4035 lines
109 KiB
JSON
4035 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.12859255449109497,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742344284057618,
|
|
"epoch": 0.00032148138622773744,
|
|
"grad_norm": 5.1875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7889,
|
|
"mean_token_accuracy": 0.0001923076924867928,
|
|
"num_tokens": 10787.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742341995239258,
|
|
"epoch": 0.0006429627724554749,
|
|
"grad_norm": 4.6875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7754,
|
|
"mean_token_accuracy": 9.199631749652326e-05,
|
|
"num_tokens": 21408.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.74231481552124,
|
|
"epoch": 0.0009644441586832123,
|
|
"grad_norm": 4.84375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7745,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 33838.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742263031005859,
|
|
"epoch": 0.0012859255449109497,
|
|
"grad_norm": 4.875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.7415,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 46592.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742058944702148,
|
|
"epoch": 0.001607406931138687,
|
|
"grad_norm": 4.21875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6278,
|
|
"mean_token_accuracy": 0.00028761792345903813,
|
|
"num_tokens": 57409.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741728782653809,
|
|
"epoch": 0.0019288883173664245,
|
|
"grad_norm": 3.71875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.5262,
|
|
"mean_token_accuracy": 0.005182119726669043,
|
|
"num_tokens": 68705.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.74124059677124,
|
|
"epoch": 0.002250369703594162,
|
|
"grad_norm": 3.09375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.4297,
|
|
"mean_token_accuracy": 0.019557270966470242,
|
|
"num_tokens": 81460.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.740433311462402,
|
|
"epoch": 0.0025718510898218995,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2725,
|
|
"mean_token_accuracy": 0.02960890345275402,
|
|
"num_tokens": 93576.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.739229869842529,
|
|
"epoch": 0.0028933324760496365,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.1857,
|
|
"mean_token_accuracy": 0.02886150423437357,
|
|
"num_tokens": 105556.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.738618659973145,
|
|
"epoch": 0.003214813862277374,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.0792,
|
|
"mean_token_accuracy": 0.031074166856706144,
|
|
"num_tokens": 117850.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.738714694976807,
|
|
"epoch": 0.0035362952485051115,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.0244,
|
|
"mean_token_accuracy": 0.031163782812654972,
|
|
"num_tokens": 130955.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.738016033172608,
|
|
"epoch": 0.003857776634732849,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.9307,
|
|
"mean_token_accuracy": 0.03660368006676436,
|
|
"num_tokens": 141959.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.736786365509033,
|
|
"epoch": 0.0041792580209605865,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8915,
|
|
"mean_token_accuracy": 0.03608605414628983,
|
|
"num_tokens": 153856.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.73516550064087,
|
|
"epoch": 0.004500739407188324,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8384,
|
|
"mean_token_accuracy": 0.043716078624129295,
|
|
"num_tokens": 166748.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.732550621032715,
|
|
"epoch": 0.0048222207934160615,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.728,
|
|
"mean_token_accuracy": 0.04251385778188706,
|
|
"num_tokens": 177172.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.728882122039796,
|
|
"epoch": 0.005143702179643799,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6737,
|
|
"mean_token_accuracy": 0.04158058017492294,
|
|
"num_tokens": 189418.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.723980331420899,
|
|
"epoch": 0.0054651835658715365,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5846,
|
|
"mean_token_accuracy": 0.039340490289032456,
|
|
"num_tokens": 201349.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.718004131317139,
|
|
"epoch": 0.005786664952099273,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.5176,
|
|
"mean_token_accuracy": 0.04345524609088898,
|
|
"num_tokens": 212298.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.710467433929443,
|
|
"epoch": 0.0061081463383270106,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.467,
|
|
"mean_token_accuracy": 0.04573878571391106,
|
|
"num_tokens": 223831.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.699566841125488,
|
|
"epoch": 0.006429627724554748,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3601,
|
|
"mean_token_accuracy": 0.04835316389799118,
|
|
"num_tokens": 236675.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.683355236053467,
|
|
"epoch": 0.0067511091107824855,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.242,
|
|
"mean_token_accuracy": 0.04662417732179165,
|
|
"num_tokens": 247634.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.66143503189087,
|
|
"epoch": 0.007072590497010223,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1162,
|
|
"mean_token_accuracy": 0.05328451320528984,
|
|
"num_tokens": 259101.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.631138801574707,
|
|
"epoch": 0.0073940718832379605,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0393,
|
|
"mean_token_accuracy": 0.05271790884435177,
|
|
"num_tokens": 272151.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.594438171386718,
|
|
"epoch": 0.007715553269465698,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8606,
|
|
"mean_token_accuracy": 0.057512912154197696,
|
|
"num_tokens": 283619.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.537901401519775,
|
|
"epoch": 0.008037034655693436,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7551,
|
|
"mean_token_accuracy": 0.05864038914442062,
|
|
"num_tokens": 296233.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.46686897277832,
|
|
"epoch": 0.008358516041921173,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6685,
|
|
"mean_token_accuracy": 0.05632430389523506,
|
|
"num_tokens": 309222.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.386721515655518,
|
|
"epoch": 0.00867999742814891,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4983,
|
|
"mean_token_accuracy": 0.05754401199519634,
|
|
"num_tokens": 322291.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.285922241210937,
|
|
"epoch": 0.009001478814376648,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.3328,
|
|
"mean_token_accuracy": 0.06465739980340005,
|
|
"num_tokens": 333520.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.168986701965332,
|
|
"epoch": 0.009322960200604385,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.1641,
|
|
"mean_token_accuracy": 0.0657901257276535,
|
|
"num_tokens": 344883.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.042533111572265,
|
|
"epoch": 0.009644441586832123,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.0727,
|
|
"mean_token_accuracy": 0.07245785929262638,
|
|
"num_tokens": 356600.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.882461643218994,
|
|
"epoch": 0.00996592297305986,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.9127,
|
|
"mean_token_accuracy": 0.0721237450838089,
|
|
"num_tokens": 368527.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.694390869140625,
|
|
"epoch": 0.010287404359287598,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8511,
|
|
"mean_token_accuracy": 0.06826285421848297,
|
|
"num_tokens": 379836.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.510071659088135,
|
|
"epoch": 0.010608885745515335,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.8393,
|
|
"mean_token_accuracy": 0.0674971140921116,
|
|
"num_tokens": 392535.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.299476528167725,
|
|
"epoch": 0.010930367131743073,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7515,
|
|
"mean_token_accuracy": 0.06860553249716758,
|
|
"num_tokens": 405259.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.040483093261718,
|
|
"epoch": 0.011251848517970809,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.599,
|
|
"mean_token_accuracy": 0.07114262394607067,
|
|
"num_tokens": 417922.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.870688247680665,
|
|
"epoch": 0.011573329904198546,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.6169,
|
|
"mean_token_accuracy": 0.07012484185397624,
|
|
"num_tokens": 429411.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.67488489151001,
