|
|
"epoch": 0.0020757893765712574,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.1085,
|
|
"mean_token_accuracy": 0.06282291375100613,
|
|
"num_tokens": 63634.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.650175952911377,
|
|
"epoch": 0.0021911110086029937,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8704,
|
|
"mean_token_accuracy": 0.07925741896033286,
|
|
"num_tokens": 66883.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.59815263748169,
|
|
"epoch": 0.0023064326406347303,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8919,
|
|
"mean_token_accuracy": 0.0715696856379509,
|
|
"num_tokens": 70491.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.569210147857666,
|
|
"epoch": 0.0024217542726664666,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.7607,
|
|
"mean_token_accuracy": 0.06703307442367076,
|
|
"num_tokens": 73852.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.54236125946045,
|
|
"epoch": 0.0025370759046982033,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.7055,
|
|
"mean_token_accuracy": 0.07926666587591172,
|
|
"num_tokens": 77442.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.49743299484253,
|
|
"epoch": 0.00265239753672994,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.5989,
|
|
"mean_token_accuracy": 0.07457204945385457,
|
|
"num_tokens": 80871.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.424223518371582,
|
|
"epoch": 0.002767719168761676,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.494,
|
|
"mean_token_accuracy": 0.07471656203269958,
|
|
"num_tokens": 84444.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.37389669418335,
|
|
"epoch": 0.002883040800793413,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.4199,
|
|
"mean_token_accuracy": 0.07393867634236813,
|
|
"num_tokens": 87939.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.249776554107665,
|
|
"epoch": 0.0029983624328251495,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.1924,
|
|
"mean_token_accuracy": 0.07107977494597435,
|
|
"num_tokens": 91189.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.182583045959472,
|
|
"epoch": 0.0031136840648568857,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.1534,
|
|
"mean_token_accuracy": 0.07604897432029248,
|
|
"num_tokens": 94777.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.081933212280273,
|
|
"epoch": 0.0032290056968886224,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.1665,
|
|
"mean_token_accuracy": 0.06793576180934906,
|
|
"num_tokens": 98581.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.924797248840331,
|
|
"epoch": 0.003344327328920359,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.052,
|
|
"mean_token_accuracy": 0.06972302682697773,
|
|
"num_tokens": 102333.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.775594520568848,
|
|
"epoch": 0.0034596489609520953,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.9256,
|
|
"mean_token_accuracy": 0.07209130860865116,
|
|
"num_tokens": 106267.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.597217845916749,
|
|
"epoch": 0.003574970592983832,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.8077,
|
|
"mean_token_accuracy": 0.07151266261935234,
|
|
"num_tokens": 109989.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.383270835876464,
|
|
"epoch": 0.003690292225015568,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.7181,
|
|
"mean_token_accuracy": 0.07966535687446594,
|
|
"num_tokens": 113378.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.220961475372315,
|
|
"epoch": 0.003805613857047305,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.594,
|
|
"mean_token_accuracy": 0.08258548378944397,
|
|
"num_tokens": 116736.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.071347045898438,
|
|
"epoch": 0.0039209354890790415,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.5045,
|
|
"mean_token_accuracy": 0.08615332692861558,
|
|
"num_tokens": 120369.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.765302276611328,
|
|
"epoch": 0.004036257121110778,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.4619,
|
|
"mean_token_accuracy": 0.08963219076395035,
|
|
"num_tokens": 124003.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.61657485961914,
|
|
"epoch": 0.004151578753142515,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4401,
|
|
"mean_token_accuracy": 0.08809128813445569,
|
|
"num_tokens": 127611.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.546231937408447,
|
|
"epoch": 0.004266900385174251,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4141,
|
|
"mean_token_accuracy": 0.08483765870332718,
|
|
"num_tokens": 130837.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.369953155517578,
|
|
"epoch": 0.004382222017205987,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4516,
|
|
"mean_token_accuracy": 0.08280227519571781,
|
|
"num_tokens": 134801.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.383075523376466,
|
|
"epoch": 0.004497543649237724,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4581,
|
|
"mean_token_accuracy": 0.07945475727319717,
|
|
"num_tokens": 138479.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.266378498077392,
|
|
"epoch": 0.004612865281269461,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.382,
|
|
"mean_token_accuracy": 0.08485323041677476,
|
|
"num_tokens": 142536.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.269253635406494,
|
|
"epoch": 0.004728186913301197,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4255,
|
|
"mean_token_accuracy": 0.07670572884380818,
|
|
"num_tokens": 146633.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.14164171218872,
|
|
"epoch": 0.004843508545332933,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2377,
|
|
"mean_token_accuracy": 0.09400919899344444,
|
|
"num_tokens": 150070.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.126185989379882,
|
|
"epoch": 0.00495883017736467,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3864,
|
|
"mean_token_accuracy": 0.08119251355528831,
|
|
"num_tokens": 153698.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.100548648834229,
|
|
"epoch": 0.0050741518093964065,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2826,
|
|
"mean_token_accuracy": 0.09440358653664589,
|
|
"num_tokens": 157381.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.097941875457764,
|
|
"epoch": 0.005189473441428143,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.267,
|
|
"mean_token_accuracy": 0.0852508783340454,
|
|
"num_tokens": 160862.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.997360134124756,
|
|
"epoch": 0.00530479507345988,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2505,
|
|
"mean_token_accuracy": 0.08932770863175392,
|
|
"num_tokens": 164231.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.9703351020812985,
|
|
"epoch": 0.0054201167054916165,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.3237,
|
|
"mean_token_accuracy": 0.08765647709369659,
|
|
"num_tokens": 168351.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.950035905838012,
|
|
"epoch": 0.005535438337523352,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1315,
|
|
"mean_token_accuracy": 0.0994307741522789,
|
|
"num_tokens": 171645.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.882783460617065,
|
|
"epoch": 0.005650759969555089,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.163,
|
|
"mean_token_accuracy": 0.09629305228590965,
|
|
"num_tokens": 175279.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.845568895339966,
|
|
"epoch": 0.005766081601586826,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2429,
|
|
"mean_token_accuracy": 0.0953019991517067,
|
|
"num_tokens": 179111.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.9421182632446286,
|
|
"epoch": 0.005881403233618562,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1945,
|
|
"mean_token_accuracy": 0.09875313565135002,
|
|
"num_tokens": 182369.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.8873544216156,
|
|
"epoch": 0.005996724865650299,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1627,
|
|
"mean_token_accuracy": 0.10178187415003777,
|
|
"num_tokens": 185671.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.945286512374878,
|
|
"epoch": 0.006112046497682035,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.089,
|
|
"mean_token_accuracy": 0.09403523430228233,
|
|
"num_tokens": 189295.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.735852289199829,
|
|
"epoch": 0.0062273681297137715,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0907,
|
|
"mean_token_accuracy": 0.0964249037206173,
|
|
"num_tokens": 192791.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.727569532394409,
|
|
"epoch": 0.006342689761745508,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1104,
|
|
"mean_token_accuracy": 0.09585446193814277,
|
|
"num_tokens": 196550.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.847873210906982,
|
|
"epoch": 0.006458011393777245,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1018,
|
|
"mean_token_accuracy": 0.09330938830971718,
|
|
"num_tokens": 200274.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.683241939544677,
|
|
"epoch": 0.0065733330258089815,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0522,
|
|
"mean_token_accuracy": 0.10625462830066681,
|
|
"num_tokens": 203748.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.778611564636231,
|
|
"epoch": 0.006688654657840718,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0655,
|
|
"mean_token_accuracy": 0.10183993503451347,
|
|
"num_tokens": 207203.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.754481458663941,
|
|
"epoch": 0.006803976289872454,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.2431,
|
|
"mean_token_accuracy": 0.08833375945687294,
|
|
"num_tokens": 210761.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.801531457901001,
|
|
"epoch": 0.006919297921904191,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.1216,
|
|
"mean_token_accuracy": 0.10129701644182205,
|
|
"num_tokens": 214362.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.7394914627075195,
|
|
"epoch": 0.007034619553935927,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.1811,
|
|
"mean_token_accuracy": 0.09345738962292671,
|
|
"num_tokens": 217938.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.6699385166168215,
|
|
"epoch": 0.007149941185967664,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.089,
|
|
"mean_token_accuracy": 0.1043456457555294,
|
|
"num_tokens": 221519.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.771025037765503,
|
|
"epoch": 0.007265262817999401,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0161,
|
|
"mean_token_accuracy": 0.10540573447942733,
|
|
"num_tokens": 224918.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.6741640090942385,
|
|
"epoch": 0.007380584450031136,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0495,
|
|
"mean_token_accuracy": 0.10322595313191414,
|
|
"num_tokens": 228334.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.7113241195678714,
|
|
"epoch": 0.007495906082062873,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.0706,
|
|
"mean_token_accuracy": 0.10638125836849213,
|
|
"num_tokens": 231826.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.706802654266357,
