1035 lines
27 KiB
JSON
1035 lines
27 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.011532163203173651,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 82976677724160.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|