|
|
"epoch": 0.011894811290426284,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.386,
|
|
"mean_token_accuracy": 0.0772802360355854,
|
|
"num_tokens": 439950.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.489033699035645,
|
|
"epoch": 0.012216292676654021,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3822,
|
|
"mean_token_accuracy": 0.07805034667253494,
|
|
"num_tokens": 451245.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.393837928771973,
|
|
"epoch": 0.012537774062881759,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4539,
|
|
"mean_token_accuracy": 0.07150709182024002,
|
|
"num_tokens": 465146.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.359711456298829,
|
|
"epoch": 0.012859255449109496,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4708,
|
|
"mean_token_accuracy": 0.07765417769551278,
|
|
"num_tokens": 477079.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.254557895660401,
|
|
"epoch": 0.013180736835337234,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2964,
|
|
"mean_token_accuracy": 0.08167731463909149,
|
|
"num_tokens": 488216.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.201800632476807,
|
|
"epoch": 0.013502218221564971,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2996,
|
|
"mean_token_accuracy": 0.08416381254792213,
|
|
"num_tokens": 501040.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.122812938690185,
|
|
"epoch": 0.013823699607792709,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2535,
|
|
"mean_token_accuracy": 0.08663205504417419,
|
|
"num_tokens": 512636.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.088646793365479,
|
|
"epoch": 0.014145180994020446,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.301,
|
|
"mean_token_accuracy": 0.08183000981807709,
|
|
"num_tokens": 525249.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.097221660614014,
|
|
"epoch": 0.014466662380248184,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3486,
|
|
"mean_token_accuracy": 0.08002460822463035,
|
|
"num_tokens": 536963.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.05938491821289,
|
|
"epoch": 0.014788143766475921,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2099,
|
|
"mean_token_accuracy": 0.08030182272195815,
|
|
"num_tokens": 549008.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.041013956069946,
|
|
"epoch": 0.015109625152703659,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2319,
|
|
"mean_token_accuracy": 0.09074903577566147,
|
|
"num_tokens": 560958.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.961578607559204,
|
|
"epoch": 0.015431106538931396,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1147,
|
|
"mean_token_accuracy": 0.09378238469362259,
|
|
"num_tokens": 572472.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.922268009185791,
|
|
"epoch": 0.015752587925159132,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1808,
|
|
"mean_token_accuracy": 0.08942435756325721,
|
|
"num_tokens": 585125.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.913467741012573,
|
|
"epoch": 0.01607406931138687,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1259,
|
|
"mean_token_accuracy": 0.09376866966485978,
|
|
"num_tokens": 597774.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.856596946716309,
|
|
"epoch": 0.016395550697614607,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1233,
|
|
"mean_token_accuracy": 0.09992039278149605,
|
|
"num_tokens": 610130.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.8915050506591795,
|
|
"epoch": 0.016717032083842346,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1488,
|
|
"mean_token_accuracy": 0.09605236500501632,
|
|
"num_tokens": 621659.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.833514070510864,
|
|
"epoch": 0.017038513470070082,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0552,
|
|
"mean_token_accuracy": 0.09916835874319077,
|
|
"num_tokens": 633172.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.854182243347168,
|
|
"epoch": 0.01735999485629782,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0595,
|
|
"mean_token_accuracy": 0.09681920260190964,
|
|
"num_tokens": 643928.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.712904834747315,
|
|
"epoch": 0.017681476242525557,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.9556,
|
|
"mean_token_accuracy": 0.09763396382331849,
|
|
"num_tokens": 654693.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.817079973220825,
|
|
"epoch": 0.018002957628753296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1338,
|
|
"mean_token_accuracy": 0.09172611869871616,
|
|
"num_tokens": 667216.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.727632856369018,
|
|
"epoch": 0.01832443901498103,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9273,
|
|
"mean_token_accuracy": 0.10492636933922768,
|
|
"num_tokens": 679605.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.6807286739349365,
|
|
"epoch": 0.01864592040120877,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.0243,
|
|
"mean_token_accuracy": 0.09721777960658073,
|
|
"num_tokens": 691581.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.730735111236572,
|
|
"epoch": 0.018967401787436507,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.064,
|
|
"mean_token_accuracy": 0.10062759071588516,
|
|
"num_tokens": 703162.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.675752592086792,
|
|
"epoch": 0.019288883173664246,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.947,
|
|
"mean_token_accuracy": 0.10126433670520782,
|
|
"num_tokens": 715024.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.604991054534912,
|
|
"epoch": 0.01961036455989198,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9484,
|
|
"mean_token_accuracy": 0.10657469853758812,
|
|
"num_tokens": 728434.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.582369089126587,
|
|
"epoch": 0.01993184594611972,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9856,
|
|
"mean_token_accuracy": 0.10384280532598496,
|
|
"num_tokens": 741049.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.556066989898682,
|
|
"epoch": 0.020253327332347457,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8995,
|
|
"mean_token_accuracy": 0.10891256630420684,
|
|
"num_tokens": 753316.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.487519884109497,
|
|
"epoch": 0.020574808718575196,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.8976,
|
|
"mean_token_accuracy": 0.10375816151499748,
|
|
"num_tokens": 766605.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.575658369064331,
|
|
"epoch": 0.02089629010480293,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8776,
|
|
"mean_token_accuracy": 0.10548696890473366,
|
|
"num_tokens": 778857.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.49542875289917,
|
|
"epoch": 0.02121777149103067,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.8801,
|
|
"mean_token_accuracy": 0.0980547919869423,
|
|
"num_tokens": 790497.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.548171281814575,
|
|
"epoch": 0.021539252877258407,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9457,
|
|
"mean_token_accuracy": 0.10034449025988579,
|
|
"num_tokens": 803021.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.514672136306762,
|
|
"epoch": 0.021860734263486146,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.8973,
|
|
"mean_token_accuracy": 0.10233017727732659,
|
|
"num_tokens": 815056.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.548324632644653,
|
|
"epoch": 0.02218221564971388,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8436,
|
|
"mean_token_accuracy": 0.11048155352473259,
|
|
"num_tokens": 827149.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.47729377746582,
|
|
"epoch": 0.022503697035941617,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8877,
|
|
"mean_token_accuracy": 0.10872978940606118,
|
|
"num_tokens": 839172.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.497987508773804,
|
|
"epoch": 0.022825178422169357,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9329,
|
|
"mean_token_accuracy": 0.10738308876752853,
|
|
"num_tokens": 851148.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.548883295059204,
|
|
"epoch": 0.023146659808397092,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8783,
|
|
"mean_token_accuracy": 0.10820900201797486,
|
|
"num_tokens": 862904.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.436304187774658,
|
|
"epoch": 0.02346814119462483,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.7668,
|
|
"mean_token_accuracy": 0.11688080206513404,
|
|
"num_tokens": 874720.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.409350728988647,
|
|
"epoch": 0.023789622580852567,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.8078,
|
|