|
|
"epoch": 0.00761122771409461,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.0128,
|
|
"mean_token_accuracy": 0.10426252707839012,
|
|
"num_tokens": 235519.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.707431745529175,
|
|
"epoch": 0.007726549346126346,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.2316,
|
|
"mean_token_accuracy": 0.09854352325201035,
|
|
"num_tokens": 239451.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.5462418556213375,
|
|
"epoch": 0.007841870978158083,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.1006,
|
|
"mean_token_accuracy": 0.09770274758338929,
|
|
"num_tokens": 243193.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.8119916915893555,
|
|
"epoch": 0.007957192610189819,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1326,
|
|
"mean_token_accuracy": 0.10739191100001336,
|
|
"num_tokens": 246859.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.6009765625,
|
|
"epoch": 0.008072514242221556,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.0056,
|
|
"mean_token_accuracy": 0.10416570231318474,
|
|
"num_tokens": 250345.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.648880338668823,
|
|
"epoch": 0.008187835874253292,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0733,
|
|
"mean_token_accuracy": 0.0952294260263443,
|
|
"num_tokens": 254201.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.7096099853515625,
|
|
"epoch": 0.00830315750628503,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.0603,
|
|
"mean_token_accuracy": 0.10599224865436555,
|
|
"num_tokens": 258073.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.540476942062378,
|
|
"epoch": 0.008418479138316766,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.076,
|
|
"mean_token_accuracy": 0.10500729605555534,
|
|
"num_tokens": 262221.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.630076503753662,
|
|
"epoch": 0.008533800770348501,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9305,
|
|
"mean_token_accuracy": 0.11372119784355164,
|
|
"num_tokens": 265920.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.52258267402649,
|
|
"epoch": 0.008649122402380239,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.9444,
|
|
"mean_token_accuracy": 0.10598652064800262,
|
|
"num_tokens": 269617.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.6615653991699215,
|
|
"epoch": 0.008764444034411975,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1068,
|
|
"mean_token_accuracy": 0.11471877321600914,
|
|
"num_tokens": 273482.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.565646600723267,
|
|
"epoch": 0.008879765666443712,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9744,
|
|
"mean_token_accuracy": 0.10696744173765182,
|
|
"num_tokens": 276877.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.5632483005523685,
|
|
"epoch": 0.008995087298475448,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8722,
|
|
"mean_token_accuracy": 0.11907225847244263,
|
|
"num_tokens": 280342.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.529973316192627,
|
|
"epoch": 0.009110408930507184,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9349,
|
|
"mean_token_accuracy": 0.11691308841109276,
|
|
"num_tokens": 283740.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.601187419891358,
|
|
"epoch": 0.009225730562538921,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9257,
|
|
"mean_token_accuracy": 0.10361649692058564,
|
|
"num_tokens": 287534.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.506116008758545,
|
|
"epoch": 0.009341052194570657,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0118,
|
|
"mean_token_accuracy": 0.10809247344732284,
|
|
"num_tokens": 290847.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.594826316833496,
|
|
"epoch": 0.009456373826602395,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.876,
|
|
"mean_token_accuracy": 0.11550146490335464,
|
|
"num_tokens": 294317.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.40671329498291,
|
|
"epoch": 0.00957169545863413,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8185,
|
|
"mean_token_accuracy": 0.11729320585727691,
|
|
"num_tokens": 297899.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.4021570682525635,
|
|
"epoch": 0.009687017090665866,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8494,
|
|
"mean_token_accuracy": 0.12330111861228943,
|
|
"num_tokens": 301076.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.417027854919434,
|
|
"epoch": 0.009802338722697604,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8474,
|
|
"mean_token_accuracy": 0.11745313182473183,
|
|
"num_tokens": 304848.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.533647346496582,
|
|
"epoch": 0.00991766035472934,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.9147,
|
|
"mean_token_accuracy": 0.11939889490604401,
|
|
"num_tokens": 308223.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.437449264526367,
|
|
"epoch": 0.010032981986761077,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.0516,
|
|
"mean_token_accuracy": 0.11606085002422332,
|
|
"num_tokens": 311755.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.477462291717529,
|
|
"epoch": 0.010148303618792813,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9156,
|
|
"mean_token_accuracy": 0.1153307244181633,
|
|
"num_tokens": 315362.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.674656772613526,
|
|
"epoch": 0.010263625250824549,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.1171,
|
|
"mean_token_accuracy": 0.10499726831912995,
|
|
"num_tokens": 319705.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.527849102020264,
|
|
"epoch": 0.010378946882856286,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.954,
|
|
"mean_token_accuracy": 0.11489417552947997,
|
|
"num_tokens": 323371.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.6116527080535885,
|
|
"epoch": 0.010494268514888022,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.0286,
|
|
"mean_token_accuracy": 0.11207885667681694,
|
|
"num_tokens": 326846.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.374996709823608,
|
|
"epoch": 0.01060959014691976,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7879,
|
|
"mean_token_accuracy": 0.13456487134099007,
|
|
"num_tokens": 330305.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.3307215690612795,
|
|
"epoch": 0.010724911778951495,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.786,
|
|
"mean_token_accuracy": 0.12738576233386995,
|
|
"num_tokens": 333532.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.416831588745117,
|
|
"epoch": 0.010840233410983233,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8221,
|
|
"mean_token_accuracy": 0.1296285279095173,
|
|
"num_tokens": 336892.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.4116466522216795,
|
|
"epoch": 0.010955555043014969,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7774,
|
|
"mean_token_accuracy": 0.1254955343902111,
|
|
"num_tokens": 340372.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.6591136932373045,
|
|
"epoch": 0.011070876675046705,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9614,
|
|
"mean_token_accuracy": 0.11712158173322677,
|
|
"num_tokens": 344204.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.256807279586792,
|
|
"epoch": 0.011186198307078442,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8478,
|
|
"mean_token_accuracy": 0.12824892923235892,
|
|
"num_tokens": 347962.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.46787543296814,
|
|
"epoch": 0.011301519939110178,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.8481,
|
|
"mean_token_accuracy": 0.12688155770301818,
|
|
"num_tokens": 351559.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.511302423477173,
|
|
"epoch": 0.011416841571141916,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.9071,
|
|
"mean_token_accuracy": 0.12064936980605126,
|
|
"num_tokens": 354878.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.546496772766114,
|
|
"epoch": 0.011532163203173651,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.9038,
|
|
"mean_token_accuracy": 0.12078217267990113,
|
|
"num_tokens": 358998.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.302989673614502,
|
|
"epoch": 0.011647484835205387,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8774,
|
|
"mean_token_accuracy": 0.12374431192874909,
|
|
"num_tokens": 362372.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.404789018630981,
|
|
"epoch": 0.011762806467237125,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7698,
|
|
"mean_token_accuracy": 0.12083231210708618,
|
|
"num_tokens": 366142.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.3332984924316404,
|
|
"epoch": 0.01187812809926886,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7953,
|
|
"mean_token_accuracy": 0.13245533555746078,
|
|
"num_tokens": 369761.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.313057947158813,
|
|
"epoch": 0.011993449731300598,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.7312,
|
|
"mean_token_accuracy": 0.12337852418422698,
|
|
"num_tokens": 373145.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.36409969329834,
|
|
"epoch": 0.012108771363332334,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.727,
|
|
"mean_token_accuracy": 0.12250038459897042,
|
|
"num_tokens": 376447.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.386349010467529,
|
|
"epoch": 0.01222409299536407,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.846,
|
|
"mean_token_accuracy": 0.1162750244140625,
|
|
"num_tokens": 379928.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.233669900894165,
|
|
"epoch": 0.012339414627395807,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8069,
|
|
"mean_token_accuracy": 0.11781511679291726,
|
|
"num_tokens": 383443.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.233025312423706,
|
|
"epoch": 0.012454736259427543,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6479,
|
|
"mean_token_accuracy": 0.13859662935137748,
|
|
"num_tokens": 386691.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.372886323928833,
|
|
"epoch": 0.01257005789145928,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.9235,
|
|
"mean_token_accuracy": 0.12282031551003456,
|
|
"num_tokens": 390600.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.353907585144043,
|
|
"epoch": 0.012685379523491016,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7458,
|
|
"mean_token_accuracy": 0.12587565630674363,
|
|
"num_tokens": 394148.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.198177480697632,
|
|
"epoch": 0.012800701155522754,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6411,
|
|
"mean_token_accuracy": 0.137128234654665,
|
|
"num_tokens": 397524.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.252558660507202,
|
|
"epoch": 0.01291602278755449,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8175,
|
|
"mean_token_accuracy": 0.12427356764674187,
|
|
"num_tokens": 401366.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.290296602249145,
|
|
"epoch": 0.013031344419586225,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7517,
|
|
"mean_token_accuracy": 0.13000626116991043,
|
|
"num_tokens": 404952.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.274889421463013,