"mean_token_accuracy": 0.11283540055155754,
|
|
"num_tokens": 887261.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.4675617694854735,
|
|
"epoch": 0.024111103967080307,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7767,
|
|
"mean_token_accuracy": 0.1088891163468361,
|
|
"num_tokens": 898048.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.336610507965088,
|
|
"epoch": 0.024432585353308042,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.8343,
|
|
"mean_token_accuracy": 0.11030351296067238,
|
|
"num_tokens": 910297.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.432262849807739,
|
|
"epoch": 0.02475406673953578,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8693,
|
|
"mean_token_accuracy": 0.10626164525747299,
|
|
"num_tokens": 923463.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.399855327606201,
|
|
"epoch": 0.025075548125763517,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8379,
|
|
"mean_token_accuracy": 0.11091364100575447,
|
|
"num_tokens": 935306.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.355379390716553,
|
|
"epoch": 0.025397029511991256,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7379,
|
|
"mean_token_accuracy": 0.115155877918005,
|
|
"num_tokens": 947827.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.281255531311035,
|
|
"epoch": 0.025718510898218992,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.6838,
|
|
"mean_token_accuracy": 0.12000245451927186,
|
|
"num_tokens": 961664.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.312869071960449,
|
|
"epoch": 0.02603999228444673,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7829,
|
|
"mean_token_accuracy": 0.09842887446284294,
|
|
"num_tokens": 973981.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.335048866271973,
|
|
"epoch": 0.026361473670674467,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8181,
|
|
"mean_token_accuracy": 0.11232979372143745,
|
|
"num_tokens": 986702.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.318604183197022,
|
|
"epoch": 0.026682955056902206,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7181,
|
|
"mean_token_accuracy": 0.11695454865694047,
|
|
"num_tokens": 998513.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.294521141052246,
|
|
"epoch": 0.027004436443129942,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7037,
|
|
"mean_token_accuracy": 0.11898418515920639,
|
|
"num_tokens": 1011111.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.186019515991211,
|
|
"epoch": 0.02732591782935768,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6799,
|
|
"mean_token_accuracy": 0.11566238775849343,
|
|
"num_tokens": 1021779.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.370555067062378,
|
|
"epoch": 0.027647399215585417,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7381,
|
|
"mean_token_accuracy": 0.11263754442334176,
|
|
"num_tokens": 1033228.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.226527881622315,
|
|
"epoch": 0.027968880601813156,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6415,
|
|
"mean_token_accuracy": 0.11632440984249115,
|
|
"num_tokens": 1044964.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.18967342376709,
|
|
"epoch": 0.028290361988040892,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6058,
|
|
"mean_token_accuracy": 0.12256524711847305,
|
|
"num_tokens": 1056450.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.242304182052612,
|
|
"epoch": 0.02861184337426863,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7402,
|
|
"mean_token_accuracy": 0.1159943200647831,
|
|
"num_tokens": 1067031.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.1245410442352295,
|
|
"epoch": 0.028933324760496367,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6051,
|
|
"mean_token_accuracy": 0.1228414848446846,
|
|
"num_tokens": 1078718.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.182725429534912,
|
|
"epoch": 0.029254806146724106,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.6217,
|
|
"mean_token_accuracy": 0.11590910404920578,
|
|
"num_tokens": 1091630.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.229373598098755,
|
|
"epoch": 0.029576287532951842,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6761,
|
|
"mean_token_accuracy": 0.11919770017266273,
|
|
"num_tokens": 1102339.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.166626882553101,
|
|
"epoch": 0.029897768919179578,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.6491,
|
|
"mean_token_accuracy": 0.11653751060366631,
|
|
"num_tokens": 1114385.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.213460397720337,
|
|
"epoch": 0.030219250305407317,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6627,
|
|
"mean_token_accuracy": 0.12240460664033889,
|
|
"num_tokens": 1126364.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.160998678207397,
|
|
"epoch": 0.030540731691635053,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6469,
|
|
"mean_token_accuracy": 0.12170583978295327,
|
|
"num_tokens": 1137492.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.108113145828247,
|
|
"epoch": 0.030862213077862792,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.5317,
|
|
"mean_token_accuracy": 0.12732059210538865,
|
|
"num_tokens": 1148734.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.14542589187622,
|
|
"epoch": 0.031183694464090528,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6757,
|
|
"mean_token_accuracy": 0.12211950346827508,
|
|
"num_tokens": 1163229.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.165257167816162,
|
|
"epoch": 0.031505175850318264,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6959,
|
|
"mean_token_accuracy": 0.1155033528804779,
|
|
"num_tokens": 1175452.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.1013343334198,
|
|
"epoch": 0.031826657236546006,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5479,
|
|
"mean_token_accuracy": 0.12406643405556679,
|
|
"num_tokens": 1187836.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.197257661819458,
|
|
"epoch": 0.03214813862277374,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6297,
|
|
"mean_token_accuracy": 0.12298163026571274,
|
|
"num_tokens": 1198867.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.00099925994873,
|
|
"epoch": 0.03246962000900148,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.5666,
|
|
"mean_token_accuracy": 0.11886920258402825,
|
|
"num_tokens": 1212534.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.123233509063721,
|
|
"epoch": 0.032791101395229214,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.4812,
|
|
"mean_token_accuracy": 0.1376668132841587,
|
|
"num_tokens": 1223028.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.040684223175049,
|
|
"epoch": 0.033112582781456956,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5438,
|
|
"mean_token_accuracy": 0.12737778946757317,
|
|
"num_tokens": 1235460.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.9980145454406735,
|
|
"epoch": 0.03343406416768469,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5329,
|
|
"mean_token_accuracy": 0.1297763057053089,
|
|
"num_tokens": 1247180.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.0526275634765625,
|
|
"epoch": 0.03375554555391243,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5934,
|
|
"mean_token_accuracy": 0.12331821173429489,
|
|
"num_tokens": 1260000.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.997141313552857,
|
|
"epoch": 0.034077026940140163,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.4284,
|
|
"mean_token_accuracy": 0.12857622653245926,
|
|
"num_tokens": 1271346.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.008873414993286,
|
|
"epoch": 0.034398508326367906,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.5163,
|
|
"mean_token_accuracy": 0.12632984891533852,
|
|
"num_tokens": 1283017.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.0776612758636475,
|
|
"epoch": 0.03471998971259564,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5481,
|
|
"mean_token_accuracy": 0.12498711422085762,
|
|
"num_tokens": 1293243.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.8785299301147464,
|
|
"epoch": 0.03504147109882338,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.439,
|
|
"mean_token_accuracy": 0.134949841350317,
|
|
"num_tokens": 1304801.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.143336582183838,
|
|
"epoch": 0.03536295248505111,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6164,
|
|
"mean_token_accuracy": 0.12681611329317094,
|
|
"num_tokens": 1315985.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.930103540420532,
|
|
"epoch": 0.035684433871278856,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4575,
|
|
"mean_token_accuracy": 0.13166105449199678,
|
|
"num_tokens": 1327771.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.966667604446411,
|
|
"epoch": 0.03600591525750659,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.412,
|
|
"mean_token_accuracy": 0.13625267744064332,
|
|
"num_tokens": 1338524.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.932982063293457,
|
|
"epoch": 0.03632739664373433,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.4579,
|
|
"mean_token_accuracy": 0.12851827815175057,
|
|
"num_tokens": 1350619.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.9319816589355465,
|
|
"epoch": 0.03664887802996206,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.4918,
|
|
"mean_token_accuracy": 0.12711360901594163,
|
|
"num_tokens": 1362781.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.932379579544067,
|
|
"epoch": 0.0369703594161898,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.4437,
|
|
"mean_token_accuracy": 0.1318575732409954,
|
|
"num_tokens": 1373784.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.074356174468994,
|
|
"epoch": 0.03729184080241754,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.705,
|
|
"mean_token_accuracy": 0.11901014372706413,
|
|
"num_tokens": 1386426.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.872195100784301,
|
|
"epoch": 0.03761332218864528,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.3845,
|
|
"mean_token_accuracy": 0.13377587869763374,
|
|
"num_tokens": 1397703.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.900994157791137,
|
|
"epoch": 0.03793480357487301,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.4395,
|
|
"mean_token_accuracy": 0.12795801013708114,
|
|
"num_tokens": 1410071.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.954240083694458,
|
|
"epoch": 0.03825628496110075,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.4338,
|
|
"mean_token_accuracy": 0.1245950624346733,
|
|
"num_tokens": 1421387.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.907541370391845,
|
|
"epoch": 0.03857776634732849,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4593,
|
|
"mean_token_accuracy": 0.12624357938766478,
|
|
"num_tokens": 1433253.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.899614143371582,
|
|
"epoch": 0.03889924773355623,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4459,
|
|
"mean_token_accuracy": 0.12114612162113189,
|
|
"num_tokens": 1444189.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.889805459976197,
|
|
"epoch": 0.03922072911978396,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.4478,
|
|
"mean_token_accuracy": 0.12658536732196807,
|
|
"num_tokens": 1455830.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.883515930175781,
|
|
"epoch": 0.0395422105060117,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.474,
|
|
"mean_token_accuracy": 0.13375141024589537,
|
|
"num_tokens": 1467442.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.902092170715332,
|
|
"epoch": 0.03986369189223944,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.378,
|
|
"mean_token_accuracy": 0.1375095695257187,
|
|
"num_tokens": 1478113.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.887974262237549,
|
|
"epoch": 0.04018517327846718,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4173,
|
|
"mean_token_accuracy": 0.1352170430123806,
|
|
"num_tokens": 1490097.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.751322174072266,
|
|
"epoch": 0.04050665466469491,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3863,
|
|
"mean_token_accuracy": 0.13129702284932138,
|
|
"num_tokens": 1501565.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.841671752929687,
|
|
"epoch": 0.04082813605092265,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.385,
|
|
"mean_token_accuracy": 0.13331395909190177,
|
|
"num_tokens": 1512434.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.838632202148437,
|
|
"epoch": 0.04114961743715039,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.3326,
|
|
"mean_token_accuracy": 0.13992721661925317,
|
|
"num_tokens": 1524212.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.717573928833008,
|
|
"epoch": 0.04147109882337813,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.2641,
|
|
"mean_token_accuracy": 0.13435597494244575,
|
|
"num_tokens": 1536588.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.813121795654297,
|
|
"epoch": 0.04179258020960586,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4067,
|
|
"mean_token_accuracy": 0.1357954926788807,
|
|
"num_tokens": 1548334.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.733175325393677,
|
|
"epoch": 0.0421140615958336,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.3691,
|
|
"mean_token_accuracy": 0.13718299865722655,
|
|
"num_tokens": 1560381.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.773969268798828,
|
|
"epoch": 0.04243554298206134,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3926,
|
|
"mean_token_accuracy": 0.1230570524930954,
|
|
"num_tokens": 1573484.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.820370149612427,
|
|
"epoch": 0.04275702436828908,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.3808,
|
|
"mean_token_accuracy": 0.13517048284411431,
|
|
"num_tokens": 1585471.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.72473177909851,
|
|
"epoch": 0.04307850575451681,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.2612,
|
|
"mean_token_accuracy": 0.14530360400676728,
|
|
"num_tokens": 1596321.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.86566572189331,
|
|
"epoch": 0.04339998714074455,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4255,
|
|
"mean_token_accuracy": 0.1406514436006546,
|
|
"num_tokens": 1608219.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.6760951519012455,
|
|
"epoch": 0.04372146852697229,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.2714,
|
|
"mean_token_accuracy": 0.13934220522642135,
|
|
"num_tokens": 1619420.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.79856915473938,
|
|
"epoch": 0.04404294991320003,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.3423,
|
|
"mean_token_accuracy": 0.13619581386446952,
|
|
"num_tokens": 1631022.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.7899799823760985,
|
|
"epoch": 0.04436443129942776,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3661,
|
|
"mean_token_accuracy": 0.13836024552583695,
|
|
"num_tokens": 1642930.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.691325092315674,
|
|
"epoch": 0.0446859126856555,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.3574,
|
|
"mean_token_accuracy": 0.1367586337029934,
|
|
"num_tokens": 1655776.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.817952728271484,
|
|
"epoch": 0.045007394071883235,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.361,
|
|
"mean_token_accuracy": 0.1369588740170002,
|
|
"num_tokens": 1669330.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.626900386810303,
|
|
"epoch": 0.04532887545811098,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3733,
|
|
"mean_token_accuracy": 0.1379777029156685,
|
|
"num_tokens": 1682126.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.74940242767334,
|
|
"epoch": 0.04565035684433871,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.2913,
|
|
"mean_token_accuracy": 0.14252828136086465,
|
|
"num_tokens": 1693070.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.821109485626221,
|
|
"epoch": 0.04597183823056645,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.3673,
|
|
"mean_token_accuracy": 0.13518087714910507,
|
|
"num_tokens": 1705254.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.716013097763062,
|
|
"epoch": 0.046293319616794185,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.3296,
|
|
"mean_token_accuracy": 0.1330641709268093,
|
|
"num_tokens": 1718214.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.745158529281616,
|
|
"epoch": 0.04661480100302193,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.36,
|
|
"mean_token_accuracy": 0.13509849980473518,
|
|
"num_tokens": 1731046.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.691292095184326,
|
|
"epoch": 0.04693628238924966,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3051,
|
|
"mean_token_accuracy": 0.13931626304984093,
|
|
"num_tokens": 1744332.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.6679082870483395,
|
|
"epoch": 0.0472577637754774,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.3141,
|
|
"mean_token_accuracy": 0.1361882694065571,
|
|
"num_tokens": 1757744.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.755461311340332,
|
|
"epoch": 0.047579245161705135,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3255,
|
|
"mean_token_accuracy": 0.13589167818427086,
|
|
"num_tokens": 1769710.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.661909294128418,
|
|