|
|
"epoch": 0.013146666051617963,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.719,
|
|
"mean_token_accuracy": 0.130663925409317,
|
|
"num_tokens": 408678.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.204096937179566,
|
|
"epoch": 0.013261987683649699,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7706,
|
|
"mean_token_accuracy": 0.13187873736023903,
|
|
"num_tokens": 411964.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.314283561706543,
|
|
"epoch": 0.013377309315681436,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.8049,
|
|
"mean_token_accuracy": 0.1299607813358307,
|
|
"num_tokens": 415533.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.283550453186035,
|
|
"epoch": 0.013492630947713172,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7803,
|
|
"mean_token_accuracy": 0.12142688482999801,
|
|
"num_tokens": 419052.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.202226257324218,
|
|
"epoch": 0.013607952579744908,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8542,
|
|
"mean_token_accuracy": 0.12060836181044579,
|
|
"num_tokens": 422775.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.199810123443603,
|
|
"epoch": 0.013723274211776645,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7513,
|
|
"mean_token_accuracy": 0.13247860968112946,
|
|
"num_tokens": 426363.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.114086627960205,
|
|
"epoch": 0.013838595843808381,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7615,
|
|
"mean_token_accuracy": 0.1301987074315548,
|
|
"num_tokens": 429815.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.156881427764892,
|
|
"epoch": 0.013953917475840119,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6489,
|
|
"mean_token_accuracy": 0.1388367809355259,
|
|
"num_tokens": 433128.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.0776512145996096,
|
|
"epoch": 0.014069239107871855,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7137,
|
|
"mean_token_accuracy": 0.12970117330551148,
|
|
"num_tokens": 436806.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.2287922382354735,
|
|
"epoch": 0.01418456073990359,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.733,
|
|
"mean_token_accuracy": 0.1362753003835678,
|
|
"num_tokens": 440601.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.313516759872437,
|
|
"epoch": 0.014299882371935328,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7598,
|
|
"mean_token_accuracy": 0.12404989525675773,
|
|
"num_tokens": 444192.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.119010877609253,
|
|
"epoch": 0.014415204003967064,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7826,
|
|
"mean_token_accuracy": 0.1288778491318226,
|
|
"num_tokens": 447680.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.195054006576538,
|
|
"epoch": 0.014530525635998801,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7608,
|
|
"mean_token_accuracy": 0.13093777373433113,
|
|
"num_tokens": 451469.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.146825218200684,
|
|
"epoch": 0.014645847268030537,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.58,
|
|
"mean_token_accuracy": 0.14002493098378183,
|
|
"num_tokens": 455099.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.225515699386596,
|
|
"epoch": 0.014761168900062273,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7085,
|
|
"mean_token_accuracy": 0.12969221770763398,
|
|
"num_tokens": 459026.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.082807302474976,
|
|
"epoch": 0.01487649053209401,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6302,
|
|
"mean_token_accuracy": 0.12743497341871263,
|
|
"num_tokens": 462901.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.1721337795257565,
|
|
"epoch": 0.014991812164125746,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7815,
|
|
"mean_token_accuracy": 0.1328444227576256,
|
|
"num_tokens": 466424.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.9974264144897464,
|
|
"epoch": 0.015107133796157484,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6073,
|
|
"mean_token_accuracy": 0.1396956041455269,
|
|
"num_tokens": 470063.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.132135534286499,
|
|
"epoch": 0.01522245542818922,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5109,
|
|
"mean_token_accuracy": 0.1423234686255455,
|
|
"num_tokens": 473513.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.0790338039398195,
|
|
"epoch": 0.015337777060220957,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5865,
|
|
"mean_token_accuracy": 0.13629688918590546,
|
|
"num_tokens": 476973.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.065616226196289,
|
|
"epoch": 0.015453098692252693,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.62,
|
|
"mean_token_accuracy": 0.13478031530976295,
|
|
"num_tokens": 480331.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.148045635223388,
|
|
"epoch": 0.015568420324284429,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6034,
|
|
"mean_token_accuracy": 0.12301738932728767,
|
|
"num_tokens": 483647.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.08280873298645,
|
|
"epoch": 0.015683741956316166,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.6582,
|
|
"mean_token_accuracy": 0.13532254993915557,
|
|
"num_tokens": 487463.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.0641885757446286,
|
|
"epoch": 0.015799063588347904,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.5872,
|
|
"mean_token_accuracy": 0.13603476136922837,
|
|
"num_tokens": 490891.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.162705183029175,
|
|
"epoch": 0.015914385220379638,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.8325,
|
|
"mean_token_accuracy": 0.12991492375731467,
|
|
"num_tokens": 494576.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.883268213272094,
|
|
"epoch": 0.016029706852411375,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5904,
|
|
"mean_token_accuracy": 0.13417865112423896,
|
|
"num_tokens": 498316.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.228239345550537,
|
|
"epoch": 0.016145028484443113,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6597,
|
|
"mean_token_accuracy": 0.1372683197259903,
|
|
"num_tokens": 501746.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.889809036254883,
|
|
"epoch": 0.016260350116474847,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5728,
|
|
"mean_token_accuracy": 0.1426199421286583,
|
|
"num_tokens": 505188.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.068775320053101,
|
|
"epoch": 0.016375671748506584,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5461,
|
|
"mean_token_accuracy": 0.14004089310765266,
|
|
"num_tokens": 509342.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.0989960670471195,
|
|
"epoch": 0.016490993380538322,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.639,
|
|
"mean_token_accuracy": 0.1397839866578579,
|
|
"num_tokens": 513284.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.900527143478394,
|
|
"epoch": 0.01660631501257006,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6493,
|
|
"mean_token_accuracy": 0.13304078131914138,
|
|
"num_tokens": 517032.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.152545356750489,
|
|
"epoch": 0.016721636644601794,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5704,
|
|
"mean_token_accuracy": 0.14788509383797646,
|
|
"num_tokens": 520629.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.90955548286438,
|
|
"epoch": 0.01683695827663353,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4758,
|
|
"mean_token_accuracy": 0.14352560713887214,
|
|
"num_tokens": 524281.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.128781986236572,
|
|
"epoch": 0.01695227990866527,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5291,
|
|
"mean_token_accuracy": 0.14776214808225632,
|
|
"num_tokens": 527726.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.897321939468384,
|
|
"epoch": 0.017067601540697003,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5776,
|
|
"mean_token_accuracy": 0.14073135927319527,
|
|
"num_tokens": 531017.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.105098581314087,
|
|
"epoch": 0.01718292317272874,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.6708,
|
|
"mean_token_accuracy": 0.1416097842156887,
|
|
"num_tokens": 535161.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.99523434638977,
|
|
"epoch": 0.017298244804760478,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5859,
|
|
"mean_token_accuracy": 0.14246859773993492,
|
|
"num_tokens": 538916.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.910310506820679,
|
|
"epoch": 0.017413566436792212,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5398,
|
|
"mean_token_accuracy": 0.1352516621351242,
|
|
"num_tokens": 542857.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.935078048706055,
|
|
"epoch": 0.01752888806882395,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6003,
|
|
"mean_token_accuracy": 0.12804893776774406,
|
|
"num_tokens": 546678.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.066189956665039,
|
|
"epoch": 0.017644209700855687,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5657,
|
|
"mean_token_accuracy": 0.1297842025756836,
|
|
"num_tokens": 550185.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.835250377655029,
|
|
"epoch": 0.017759531332887424,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5668,
|
|
"mean_token_accuracy": 0.1325291983783245,
|
|
"num_tokens": 553769.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.0729241371154785,
|
|
"epoch": 0.01787485296491916,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4777,
|
|
"mean_token_accuracy": 0.15337770953774452,
|
|
"num_tokens": 557235.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.63848090171814,
|
|
"epoch": 0.017990174596950896,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4892,
|
|
"mean_token_accuracy": 0.14428077042102813,
|
|
"num_tokens": 560646.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.8471729278564455,
|
|
"epoch": 0.018105496228982634,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4017,
|
|
"mean_token_accuracy": 0.1493760883808136,
|
|
"num_tokens": 563945.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.9516904830932615,
|
|
"epoch": 0.018220817861014368,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.6442,
|
|
"mean_token_accuracy": 0.13188150003552437,
|
|
"num_tokens": 568010.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.9877588748931885,
|
|
"epoch": 0.018336139493046105,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5412,
|
|
"mean_token_accuracy": 0.13552800193428993,
|
|
"num_tokens": 571643.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.975119495391846,
|
|
"epoch": 0.018451461125077843,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4857,
|
|
"mean_token_accuracy": 0.1437096744775772,
|
|
"num_tokens": 575128.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.733108949661255,
|
|
"epoch": 0.01856678275710958,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.477,
|
|
"mean_token_accuracy": 0.1386350005865097,
|
|
"num_tokens": 578471.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.899811506271362,