"epoch": 0.04790072654793288,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3197,
|
|
"mean_token_accuracy": 0.134683046489954,
|
|
"num_tokens": 1782299.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.686239433288574,
|
|
"epoch": 0.04822220793416061,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2854,
|
|
"mean_token_accuracy": 0.1418588526546955,
|
|
"num_tokens": 1794407.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.681186580657959,
|
|
"epoch": 0.04854368932038835,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.2539,
|
|
"mean_token_accuracy": 0.13846987187862397,
|
|
"num_tokens": 1806512.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.679962110519409,
|
|
"epoch": 0.048865170706616085,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.2985,
|
|
"mean_token_accuracy": 0.14243988171219826,
|
|
"num_tokens": 1819456.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.526072311401367,
|
|
"epoch": 0.04918665209284383,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.1211,
|
|
"mean_token_accuracy": 0.14557768478989602,
|
|
"num_tokens": 1831866.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.57554988861084,
|
|
"epoch": 0.04950813347907156,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.141,
|
|
"mean_token_accuracy": 0.14539860635995866,
|
|
"num_tokens": 1842662.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.579600238800049,
|
|
"epoch": 0.0498296148652993,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.1807,
|
|
"mean_token_accuracy": 0.14677832573652266,
|
|
"num_tokens": 1855599.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.664310932159424,
|
|
"epoch": 0.050151096251527034,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.2611,
|
|
"mean_token_accuracy": 0.13455238118767737,
|
|
"num_tokens": 1866656.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.580622625350952,
|
|
"epoch": 0.05047257763775478,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2491,
|
|
"mean_token_accuracy": 0.13729645386338235,
|
|
"num_tokens": 1877473.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.699300527572632,
|
|
"epoch": 0.05079405902398251,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.2869,
|
|
"mean_token_accuracy": 0.13934579417109488,
|
|
"num_tokens": 1889113.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.575865650177002,
|
|
"epoch": 0.05111554041021025,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.1885,
|
|
"mean_token_accuracy": 0.1417413368821144,
|
|
"num_tokens": 1901528.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.590616846084595,
|
|
"epoch": 0.051437021796437984,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.2219,
|
|
"mean_token_accuracy": 0.13816025704145432,
|
|
"num_tokens": 1912036.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.601884841918945,
|
|
"epoch": 0.05175850318266572,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.1707,
|
|
"mean_token_accuracy": 0.14483692795038222,
|
|
"num_tokens": 1923939.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.482674884796142,
|
|
"epoch": 0.05207998456889346,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.1383,
|
|
"mean_token_accuracy": 0.15191132128238677,
|
|
"num_tokens": 1935248.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.679584169387818,
|
|
"epoch": 0.0524014659551212,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.2512,
|
|
"mean_token_accuracy": 0.1460711881518364,
|
|
"num_tokens": 1946732.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.574809503555298,
|
|
"epoch": 0.052722947341348934,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.226,
|
|
"mean_token_accuracy": 0.1427159160375595,
|
|
"num_tokens": 1958316.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.472147607803345,
|
|
"epoch": 0.05304442872757667,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.0917,
|
|
"mean_token_accuracy": 0.15051717907190323,
|
|
"num_tokens": 1969543.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.623300170898437,
|
|
"epoch": 0.05336591011380441,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.313,
|
|
"mean_token_accuracy": 0.1366821512579918,
|
|
"num_tokens": 1981085.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.603462076187133,
|
|
"epoch": 0.05368739150003215,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.1986,
|
|
"mean_token_accuracy": 0.14714324995875358,
|
|
"num_tokens": 1992610.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.533374309539795,
|
|
"epoch": 0.054008872886259884,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.1716,
|
|
"mean_token_accuracy": 0.14295029491186143,
|
|
"num_tokens": 2004154.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.516127395629883,
|
|
"epoch": 0.05433035427248762,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.0815,
|
|
"mean_token_accuracy": 0.14348058253526688,
|
|
"num_tokens": 2015141.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.5279299259185795,
|
|
"epoch": 0.05465183565871536,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.1455,
|
|
"mean_token_accuracy": 0.15023983269929886,
|
|
"num_tokens": 2028173.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.429127645492554,
|
|
"epoch": 0.0549733170449431,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.1486,
|
|
"mean_token_accuracy": 0.14123030006885529,
|
|
"num_tokens": 2039979.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.57825779914856,
|
|
"epoch": 0.055294798431170834,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.0981,
|
|
"mean_token_accuracy": 0.14770488440990448,
|
|
"num_tokens": 2051052.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.48339581489563,
|
|
"epoch": 0.05561627981739857,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.2736,
|
|
"mean_token_accuracy": 0.13757081255316733,
|
|
"num_tokens": 2064252.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.527880430221558,
|
|
"epoch": 0.05593776120362631,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.1659,
|
|
"mean_token_accuracy": 0.14587004482746124,
|
|
"num_tokens": 2075697.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.587885761260987,
|
|
"epoch": 0.05625924258985405,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2902,
|
|
"mean_token_accuracy": 0.1394598327577114,
|
|
"num_tokens": 2088354.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.597920179367065,
|
|
"epoch": 0.056580723976081784,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2794,
|
|
"mean_token_accuracy": 0.14314963445067405,
|
|
"num_tokens": 2101291.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.531395626068115,
|
|
"epoch": 0.05690220536230952,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.1794,
|
|
"mean_token_accuracy": 0.15109536275267602,
|
|
"num_tokens": 2114151.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.5403975486755375,
|
|
"epoch": 0.05722368674853726,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2204,
|
|
"mean_token_accuracy": 0.140879013389349,
|
|
"num_tokens": 2127638.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.454720592498779,
|
|
"epoch": 0.057545168134765,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.1054,
|
|
"mean_token_accuracy": 0.15080050081014634,
|
|
"num_tokens": 2139773.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.582310771942138,
|
|
"epoch": 0.057866649520992734,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2261,
|
|
"mean_token_accuracy": 0.13719287440180777,
|
|
"num_tokens": 2151167.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.373457574844361,
|
|
"epoch": 0.05818813090722047,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.1103,
|
|
"mean_token_accuracy": 0.14882433116436006,
|
|
"num_tokens": 2163150.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.511533308029175,
|
|
"epoch": 0.05850961229344821,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.138,
|
|
"mean_token_accuracy": 0.1470898576080799,
|
|
"num_tokens": 2175116.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.402334594726563,
|
|
"epoch": 0.05883109367967595,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.0541,
|
|
"mean_token_accuracy": 0.1523958332836628,
|
|
"num_tokens": 2188074.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.463513612747192,
|
|
"epoch": 0.059152575065903684,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.1002,
|
|
"mean_token_accuracy": 0.1470402017235756,
|
|
"num_tokens": 2200706.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.349210262298584,
|
|
"epoch": 0.05947405645213142,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.1154,
|
|
"mean_token_accuracy": 0.15224614143371581,
|
|
"num_tokens": 2212493.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.507826137542724,
|
|
"epoch": 0.059795537838359156,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1258,
|
|