|
|
"epoch": 0.018682104389141314,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4878,
|
|
"mean_token_accuracy": 0.14222341030836105,
|
|
"num_tokens": 581865.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.829076910018921,
|
|
"epoch": 0.018797426021173052,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5079,
|
|
"mean_token_accuracy": 0.1328883461654186,
|
|
"num_tokens": 585277.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.954855728149414,
|
|
"epoch": 0.01891274765320479,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.6236,
|
|
"mean_token_accuracy": 0.12898606956005096,
|
|
"num_tokens": 589313.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.847860956192017,
|
|
"epoch": 0.019028069285236524,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4541,
|
|
"mean_token_accuracy": 0.14378139823675157,
|
|
"num_tokens": 592907.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.787905931472778,
|
|
"epoch": 0.01914339091726826,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4968,
|
|
"mean_token_accuracy": 0.14837865382432938,
|
|
"num_tokens": 596406.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.846011734008789,
|
|
"epoch": 0.0192587125493,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4548,
|
|
"mean_token_accuracy": 0.15046936571598052,
|
|
"num_tokens": 599978.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.7639833927154545,
|
|
"epoch": 0.019374034181331733,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4207,
|
|
"mean_token_accuracy": 0.14280462637543678,
|
|
"num_tokens": 603466.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.822991180419922,
|
|
"epoch": 0.01948935581336347,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4501,
|
|
"mean_token_accuracy": 0.14506957828998565,
|
|
"num_tokens": 606976.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.856816244125366,
|
|
"epoch": 0.019604677445395208,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5081,
|
|
"mean_token_accuracy": 0.1306887984275818,
|
|
"num_tokens": 610475.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.865983200073242,
|
|
"epoch": 0.019719999077426945,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5345,
|
|
"mean_token_accuracy": 0.13522887825965882,
|
|
"num_tokens": 614416.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.83031644821167,
|
|
"epoch": 0.01983532070945868,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4483,
|
|
"mean_token_accuracy": 0.14981459826231003,
|
|
"num_tokens": 618398.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.79636173248291,
|
|
"epoch": 0.019950642341490417,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3795,
|
|
"mean_token_accuracy": 0.14684337005019188,
|
|
"num_tokens": 622029.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.708897495269776,
|
|
"epoch": 0.020065963973522154,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.447,
|
|
"mean_token_accuracy": 0.14757393524050713,
|
|
"num_tokens": 625438.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.778446292877197,
|
|
"epoch": 0.02018128560555389,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3749,
|
|
"mean_token_accuracy": 0.14992502555251122,
|
|
"num_tokens": 628936.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.848304557800293,
|
|
"epoch": 0.020296607237585626,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.428,
|
|
"mean_token_accuracy": 0.14919070899486542,
|
|
"num_tokens": 632190.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.826952219009399,
|
|
"epoch": 0.020411928869617364,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.5018,
|
|
"mean_token_accuracy": 0.13902354538440703,
|
|
"num_tokens": 635786.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.7481951236724855,
|
|
"epoch": 0.020527250501649098,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4143,
|
|
"mean_token_accuracy": 0.1464080534875393,
|
|
"num_tokens": 639423.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.848554182052612,
|
|
"epoch": 0.020642572133680835,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.3874,
|
|
"mean_token_accuracy": 0.15259440019726753,
|
|
"num_tokens": 642880.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.596447849273682,
|
|
"epoch": 0.020757893765712573,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4039,
|
|
"mean_token_accuracy": 0.14992632791399957,
|
|
"num_tokens": 646417.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.714345693588257,
|
|
"epoch": 0.02087321539774431,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4423,
|
|
"mean_token_accuracy": 0.1356890842318535,
|
|
"num_tokens": 650088.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.844691753387451,
|
|
"epoch": 0.020988537029776044,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5516,
|
|
"mean_token_accuracy": 0.13994285166263581,
|
|
"num_tokens": 653648.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.726529121398926,
|
|
"epoch": 0.021103858661807782,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.6157,
|
|
"mean_token_accuracy": 0.1448817826807499,
|
|
"num_tokens": 657597.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.7943517684936525,
|
|
"epoch": 0.02121918029383952,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4777,
|
|
"mean_token_accuracy": 0.1404040850698948,
|
|
"num_tokens": 661466.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.934364891052246,
|
|
"epoch": 0.021334501925871253,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.547,
|
|
"mean_token_accuracy": 0.1400091178715229,
|
|
"num_tokens": 665662.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.662406778335571,
|
|
"epoch": 0.02144982355790299,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3434,
|
|
"mean_token_accuracy": 0.14479073956608773,
|
|
"num_tokens": 669190.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.637304782867432,
|
|
"epoch": 0.02156514518993473,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.413,
|
|
"mean_token_accuracy": 0.13896835520863532,
|
|
"num_tokens": 672785.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.61347222328186,
|
|
"epoch": 0.021680466821966466,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3792,
|
|
"mean_token_accuracy": 0.1475941888988018,
|
|
"num_tokens": 676090.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.706289768218994,
|
|
"epoch": 0.0217957884539982,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.472,
|
|
"mean_token_accuracy": 0.14431513100862503,
|
|
"num_tokens": 679765.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.766093397140503,
|
|
"epoch": 0.021911110086029938,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4545,
|
|
"mean_token_accuracy": 0.14060847386717795,
|
|
"num_tokens": 683289.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.756227016448975,
|
|
"epoch": 0.022026431718061675,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3061,
|
|
"mean_token_accuracy": 0.1499037891626358,
|
|
"num_tokens": 686551.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.574160623550415,
|
|
"epoch": 0.02214175335009341,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5524,
|
|
"mean_token_accuracy": 0.13812233954668046,
|
|
"num_tokens": 689876.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.812145614624024,
|
|
"epoch": 0.022257074982125147,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3679,
|
|
"mean_token_accuracy": 0.15085181668400766,
|
|
"num_tokens": 693258.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.5659544467926025,
|
|
"epoch": 0.022372396614156884,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.453,
|
|
"mean_token_accuracy": 0.15052863135933875,
|
|
"num_tokens": 696908.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.932853555679321,
|
|
"epoch": 0.02248771824618862,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.491,
|
|
"mean_token_accuracy": 0.14638491868972778,
|
|
"num_tokens": 700621.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.573339557647705,
|
|
"epoch": 0.022603039878220356,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3537,
|
|
"mean_token_accuracy": 0.14875999763607978,
|
|
"num_tokens": 703896.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.6739246368408205,
|
|
"epoch": 0.022718361510252093,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1758,
|
|
"mean_token_accuracy": 0.1613540157675743,
|
|
"num_tokens": 707309.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.515360355377197,
|
|
"epoch": 0.02283368314228383,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3102,
|
|
"mean_token_accuracy": 0.1500277727842331,
|
|
"num_tokens": 710719.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.7645505428314205,
|
|
"epoch": 0.022949004774315565,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.4921,
|
|
"mean_token_accuracy": 0.15008721128106117,
|
|
"num_tokens": 715016.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.729737281799316,
|
|
"epoch": 0.023064326406347303,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.4748,
|
|
"mean_token_accuracy": 0.13950299099087715,
|
|
"num_tokens": 718863.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.541002511978149,
|
|
"epoch": 0.02317964803837904,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.2894,
|
|
"mean_token_accuracy": 0.16091133207082747,
|
|
"num_tokens": 722399.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.576257181167603,
|
|
"epoch": 0.023294969670410774,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.3785,
|
|
"mean_token_accuracy": 0.15231261402368546,
|
|
"num_tokens": 726090.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.629475831985474,
|
|
"epoch": 0.023410291302442512,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.333,
|
|
"mean_token_accuracy": 0.1444718509912491,
|
|
"num_tokens": 729749.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.595548295974732,
|
|
"epoch": 0.02352561293447425,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.3985,
|
|
"mean_token_accuracy": 0.14387279450893403,
|
|
"num_tokens": 733370.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.613503074645996,
|
|
"epoch": 0.023640934566505987,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.4033,
|
|
"mean_token_accuracy": 0.15736466497182847,
|
|
"num_tokens": 737084.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.582910394668579,
|
|
"epoch": 0.02375625619853772,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.4187,
|
|
"mean_token_accuracy": 0.14997555688023567,
|
|
"num_tokens": 740907.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.7457200050354,
|
|
"epoch": 0.02387157783056946,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.3403,
|
|
"mean_token_accuracy": 0.1474141374230385,
|
|
"num_tokens": 744546.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.664389944076538,
|
|
"epoch": 0.023986899462601196,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.5069,
|
|
"mean_token_accuracy": 0.14562955796718596,
|
|
"num_tokens": 748289.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.47734580039978,
|
|
"epoch": 0.02410222109463293,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.2938,
|
|
"mean_token_accuracy": 0.15487776100635528,
|
|
"num_tokens": 752034.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.430880784988403,
|
|
"epoch": 0.024217542726664668,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.1899,
|
|
"mean_token_accuracy": 0.15236498713493346,
|
|
"num_tokens": 755312.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.59245982170105,
|
|
"epoch": 0.024332864358696405,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.3682,
|
|
"mean_token_accuracy": 0.14548618867993354,
|
|
"num_tokens": 758921.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.485173511505127,
|
|
"epoch": 0.02444818599072814,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.2434,
|
|
"mean_token_accuracy": 0.1515353113412857,
|
|
"num_tokens": 762392.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.3894353866577145,
|
|
"epoch": 0.024563507622759877,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.2504,
|
|
"mean_token_accuracy": 0.15577544420957565,
|
|
"num_tokens": 766151.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.5687884330749515,
|
|
"epoch": 0.024678829254791614,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.3492,
|
|
"mean_token_accuracy": 0.14603114277124404,
|
|
"num_tokens": 769799.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.479031753540039,
|
|
"epoch": 0.024794150886823352,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.3657,
|
|
"mean_token_accuracy": 0.13736803457140923,
|
|
"num_tokens": 773345.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.408109903335571,
|
|
"epoch": 0.024909472518855086,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.2535,
|
|
"mean_token_accuracy": 0.1497886911034584,
|
|
"num_tokens": 777121.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.575307941436767,
|
|
"epoch": 0.025024794150886823,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.2192,
|
|
"mean_token_accuracy": 0.1585264466702938,
|
|
"num_tokens": 780601.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.532757568359375,
|
|
"epoch": 0.02514011578291856,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.44,
|
|
"mean_token_accuracy": 0.14874088764190674,
|
|
"num_tokens": 783995.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.544535875320435,
|
|
"epoch": 0.025255437414950295,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.3367,
|
|
"mean_token_accuracy": 0.14363013133406638,
|
|
"num_tokens": 787810.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.709648180007934,
|
|
"epoch": 0.025370759046982033,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.4243,
|
|
"mean_token_accuracy": 0.14066330567002297,
|
|
"num_tokens": 791554.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.605457448959351,
|
|
"epoch": 0.02548608067901377,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.2425,
|
|
"mean_token_accuracy": 0.15509673207998276,
|
|
"num_tokens": 795210.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.4382562160491945,
|
|
"epoch": 0.025601402311045508,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.1771,
|
|
"mean_token_accuracy": 0.160766139626503,
|
|
"num_tokens": 798225.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.367309522628784,
|
|
"epoch": 0.02571672394307724,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.2478,
|
|
"mean_token_accuracy": 0.16239843666553497,
|
|
"num_tokens": 801465.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.519221496582031,
|
|
"epoch": 0.02583204557510898,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.2276,
|
|
"mean_token_accuracy": 0.15966097712516786,
|
|
"num_tokens": 804840.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.3273594856262205,
|
|
"epoch": 0.025947367207140717,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.267,
|
|
"mean_token_accuracy": 0.15511202141642572,
|
|
"num_tokens": 808594.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.569507169723511,
|
|
"epoch": 0.02606268883917245,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.3432,
|
|
"mean_token_accuracy": 0.14767091646790503,
|
|
"num_tokens": 812495.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.642676496505738,
|
|
"epoch": 0.02617801047120419,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.3796,
|
|
"mean_token_accuracy": 0.1536947324872017,
|
|
"num_tokens": 816424.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.446162605285645,
|
|
"epoch": 0.026293332103235926,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.4062,
|
|
"mean_token_accuracy": 0.14982397630810737,
|
|
"num_tokens": 820757.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.716174554824829,
|
|
"epoch": 0.02640865373526766,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.3593,
|
|
"mean_token_accuracy": 0.14376568794250488,
|
|
"num_tokens": 824456.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.497501611709595,
|
|
"epoch": 0.026523975367299397,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.256,
|
|
"mean_token_accuracy": 0.1652655228972435,
|
|
"num_tokens": 828144.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.342379236221314,
|
|
"epoch": 0.026639296999331135,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.2368,
|
|
"mean_token_accuracy": 0.16241760402917862,
|
|
"num_tokens": 831802.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.5346297264099125,
|
|
"epoch": 0.026754618631362873,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.3111,
|
|
"mean_token_accuracy": 0.16149049252271652,
|
|
"num_tokens": 835355.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.550338697433472,
|
|
"epoch": 0.026869940263394607,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.3718,
|
|
"mean_token_accuracy": 0.14529037550091745,
|
|
"num_tokens": 839193.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.386302709579468,
|
|
"epoch": 0.026985261895426344,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.2242,
|
|
"mean_token_accuracy": 0.1485719881951809,
|
|
"num_tokens": 842601.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.47716908454895,
|
|
"epoch": 0.02710058352745808,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.2308,
|
|
"mean_token_accuracy": 0.16209668293595314,
|
|
"num_tokens": 846335.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.4549352645874025,
|
|
"epoch": 0.027215905159489816,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.2183,
|
|
"mean_token_accuracy": 0.16738357841968537,
|
|
"num_tokens": 849944.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.530093336105347,
|
|
"epoch": 0.027331226791521553,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.2009,
|
|
"mean_token_accuracy": 0.15745505169034005,
|
|
"num_tokens": 853417.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.36098804473877,
|
|
"epoch": 0.02744654842355329,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.3304,
|
|
"mean_token_accuracy": 0.1461167961359024,
|
|
"num_tokens": 856935.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.608738279342651,
|
|
"epoch": 0.02756187005558503,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.3565,
|
|
"mean_token_accuracy": 0.1573301449418068,
|
|
"num_tokens": 860334.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.462534856796265,
|
|
"epoch": 0.027677191687616762,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.2583,
|
|
"mean_token_accuracy": 0.16238447278738022,
|
|
"num_tokens": 864041.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.404558229446411,
|
|
"epoch": 0.0277925133196485,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.2631,
|
|
"mean_token_accuracy": 0.16061173155903816,
|
|
"num_tokens": 867577.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.463160705566406,
|
|
"epoch": 0.027907834951680238,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.311,
|
|
"mean_token_accuracy": 0.15379835441708564,
|
|
"num_tokens": 871509.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.515782499313355,
|
|
"epoch": 0.02802315658371197,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.2511,
|
|
"mean_token_accuracy": 0.16381916627287865,
|
|
"num_tokens": 874851.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.399502849578857,
|
|
"epoch": 0.02813847821574371,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.2707,
|
|
"mean_token_accuracy": 0.15872995257377626,
|
|
"num_tokens": 878516.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.335233688354492,
|
|
"epoch": 0.028253799847775447,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.1828,
|
|
"mean_token_accuracy": 0.15398550629615784,
|
|
"num_tokens": 882339.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.529935646057129,
|
|
"epoch": 0.02836912147980718,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.4462,
|
|
"mean_token_accuracy": 0.1562889628112316,
|
|
"num_tokens": 886439.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.470861434936523,
|
|
"epoch": 0.028484443111838918,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.3035,
|
|
"mean_token_accuracy": 0.1598047912120819,
|
|
"num_tokens": 890247.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.28164701461792,
|
|
"epoch": 0.028599764743870656,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.266,
|
|
"mean_token_accuracy": 0.14978654831647872,
|
|
"num_tokens": 893576.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.546269941329956,
|
|
"epoch": 0.028715086375902393,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.3187,
|
|
"mean_token_accuracy": 0.15078958198428155,
|
|
"num_tokens": 897163.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.42095251083374,
|
|
"epoch": 0.028830408007934127,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.2401,
|
|
"mean_token_accuracy": 0.16046840250492095,
|
|
"num_tokens": 900787.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.369808721542358,
|
|
"epoch": 0.028945729639965865,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.237,
|
|
"mean_token_accuracy": 0.16006658524274825,
|
|
"num_tokens": 904420.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.41910490989685,
|
|
"epoch": 0.029061051271997602,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.3838,
|
|
"mean_token_accuracy": 0.14555097371339798,
|
|
"num_tokens": 908244.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.552026271820068,
|
|
"epoch": 0.029176372904029337,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.1964,
|
|
"mean_token_accuracy": 0.16036371216177941,
|
|
"num_tokens": 911715.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.2189428329467775,
|
|
"epoch": 0.029291694536061074,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.1044,
|
|
"mean_token_accuracy": 0.17435312271118164,
|
|
"num_tokens": 915039.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.398732280731201,
|
|
"epoch": 0.02940701616809281,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.1189,
|
|
"mean_token_accuracy": 0.16803320348262787,
|
|
"num_tokens": 918744.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.379372930526733,
|
|
"epoch": 0.029522337800124546,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.2731,
|
|
"mean_token_accuracy": 0.15851129293441774,
|
|
"num_tokens": 922235.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.595360231399536,
|
|
"epoch": 0.029637659432156283,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.287,
|
|
"mean_token_accuracy": 0.15708497390151024,
|
|
"num_tokens": 926282.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.537356853485107,