"mean_token_accuracy": 0.14575668349862098,
|
|
"num_tokens": 2224668.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.432238912582397,
|
|
"epoch": 0.0601170192245869,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1242,
|
|
"mean_token_accuracy": 0.14432022348046303,
|
|
"num_tokens": 2236280.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.493103885650635,
|
|
"epoch": 0.060438500610814634,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1341,
|
|
"mean_token_accuracy": 0.14489658921957016,
|
|
"num_tokens": 2246947.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.486361408233643,
|
|
"epoch": 0.06075998199704237,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.1497,
|
|
"mean_token_accuracy": 0.1474609225988388,
|
|
"num_tokens": 2258591.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.394432783126831,
|
|
"epoch": 0.061081463383270106,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1055,
|
|
"mean_token_accuracy": 0.1405967153608799,
|
|
"num_tokens": 2269725.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.481177186965942,
|
|
"epoch": 0.06140294476949785,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.0784,
|
|
"mean_token_accuracy": 0.1547504723072052,
|
|
"num_tokens": 2282172.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.446505546569824,
|
|
"epoch": 0.061724426155725584,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.2106,
|
|
"mean_token_accuracy": 0.14404927641153337,
|
|
"num_tokens": 2293991.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.479608154296875,
|
|
"epoch": 0.06204590754195332,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1149,
|
|
"mean_token_accuracy": 0.1508398488163948,
|
|
"num_tokens": 2306715.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.330813074111939,
|
|
"epoch": 0.062367388928181056,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1465,
|
|
"mean_token_accuracy": 0.14545181691646575,
|
|
"num_tokens": 2319096.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.360287046432495,
|
|
"epoch": 0.0626888703144088,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.0329,
|
|
"mean_token_accuracy": 0.14853936582803726,
|
|
"num_tokens": 2332239.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.3763471126556395,
|
|
"epoch": 0.06301035170063653,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.1228,
|
|
"mean_token_accuracy": 0.14723728373646736,
|
|
"num_tokens": 2343897.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.353127479553223,
|
|
"epoch": 0.06333183308686427,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0663,
|
|
"mean_token_accuracy": 0.14970697611570358,
|
|
"num_tokens": 2356811.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.3828963279724125,
|
|
"epoch": 0.06365331447309201,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.0559,
|
|
"mean_token_accuracy": 0.15341916903853417,
|
|
"num_tokens": 2369146.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.346980619430542,
|
|
"epoch": 0.06397479585931974,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.0164,
|
|
"mean_token_accuracy": 0.1514612004160881,
|
|
"num_tokens": 2381154.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.36736650466919,
|
|
"epoch": 0.06429627724554748,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.1074,
|
|
"mean_token_accuracy": 0.14672497659921646,
|
|
"num_tokens": 2392095.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.426885080337525,
|
|
"epoch": 0.06461775863177523,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.2126,
|
|
"mean_token_accuracy": 0.14347948506474495,
|
|
"num_tokens": 2405446.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.350338315963745,
|
|
"epoch": 0.06493924001800296,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.0627,
|
|
"mean_token_accuracy": 0.1458252727985382,
|
|
"num_tokens": 2418506.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.34899787902832,
|
|
"epoch": 0.0652607214042307,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.0568,
|
|
"mean_token_accuracy": 0.15567989647388458,
|
|
"num_tokens": 2430125.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.37267050743103,
|
|
"epoch": 0.06558220279045843,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.0766,
|
|
"mean_token_accuracy": 0.14890180379152299,
|
|
"num_tokens": 2442000.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.391686868667603,
|
|
"epoch": 0.06590368417668617,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.095,
|
|
"mean_token_accuracy": 0.1505647674202919,
|
|
"num_tokens": 2453389.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.414132595062256,
|
|
"epoch": 0.06622516556291391,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.1065,
|
|
"mean_token_accuracy": 0.1472689539194107,
|
|
"num_tokens": 2464988.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.3225417137146,
|
|
"epoch": 0.06654664694914164,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.0009,
|
|
"mean_token_accuracy": 0.15887372940778732,
|
|
"num_tokens": 2476820.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.353922176361084,
|
|
"epoch": 0.06686812833536938,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.0069,
|
|
"mean_token_accuracy": 0.14979753121733666,
|
|
"num_tokens": 2489427.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.2835996627807615,
|
|
"epoch": 0.06718960972159711,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.0293,
|
|
"mean_token_accuracy": 0.15290547013282776,
|
|
"num_tokens": 2501084.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.393969631195068,
|
|
"epoch": 0.06751109110782486,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.1253,
|
|
"mean_token_accuracy": 0.1501571610569954,
|
|
"num_tokens": 2513005.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.297085285186768,
|
|
"epoch": 0.0678325724940526,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 5.9784,
|
|
"mean_token_accuracy": 0.15144816786050797,
|
|
"num_tokens": 2526217.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.364409065246582,
|
|
"epoch": 0.06815405388028033,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.0721,
|
|
"mean_token_accuracy": 0.15043441355228424,
|
|
"num_tokens": 2538892.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.305640745162964,
|
|
"epoch": 0.06847553526650807,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.0322,
|
|
"mean_token_accuracy": 0.15073884129524232,
|
|
"num_tokens": 2549929.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.316273498535156,
|
|
"epoch": 0.06879701665273581,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 5.967,
|
|
"mean_token_accuracy": 0.16290880739688873,
|
|
"num_tokens": 2561322.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.35417103767395,
|
|
"epoch": 0.06911849803896354,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 5.9635,
|
|
"mean_token_accuracy": 0.16284787505865098,
|
|
"num_tokens": 2572908.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.189544916152954,
|
|
"epoch": 0.06943997942519128,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.0468,
|
|
"mean_token_accuracy": 0.14753600358963012,
|
|
"num_tokens": 2584880.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.304679203033447,
|
|
"epoch": 0.06976146081141901,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 5.9537,
|
|
"mean_token_accuracy": 0.15288592427968978,
|
|
"num_tokens": 2596663.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.306663751602173,
|
|
"epoch": 0.07008294219764676,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.0103,
|
|
"mean_token_accuracy": 0.150350858271122,
|
|
"num_tokens": 2608609.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.189498329162598,
|
|
"epoch": 0.0704044235838745,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 5.9271,
|
|
"mean_token_accuracy": 0.15373560041189194,
|
|
"num_tokens": 2620535.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.272855806350708,
|
|
"epoch": 0.07072590497010223,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.0391,
|
|
"mean_token_accuracy": 0.14719461053609847,
|
|
"num_tokens": 2631718.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.330906057357788,
|
|
"epoch": 0.07104738635632997,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 5.9971,
|
|
"mean_token_accuracy": 0.1513780727982521,
|
|
"num_tokens": 2644641.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.242729377746582,
|
|
"epoch": 0.07136886774255771,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.0712,
|
|
"mean_token_accuracy": 0.1472596064209938,
|
|
"num_tokens": 2656445.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.2387518882751465,
|
|
"epoch": 0.07169034912878544,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 5.9288,
|
|
"mean_token_accuracy": 0.16197773665189744,
|
|