|
|
"epoch": 0.02975298106418802,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.4151,
|
|
"mean_token_accuracy": 0.14585280492901803,
|
|
"num_tokens": 930226.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.335377597808838,
|
|
"epoch": 0.02986830269621976,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.113,
|
|
"mean_token_accuracy": 0.16433655470609665,
|
|
"num_tokens": 933334.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.34308090209961,
|
|
"epoch": 0.029983624328251492,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.0999,
|
|
"mean_token_accuracy": 0.16376755833625795,
|
|
"num_tokens": 936710.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.354935884475708,
|
|
"epoch": 0.03009894596028323,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.1994,
|
|
"mean_token_accuracy": 0.16085999011993407,
|
|
"num_tokens": 940045.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.314391422271728,
|
|
"epoch": 0.030214267592314967,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.1112,
|
|
"mean_token_accuracy": 0.1725856378674507,
|
|
"num_tokens": 943244.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.347221231460571,
|
|
"epoch": 0.0303295892243467,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.178,
|
|
"mean_token_accuracy": 0.16226338669657708,
|
|
"num_tokens": 946590.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.303072881698609,
|
|
"epoch": 0.03044491085637844,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.0578,
|
|
"mean_token_accuracy": 0.16810549795627594,
|
|
"num_tokens": 950027.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.203867053985595,
|
|
"epoch": 0.030560232488410177,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.1479,
|
|
"mean_token_accuracy": 0.1594789668917656,
|
|
"num_tokens": 953542.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.2548261165618895,
|
|
"epoch": 0.030675554120441914,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.0418,
|
|
"mean_token_accuracy": 0.1676800087094307,
|
|
"num_tokens": 957031.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.4141487121582035,
|
|
"epoch": 0.030790875752473648,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.2319,
|
|
"mean_token_accuracy": 0.1546824663877487,
|
|
"num_tokens": 960895.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.184122848510742,
|
|
"epoch": 0.030906197384505386,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.1137,
|
|
"mean_token_accuracy": 0.1634022995829582,
|
|
"num_tokens": 964532.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.253350305557251,
|
|
"epoch": 0.031021519016537123,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.1711,
|
|
"mean_token_accuracy": 0.15790961310267448,
|
|
"num_tokens": 968256.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.305469131469726,
|
|
"epoch": 0.031136840648568857,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.0398,
|
|
"mean_token_accuracy": 0.1720492035150528,
|
|
"num_tokens": 971626.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.1880847930908205,
|
|
"epoch": 0.0312521622806006,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.1004,
|
|
"mean_token_accuracy": 0.16066511869430541,
|
|
"num_tokens": 975116.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.3699556350708,
|
|
"epoch": 0.03136748391263233,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.2177,
|
|
"mean_token_accuracy": 0.1631898730993271,
|
|
"num_tokens": 978899.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.378862428665161,
|
|
"epoch": 0.031482805544664066,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.3456,
|
|
"mean_token_accuracy": 0.14527880921959876,
|
|
"num_tokens": 982993.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.235144138336182,
|
|
"epoch": 0.03159812717669581,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.0272,
|
|
"mean_token_accuracy": 0.16856341138482095,
|
|
"num_tokens": 986580.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.247233915328979,
|
|
"epoch": 0.03171344880872754,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.0216,
|
|
"mean_token_accuracy": 0.16976505219936372,
|
|
"num_tokens": 990248.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.2317156314849855,
|
|
"epoch": 0.031828770440759276,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.1127,
|
|
"mean_token_accuracy": 0.14935898035764694,
|
|
"num_tokens": 994054.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.20282769203186,
|
|
"epoch": 0.03194409207279102,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.1959,
|
|
"mean_token_accuracy": 0.15316805019974708,
|
|
"num_tokens": 997618.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.41049427986145,
|
|
"epoch": 0.03205941370482275,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.1908,
|
|
"mean_token_accuracy": 0.15548152029514312,
|
|
"num_tokens": 1001715.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.2962079524993895,
|
|
"epoch": 0.032174735336854485,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.1458,
|
|
"mean_token_accuracy": 0.1648909866809845,
|
|
"num_tokens": 1005405.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.3217706203460695,
|
|
"epoch": 0.032290056968886226,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.0409,
|
|
"mean_token_accuracy": 0.1789014533162117,
|
|
"num_tokens": 1008892.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.278942584991455,
|
|
"epoch": 0.03240537860091796,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.2614,
|
|
"mean_token_accuracy": 0.15238809138536452,
|
|
"num_tokens": 1012658.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.321128797531128,
|
|
"epoch": 0.032520700232949694,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.202,
|
|
"mean_token_accuracy": 0.1605304330587387,
|
|
"num_tokens": 1016187.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.450961542129517,
|
|
"epoch": 0.032636021864981435,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.124,
|
|
"mean_token_accuracy": 0.15989470332860947,
|
|
"num_tokens": 1019598.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.216757917404175,
|
|
"epoch": 0.03275134349701317,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.0581,
|
|
"mean_token_accuracy": 0.16423142403364183,
|
|
"num_tokens": 1022948.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.099599695205688,
|
|
"epoch": 0.0328666651290449,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.091,
|
|
"mean_token_accuracy": 0.16470970213413239,
|
|
"num_tokens": 1026376.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.215214109420776,
|
|
"epoch": 0.032981986761076644,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.2133,
|
|
"mean_token_accuracy": 0.16824086904525756,
|
|
"num_tokens": 1029882.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.254010725021362,
|
|
"epoch": 0.03309730839310838,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.1434,
|
|
"mean_token_accuracy": 0.15644071772694587,
|
|
"num_tokens": 1033663.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.339106035232544,
|
|
"epoch": 0.03321263002514012,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.2149,
|
|
"mean_token_accuracy": 0.1582715891301632,
|
|
"num_tokens": 1037250.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.265019702911377,
|
|
"epoch": 0.03332795165717185,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.2503,
|
|
"mean_token_accuracy": 0.14860587269067765,
|
|
"num_tokens": 1040848.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.171944618225098,
|
|
"epoch": 0.03344327328920359,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.1657,
|
|
"mean_token_accuracy": 0.1681614875793457,
|
|
"num_tokens": 1044268.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.490657043457031,
|
|
"epoch": 0.03355859492123533,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.2097,
|
|
"mean_token_accuracy": 0.15496580824255943,
|
|
"num_tokens": 1048374.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.245100593566894,
|
|
"epoch": 0.03367391655326706,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 5.9471,
|
|
"mean_token_accuracy": 0.16349475234746932,
|
|
"num_tokens": 1051799.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.282435035705566,
|
|
"epoch": 0.033789238185298796,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.0923,
|
|
"mean_token_accuracy": 0.1628187417984009,
|
|
"num_tokens": 1055460.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.252931547164917,
|
|
"epoch": 0.03390455981733054,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.0997,
|
|
"mean_token_accuracy": 0.15778620466589927,
|
|
"num_tokens": 1058995.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.176799249649048,
|
|
"epoch": 0.03401988144936227,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.1032,
|
|
"mean_token_accuracy": 0.16188894361257553,
|
|
"num_tokens": 1062631.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.22296371459961,
|
|
"epoch": 0.034135203081394005,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.0324,
|
|
"mean_token_accuracy": 0.16586243510246276,
|
|
"num_tokens": 1065671.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.264580965042114,
|
|
"epoch": 0.034250524713425747,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.0644,
|
|
"mean_token_accuracy": 0.16532386392354964,
|
|
"num_tokens": 1069123.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.292820930480957,
|
|
"epoch": 0.03436584634545748,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.1888,
|
|
"mean_token_accuracy": 0.15297986268997193,
|
|
"num_tokens": 1073063.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.18399076461792,
|
|
"epoch": 0.034481167977489215,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.1284,
|
|
"mean_token_accuracy": 0.1573845401406288,
|
|
"num_tokens": 1076914.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.254384469985962,
|
|
"epoch": 0.034596489609520956,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.1672,
|
|
"mean_token_accuracy": 0.1531804010272026,
|
|
"num_tokens": 1080422.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.238947200775146,
|
|
"epoch": 0.03471181124155269,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.1903,
|
|
"mean_token_accuracy": 0.15972661301493646,
|
|
"num_tokens": 1084000.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.227514696121216,
|
|
"epoch": 0.034827132873584424,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.1104,
|
|
"mean_token_accuracy": 0.165690079331398,
|
|
"num_tokens": 1087296.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.450083017349243,
|
|
"epoch": 0.034942454505616165,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.0305,
|
|
"mean_token_accuracy": 0.16898806318640708,
|
|
"num_tokens": 1090868.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.215689706802368,
|
|
"epoch": 0.0350577761376479,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.09,
|
|
"mean_token_accuracy": 0.15983846932649612,
|
|
"num_tokens": 1094940.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.183223676681519,
|
|
"epoch": 0.03517309776967964,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.0616,
|
|
"mean_token_accuracy": 0.15684643536806106,
|
|
"num_tokens": 1098556.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.268905258178711,
|
|
"epoch": 0.035288419401711374,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.1923,
|
|
"mean_token_accuracy": 0.1598368391394615,
|
|
"num_tokens": 1102433.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.101678419113159,
|
|
"epoch": 0.03540374103374311,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.0911,
|
|
"mean_token_accuracy": 0.16665463000535966,
|
|
"num_tokens": 1105919.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.277998304367065,
|
|
"epoch": 0.03551906266577485,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.1602,
|
|
"mean_token_accuracy": 0.16617074608802795,
|
|
"num_tokens": 1109756.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.236911725997925,
|
|
"epoch": 0.03563438429780658,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.0208,
|
|
"mean_token_accuracy": 0.17212379723787308,
|
|
"num_tokens": 1113199.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.124103689193726,
|
|
"epoch": 0.03574970592983832,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.1149,
|
|
"mean_token_accuracy": 0.16162882149219512,
|
|
"num_tokens": 1116834.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.255211210250854,
|
|
"epoch": 0.03586502756187006,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.1283,
|
|
"mean_token_accuracy": 0.15524285286664963,
|
|
"num_tokens": 1120429.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.223068618774414,
|
|
"epoch": 0.03598034919390179,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.1094,
|
|
"mean_token_accuracy": 0.1781144469976425,
|
|
"num_tokens": 1123775.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.1774396896362305,
|
|
"epoch": 0.036095670825933526,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.1552,
|
|
"mean_token_accuracy": 0.15898154228925704,
|
|
"num_tokens": 1127752.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.185403776168823,
|
|
"epoch": 0.03621099245796527,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 5.9629,
|
|
"mean_token_accuracy": 0.18207648098468782,
|
|
"num_tokens": 1131238.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.226887369155884,
|
|
"epoch": 0.036326314089997,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.1029,
|
|
"mean_token_accuracy": 0.15455495268106462,
|
|
"num_tokens": 1134882.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.236399078369141,
|
|
"epoch": 0.036441635722028735,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 6.0193,
|
|
"mean_token_accuracy": 0.1584901764988899,
|
|
"num_tokens": 1138410.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.137038803100586,
|
|
"epoch": 0.036556957354060476,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.0027,
|
|
"mean_token_accuracy": 0.17161184996366502,
|
|
"num_tokens": 1141737.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.244590997695923,
|
|
"epoch": 0.03667227898609221,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.1066,
|
|
"mean_token_accuracy": 0.159149469435215,
|
|
"num_tokens": 1145647.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.216052436828614,
|
|
"epoch": 0.036787600618123945,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.1649,
|
|
"mean_token_accuracy": 0.1610773205757141,
|
|
"num_tokens": 1149546.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.2309596061706545,
|
|
"epoch": 0.036902922250155686,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 5.8762,
|
|
"mean_token_accuracy": 0.17929302603006364,
|
|
"num_tokens": 1152816.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.978598880767822,
|
|
"epoch": 0.03701824388218742,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 5.9288,
|
|
"mean_token_accuracy": 0.17676628082990647,
|
|
"num_tokens": 1155993.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.164655876159668,
|
|
"epoch": 0.03713356551421916,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.0275,
|
|
"mean_token_accuracy": 0.17046592831611634,
|
|
"num_tokens": 1159497.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.065205383300781,
|
|
"epoch": 0.037248887146250895,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 5.9666,
|
|
"mean_token_accuracy": 0.1708380624651909,
|
|
"num_tokens": 1163040.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.210390901565551,
|
|
"epoch": 0.03736420877828263,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.1469,
|
|
"mean_token_accuracy": 0.16143751963973046,
|
|
"num_tokens": 1166645.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.357365942001342,
|
|
"epoch": 0.03747953041031437,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.1428,
|
|
"mean_token_accuracy": 0.15922287330031396,
|
|
"num_tokens": 1170178.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.210607433319092,
|
|
"epoch": 0.037594852042346104,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.073,
|
|
"mean_token_accuracy": 0.15819956809282304,
|
|
"num_tokens": 1173773.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.24861855506897,
|
|
"epoch": 0.03771017367437784,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.1206,
|
|
"mean_token_accuracy": 0.15854904279112816,
|
|
"num_tokens": 1177233.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.274668598175049,
|
|
"epoch": 0.03782549530640958,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 5.9639,
|
|
"mean_token_accuracy": 0.16956764906644822,
|
|
"num_tokens": 1180870.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.986456394195557,
|
|
"epoch": 0.03794081693844131,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 5.9269,
|
|
"mean_token_accuracy": 0.16868334710597993,
|
|
"num_tokens": 1184313.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.08156681060791,
|
|
"epoch": 0.03805613857047305,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.0561,
|
|
"mean_token_accuracy": 0.16173017472028733,
|
|
"num_tokens": 1187743.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.25056734085083,
|
|
"epoch": 0.03817146020250479,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 5.9517,
|
|
"mean_token_accuracy": 0.17180202454328536,
|
|
"num_tokens": 1191486.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.103668355941773,
|
|
"epoch": 0.03828678183453652,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 5.9847,
|
|
"mean_token_accuracy": 0.17403454631567,
|
|
"num_tokens": 1194668.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.132091379165649,
|
|
"epoch": 0.038402103466568256,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.0378,
|
|
"mean_token_accuracy": 0.1682954505085945,
|
|
"num_tokens": 1198465.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.10410966873169,
|
|
"epoch": 0.0385174250986,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.0501,
|
|
"mean_token_accuracy": 0.16741879433393478,
|
|
"num_tokens": 1202343.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.20172085762024,
|
|
"epoch": 0.03863274673063173,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 5.9804,
|
|
"mean_token_accuracy": 0.16315271854400634,
|
|
"num_tokens": 1205680.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.006396102905273,
|
|
"epoch": 0.038748068362663465,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 5.9253,
|
|
"mean_token_accuracy": 0.18567429184913636,
|
|
"num_tokens": 1209105.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.081904554367066,
|
|
"epoch": 0.038863389994695206,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 6.0643,
|
|
"mean_token_accuracy": 0.16113972365856172,
|
|
"num_tokens": 1212485.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.156144905090332,
|
|
"epoch": 0.03897871162672694,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.0159,
|
|
"mean_token_accuracy": 0.16522784233093263,
|
|
"num_tokens": 1216052.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.212941694259643,
|
|
"epoch": 0.03909403325875868,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.0762,
|
|
"mean_token_accuracy": 0.16660564690828322,
|
|
"num_tokens": 1219542.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.074598503112793,
|
|
"epoch": 0.039209354890790415,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.056,
|
|
"mean_token_accuracy": 0.16291215270757675,
|
|
"num_tokens": 1223243.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.229275178909302,
|
|
"epoch": 0.03932467652282215,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.0463,
|
|
"mean_token_accuracy": 0.16674332916736603,
|
|
"num_tokens": 1226764.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.23920693397522,
|
|
"epoch": 0.03943999815485389,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 6.0357,
|
|
"mean_token_accuracy": 0.16611389368772506,
|
|
"num_tokens": 1230361.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.145705604553223,
|
|
"epoch": 0.039555319786885625,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.1405,
|
|
"mean_token_accuracy": 0.14990901276469232,
|
|
"num_tokens": 1233789.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.144283628463745,
|
|
"epoch": 0.03967064141891736,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 5.9701,
|
|
"mean_token_accuracy": 0.1690068542957306,
|
|
"num_tokens": 1237552.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.206836986541748,
|
|
"epoch": 0.0397859630509491,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 5.9928,
|
|
"mean_token_accuracy": 0.15885717421770096,
|
|
"num_tokens": 1241273.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.26685733795166,
|
|
"epoch": 0.039901284682980834,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.1141,
|
|
"mean_token_accuracy": 0.16296201050281525,
|
|
"num_tokens": 1245072.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.135778427124023,
|
|
"epoch": 0.04001660631501257,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 5.9754,
|
|
"mean_token_accuracy": 0.17653044760227204,
|
|
"num_tokens": 1248744.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.121457862854004,
|
|
"epoch": 0.04013192794704431,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.0296,
|
|
"mean_token_accuracy": 0.16909366846084595,
|
|
"num_tokens": 1252373.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.1641534805297855,
|
|
"epoch": 0.04024724957907604,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 5.927,
|
|
"mean_token_accuracy": 0.17557682394981383,
|
|
"num_tokens": 1255866.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.134204149246216,
|
|
"epoch": 0.04036257121110778,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.0505,
|
|
"mean_token_accuracy": 0.1615603044629097,
|
|
"num_tokens": 1259456.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.084139108657837,
|
|
"epoch": 0.04047789284313952,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 5.9606,
|
|
"mean_token_accuracy": 0.17955740839242934,
|
|
"num_tokens": 1263038.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.167677164077759,
|
|
"epoch": 0.04059321447517125,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.0044,
|
|
"mean_token_accuracy": 0.1757432699203491,
|
|
"num_tokens": 1266905.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.2237554550170895,
|
|
"epoch": 0.040708536107202986,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.0491,
|
|
"mean_token_accuracy": 0.16226842552423476,