"num_tokens": 2667782.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.297261524200439,
|
|
"epoch": 0.07201183051501318,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 5.8888,
|
|
"mean_token_accuracy": 0.16516102403402327,
|
|
"num_tokens": 2679421.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.153607034683228,
|
|
"epoch": 0.07233331190124091,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 5.8252,
|
|
"mean_token_accuracy": 0.16737353205680847,
|
|
"num_tokens": 2690045.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.256674098968506,
|
|
"epoch": 0.07265479328746866,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 5.9358,
|
|
"mean_token_accuracy": 0.1594891406595707,
|
|
"num_tokens": 2702725.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.189019870758057,
|
|
"epoch": 0.0729762746736964,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 5.9325,
|
|
"mean_token_accuracy": 0.1560269773006439,
|
|
"num_tokens": 2714874.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.31310076713562,
|
|
"epoch": 0.07329775605992413,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.0494,
|
|
"mean_token_accuracy": 0.14427177235484123,
|
|
"num_tokens": 2727237.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.140863609313965,
|
|
"epoch": 0.07361923744615187,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 5.9221,
|
|
"mean_token_accuracy": 0.15012870132923126,
|
|
"num_tokens": 2738201.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.239338064193726,
|
|
"epoch": 0.0739407188323796,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 5.967,
|
|
"mean_token_accuracy": 0.161196768283844,
|
|
"num_tokens": 2749926.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.244922113418579,
|
|
"epoch": 0.07426220021860734,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 5.9984,
|
|
"mean_token_accuracy": 0.146126826107502,
|
|
"num_tokens": 2762454.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.172313117980957,
|
|
"epoch": 0.07458368160483508,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 5.928,
|
|
"mean_token_accuracy": 0.1524214893579483,
|
|
"num_tokens": 2775340.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.191201305389404,
|
|
"epoch": 0.07490516299106281,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 5.8407,
|
|
"mean_token_accuracy": 0.1651138797402382,
|
|
"num_tokens": 2787350.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.163157844543457,
|
|
"epoch": 0.07522664437729056,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 5.8869,
|
|
"mean_token_accuracy": 0.16123317033052445,
|
|
"num_tokens": 2797891.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.166656732559204,
|
|
"epoch": 0.0755481257635183,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 5.8488,
|
|
"mean_token_accuracy": 0.15674188137054443,
|
|
"num_tokens": 2809199.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.317021608352661,
|
|
"epoch": 0.07586960714974603,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.1259,
|
|
"mean_token_accuracy": 0.14605942070484162,
|
|
"num_tokens": 2820881.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.217242765426636,
|
|
"epoch": 0.07619108853597377,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 5.9131,
|
|
"mean_token_accuracy": 0.158234640955925,
|
|
"num_tokens": 2833203.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.085934734344482,
|
|
"epoch": 0.0765125699222015,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 5.8158,
|
|
"mean_token_accuracy": 0.1622716337442398,
|
|
"num_tokens": 2844695.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.18882040977478,
|
|
"epoch": 0.07683405130842924,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 5.9218,
|
|
"mean_token_accuracy": 0.15580735355615616,
|
|
"num_tokens": 2855757.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.096655941009521,
|
|
"epoch": 0.07715553269465698,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 5.8256,
|
|
"mean_token_accuracy": 0.1585150107741356,
|
|
"num_tokens": 2867534.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.212085294723511,
|
|
"epoch": 0.07747701408088471,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 5.9379,
|
|
"mean_token_accuracy": 0.15673388540744781,
|
|
"num_tokens": 2879984.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.263598108291626,
|
|
"epoch": 0.07779849546711246,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 5.9533,
|
|
"mean_token_accuracy": 0.1572867900133133,
|
|
"num_tokens": 2892920.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.116982269287109,
|
|
"epoch": 0.0781199768533402,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 5.9162,
|
|
"mean_token_accuracy": 0.1617930367588997,
|
|
"num_tokens": 2903821.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.2275710105896,
|
|
"epoch": 0.07844145823956793,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 5.9182,
|
|
"mean_token_accuracy": 0.15855470597743987,
|
|
"num_tokens": 2916280.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.233517837524414,
|
|
"epoch": 0.07876293962579567,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 5.9979,
|
|
"mean_token_accuracy": 0.15871625244617463,
|
|
"num_tokens": 2928737.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.193160438537598,
|
|
"epoch": 0.0790844210120234,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 5.891,
|
|
"mean_token_accuracy": 0.16073769181966782,
|
|
"num_tokens": 2940833.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.115055561065674,
|
|
"epoch": 0.07940590239825114,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 5.8428,
|
|
"mean_token_accuracy": 0.15946858525276184,
|
|
"num_tokens": 2952662.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.204311752319336,
|
|
"epoch": 0.07972738378447888,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 5.9119,
|
|
"mean_token_accuracy": 0.1619669482111931,
|
|
"num_tokens": 2965504.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.2926552295684814,
|
|
"epoch": 0.08004886517070661,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.15522371083498002,
|
|
"num_tokens": 2977610.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.112452220916748,
|
|
"epoch": 0.08037034655693436,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 5.8564,
|
|
"mean_token_accuracy": 0.15934567600488664,
|
|
"num_tokens": 2989677.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.151176643371582,
|
|
"epoch": 0.0806918279431621,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 5.9276,
|
|
"mean_token_accuracy": 0.1602254882454872,
|
|
"num_tokens": 3002164.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.151615953445434,
|
|
"epoch": 0.08101330932938983,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 5.839,
|
|
"mean_token_accuracy": 0.1672609567642212,
|
|
"num_tokens": 3013487.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.091053056716919,
|
|
"epoch": 0.08133479071561757,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 5.7592,
|
|
"mean_token_accuracy": 0.1637980043888092,
|
|
"num_tokens": 3025240.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.198080825805664,
|
|
"epoch": 0.0816562721018453,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 5.9558,
|
|
"mean_token_accuracy": 0.15986837595701217,
|
|
"num_tokens": 3037931.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.180276155471802,
|
|
"epoch": 0.08197775348807304,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 5.8599,
|
|
"mean_token_accuracy": 0.16204103082418442,
|
|
"num_tokens": 3050977.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.0987457752227785,
|
|
"epoch": 0.08229923487430078,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 5.7864,
|
|
"mean_token_accuracy": 0.15847471207380295,
|
|
"num_tokens": 3064797.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.059926891326905,
|
|
"epoch": 0.08262071626052851,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.16135916709899903,
|
|
"num_tokens": 3077595.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.165657234191895,
|
|
"epoch": 0.08294219764675625,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 5.9139,
|
|
"mean_token_accuracy": 0.15417862236499785,
|
|
"num_tokens": 3089913.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.133871126174927,
|
|
"epoch": 0.08326367903298398,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 5.9087,
|
|
"mean_token_accuracy": 0.1627100259065628,
|
|
"num_tokens": 3102554.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.111787605285644,
|
|
"epoch": 0.08358516041921173,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 5.8258,
|
|
"mean_token_accuracy": 0.16240279525518417,
|
|
"num_tokens": 3113511.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.131465053558349,
|
|