|
|
"num_tokens": 1270614.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.220596837997436,
|
|
"epoch": 0.04082385773923473,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 5.9523,
|
|
"mean_token_accuracy": 0.16239065304398537,
|
|
"num_tokens": 1274214.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.244525480270386,
|
|
"epoch": 0.04093917937126646,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 5.8409,
|
|
"mean_token_accuracy": 0.18035510331392288,
|
|
"num_tokens": 1277611.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.969187927246094,
|
|
"epoch": 0.041054501003298195,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 5.9211,
|
|
"mean_token_accuracy": 0.17024255245923997,
|
|
"num_tokens": 1281168.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.120304203033447,
|
|
"epoch": 0.041169822635329936,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.0313,
|
|
"mean_token_accuracy": 0.17333071082830429,
|
|
"num_tokens": 1284869.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.335909366607666,
|
|
"epoch": 0.04128514426736167,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.0544,
|
|
"mean_token_accuracy": 0.16203091889619828,
|
|
"num_tokens": 1288666.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.01842565536499,
|
|
"epoch": 0.04140046589939341,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 6.0631,
|
|
"mean_token_accuracy": 0.17271315678954124,
|
|
"num_tokens": 1292139.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.2019682884216305,
|
|
"epoch": 0.041515787531425145,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.0193,
|
|
"mean_token_accuracy": 0.15610606372356414,
|
|
"num_tokens": 1295951.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.293931865692139,
|
|
"epoch": 0.04163110916345688,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.0272,
|
|
"mean_token_accuracy": 0.16180424988269806,
|
|
"num_tokens": 1299512.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.119784879684448,
|
|
"epoch": 0.04174643079548862,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 5.9606,
|
|
"mean_token_accuracy": 0.16695330440998077,
|
|
"num_tokens": 1303159.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.185426712036133,
|
|
"epoch": 0.041861752427520355,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 6.0686,
|
|
"mean_token_accuracy": 0.1705578237771988,
|
|
"num_tokens": 1307109.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.2974461078643795,
|
|
"epoch": 0.04197707405955209,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 5.9839,
|
|
"mean_token_accuracy": 0.16454261690378189,
|
|
"num_tokens": 1310742.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.310203504562378,
|
|
"epoch": 0.04209239569158383,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 6.1416,
|
|
"mean_token_accuracy": 0.16858258694410325,
|
|
"num_tokens": 1314321.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.108058595657349,
|
|
"epoch": 0.042207717323615564,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 6.0898,
|
|
"mean_token_accuracy": 0.16480854004621506,
|
|
"num_tokens": 1317760.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.2883891582489015,
|
|
"epoch": 0.0423230389556473,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.0478,
|
|
"mean_token_accuracy": 0.16385416239500045,
|
|
"num_tokens": 1321585.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.10967869758606,
|
|
"epoch": 0.04243836058767904,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 5.9995,
|
|
"mean_token_accuracy": 0.16069755405187608,
|
|
"num_tokens": 1325132.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.031411075592041,
|
|
"epoch": 0.04255368221971077,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 5.9887,
|
|
"mean_token_accuracy": 0.17086584120988846,
|
|
"num_tokens": 1328522.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.050738000869751,
|
|
"epoch": 0.04266900385174251,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 5.8527,
|
|
"mean_token_accuracy": 0.1780827283859253,
|
|
"num_tokens": 1331720.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.090122127532959,
|
|
"epoch": 0.04278432548377425,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.0946,
|
|
"mean_token_accuracy": 0.16792099177837372,
|
|
"num_tokens": 1335473.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.222653913497925,
|
|
"epoch": 0.04289964711580598,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 5.998,
|
|
"mean_token_accuracy": 0.1554895505309105,
|
|
"num_tokens": 1338848.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.259007310867309,
|
|
"epoch": 0.043014968747837716,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.0702,
|
|
"mean_token_accuracy": 0.16148973554372786,
|
|
"num_tokens": 1342260.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.13902096748352,
|
|
"epoch": 0.04313029037986946,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 5.9242,
|
|
"mean_token_accuracy": 0.17190041840076448,
|
|
"num_tokens": 1345864.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.149142265319824,
|
|
"epoch": 0.04324561201190119,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 5.8591,
|
|
"mean_token_accuracy": 0.17256597727537154,
|
|
"num_tokens": 1349787.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.214622879028321,
|
|
"epoch": 0.04336093364393293,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.0664,
|
|
"mean_token_accuracy": 0.17647800892591475,
|
|
"num_tokens": 1353515.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.0533750534057615,
|
|
"epoch": 0.043476255275964666,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 5.8783,
|
|
"mean_token_accuracy": 0.18349718004465104,
|
|
"num_tokens": 1356918.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.170057153701782,
|
|
"epoch": 0.0435915769079964,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.0514,
|
|
"mean_token_accuracy": 0.16036743074655532,
|
|
"num_tokens": 1360804.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.148578119277954,
|
|
"epoch": 0.04370689854002814,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.0567,
|
|
"mean_token_accuracy": 0.1664833903312683,
|
|
"num_tokens": 1364246.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.159447717666626,
|
|
"epoch": 0.043822220172059875,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.1766192525625229,
|
|
"num_tokens": 1367607.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.023313426971436,
|
|
"epoch": 0.04393754180409161,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 5.9252,
|
|
"mean_token_accuracy": 0.17720281779766084,
|
|
"num_tokens": 1370886.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.068740653991699,
|
|
"epoch": 0.04405286343612335,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.0774,
|
|
"mean_token_accuracy": 0.16251763105392455,
|
|
"num_tokens": 1375071.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.10937271118164,
|
|
"epoch": 0.044168185068155084,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 5.927,
|
|
"mean_token_accuracy": 0.1618012472987175,
|
|
"num_tokens": 1378735.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.125348854064941,
|
|
"epoch": 0.04428350670018682,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.1631,
|
|
"mean_token_accuracy": 0.1595492482185364,
|
|
"num_tokens": 1382397.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.2215968608856205,
|
|
"epoch": 0.04439882833221856,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.0301,
|
|
"mean_token_accuracy": 0.16059124022722243,
|
|
"num_tokens": 1385953.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.0450019359588625,
|
|
"epoch": 0.044514149964250294,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 5.9631,
|
|
"mean_token_accuracy": 0.16831020414829254,
|
|
"num_tokens": 1389621.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.117081499099731,
|
|
"epoch": 0.04462947159628203,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 5.8663,
|
|
"mean_token_accuracy": 0.18380117416381836,
|
|
"num_tokens": 1392750.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.168520879745484,
|
|
"epoch": 0.04474479322831377,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 5.9942,
|
|
"mean_token_accuracy": 0.17892816960811614,
|
|
"num_tokens": 1396410.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.10781831741333,
|
|
"epoch": 0.0448601148603455,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.1357,
|
|
"mean_token_accuracy": 0.158907001465559,
|
|
"num_tokens": 1400340.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.096987581253051,
|
|
"epoch": 0.04497543649237724,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 5.8511,
|
|
"mean_token_accuracy": 0.17481426373124123,
|
|
"num_tokens": 1403841.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.130824041366577,
|
|
"epoch": 0.04509075812440898,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 5.9086,
|
|
"mean_token_accuracy": 0.17142019122838975,
|
|
"num_tokens": 1407264.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.09589409828186,
|
|
"epoch": 0.04520607975644071,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.0241,
|
|
"mean_token_accuracy": 0.1625817283987999,
|
|
"num_tokens": 1411212.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.081521701812744,
|
|
"epoch": 0.04532140138847245,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 5.9585,
|
|
"mean_token_accuracy": 0.1761806234717369,
|
|
"num_tokens": 1414929.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.128235244750977,
|
|
"epoch": 0.04543672302050419,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.8591,
|
|
"mean_token_accuracy": 0.17452277839183808,
|
|
"num_tokens": 1418211.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.017612552642822,
|
|
"epoch": 0.04555204465253592,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 5.8183,
|
|
"mean_token_accuracy": 0.18118403851985931,
|
|
"num_tokens": 1421712.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.103037357330322,
|
|
"epoch": 0.04566736628456766,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 5.8749,
|
|
"mean_token_accuracy": 0.1705790713429451,
|
|
"num_tokens": 1424949.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.961900520324707,
|
|
"epoch": 0.045782687916599396,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.9654,
|
|
"mean_token_accuracy": 0.17369249761104583,
|
|
"num_tokens": 1428304.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.027649593353272,
|
|
"epoch": 0.04589800954863113,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 5.8704,
|
|
"mean_token_accuracy": 0.18264755457639695,
|
|
"num_tokens": 1432157.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.1183741092681885,
|
|
"epoch": 0.04601333118066287,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.035,
|
|
"mean_token_accuracy": 0.16388218998908996,
|
|
"num_tokens": 1435723.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.0956000804901125,
|
|
"epoch": 0.046128652812694605,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 5.8406,
|
|
"mean_token_accuracy": 0.17017288506031036,
|
|
"num_tokens": 1439105.0,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 329616229662720.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|