"epoch": 0.08390664180543947,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 5.9213,
|
|
"mean_token_accuracy": 0.15791668891906738,
|
|
"num_tokens": 3126829.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.069682979583741,
|
|
"epoch": 0.0842281231916672,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 5.7673,
|
|
"mean_token_accuracy": 0.16573094874620437,
|
|
"num_tokens": 3139836.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.073161935806274,
|
|
"epoch": 0.08454960457789494,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 5.8286,
|
|
"mean_token_accuracy": 0.16145224422216414,
|
|
"num_tokens": 3152047.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.066088485717773,
|
|
"epoch": 0.08487108596412268,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 5.7994,
|
|
"mean_token_accuracy": 0.16172150075435637,
|
|
"num_tokens": 3163501.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.123111963272095,
|
|
"epoch": 0.08519256735035041,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.8752,
|
|
"mean_token_accuracy": 0.16654538810253144,
|
|
"num_tokens": 3175918.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.194814872741699,
|
|
"epoch": 0.08551404873657815,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 5.9495,
|
|
"mean_token_accuracy": 0.14875137954950332,
|
|
"num_tokens": 3187911.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.065830373764038,
|
|
"epoch": 0.08583553012280588,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 5.7759,
|
|
"mean_token_accuracy": 0.16410106271505356,
|
|
"num_tokens": 3200143.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.027771663665772,
|
|
"epoch": 0.08615701150903363,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 5.7746,
|
|
"mean_token_accuracy": 0.1614765614271164,
|
|
"num_tokens": 3211570.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.113853216171265,
|
|
"epoch": 0.08647849289526137,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 5.8393,
|
|
"mean_token_accuracy": 0.15796211659908294,
|
|
"num_tokens": 3222014.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.093673276901245,
|
|
"epoch": 0.0867999742814891,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 5.8008,
|
|
"mean_token_accuracy": 0.16573861241340637,
|
|
"num_tokens": 3233115.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.114884376525879,
|
|
"epoch": 0.08712145566771684,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 5.9123,
|
|
"mean_token_accuracy": 0.16168183982372283,
|
|
"num_tokens": 3244084.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.098208856582642,
|
|
"epoch": 0.08744293705394458,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 5.846,
|
|
"mean_token_accuracy": 0.1593250960111618,
|
|
"num_tokens": 3257476.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.054478931427002,
|
|
"epoch": 0.08776441844017231,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 5.7764,
|
|
"mean_token_accuracy": 0.17061060965061187,
|
|
"num_tokens": 3268067.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.125921297073364,
|
|
"epoch": 0.08808589982640005,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 5.8603,
|
|
"mean_token_accuracy": 0.16385477036237717,
|
|
"num_tokens": 3279097.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.049725770950317,
|
|
"epoch": 0.08840738121262778,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 5.9084,
|
|
"mean_token_accuracy": 0.16008279025554656,
|
|
"num_tokens": 3291198.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.082431793212891,
|
|
"epoch": 0.08872886259885553,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 5.7406,
|
|
"mean_token_accuracy": 0.16886814534664155,
|
|
"num_tokens": 3303316.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.06479926109314,
|
|
"epoch": 0.08905034398508327,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.8514,
|
|
"mean_token_accuracy": 0.16055994108319283,
|
|
"num_tokens": 3316069.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.008887243270874,
|
|
"epoch": 0.089371825371311,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 5.7445,
|
|
"mean_token_accuracy": 0.17007820010185243,
|
|
"num_tokens": 3328657.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.10736780166626,
|
|
"epoch": 0.08969330675753874,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 5.8555,
|
|
"mean_token_accuracy": 0.16169255524873732,
|
|
"num_tokens": 3340486.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.1321416854858395,
|
|
"epoch": 0.09001478814376647,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 5.8723,
|
|
"mean_token_accuracy": 0.16864815205335618,
|
|
"num_tokens": 3350936.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.087703371047974,
|
|
"epoch": 0.09033626952999421,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 5.8601,
|
|
"mean_token_accuracy": 0.16583744436502457,
|
|
"num_tokens": 3362238.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.051072263717652,
|
|
"epoch": 0.09065775091622195,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 5.794,
|
|
"mean_token_accuracy": 0.16016314327716827,
|
|
"num_tokens": 3373818.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.093393230438233,
|
|
"epoch": 0.09097923230244968,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 5.8118,
|
|
"mean_token_accuracy": 0.16103656888008117,
|
|
"num_tokens": 3385241.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.9892784595489506,
|
|
"epoch": 0.09130071368867743,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 5.8093,
|
|
"mean_token_accuracy": 0.15792890265583992,
|
|
"num_tokens": 3398274.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.129896736145019,
|
|
"epoch": 0.09162219507490517,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 5.8595,
|
|
"mean_token_accuracy": 0.16091282665729523,
|
|
"num_tokens": 3411798.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.027444553375244,
|
|
"epoch": 0.0919436764611329,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 5.8085,
|
|
"mean_token_accuracy": 0.1594970703125,
|
|
"num_tokens": 3423646.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.086286878585815,
|
|
"epoch": 0.09226515784736064,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 5.8268,
|
|
"mean_token_accuracy": 0.16200227588415145,
|
|
"num_tokens": 3436255.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.025093269348145,
|
|
"epoch": 0.09258663923358837,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 5.8122,
|
|
"mean_token_accuracy": 0.16047175079584122,
|
|
"num_tokens": 3447724.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.141870403289795,
|
|
"epoch": 0.09290812061981611,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.8945,
|
|
"mean_token_accuracy": 0.15279303789138793,
|
|
"num_tokens": 3460959.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.038381910324096,
|
|
"epoch": 0.09322960200604385,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 5.7696,
|
|
"mean_token_accuracy": 0.16729041934013367,
|
|
"num_tokens": 3472283.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.012684297561646,
|
|
"epoch": 0.09355108339227158,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 5.7736,
|
|
"mean_token_accuracy": 0.1646544575691223,
|
|
"num_tokens": 3484296.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.002734279632568,
|
|
"epoch": 0.09387256477849933,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 5.7357,
|
|
"mean_token_accuracy": 0.16739338636398315,
|
|
"num_tokens": 3496553.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.06849160194397,
|
|
"epoch": 0.09419404616472707,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 5.8064,
|
|
"mean_token_accuracy": 0.16018352508544922,
|
|
"num_tokens": 3508515.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.06675968170166,
|
|
"epoch": 0.0945155275509548,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 5.8216,
|
|
"mean_token_accuracy": 0.16168036088347434,
|
|
"num_tokens": 3520883.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.9559320449829105,
|
|
"epoch": 0.09483700893718254,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 5.7903,
|
|
"mean_token_accuracy": 0.16948920488357544,
|
|
"num_tokens": 3533786.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.1463991641998295,
|
|
"epoch": 0.09515849032341027,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 5.7776,
|
|
"mean_token_accuracy": 0.16706227362155915,
|
|
"num_tokens": 3544573.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.094358777999878,
|
|
"epoch": 0.09547997170963801,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 5.8846,
|
|
"mean_token_accuracy": 0.15878364443778992,
|
|
"num_tokens": 3556148.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.980588340759278,
|
|
"epoch": 0.09580145309586575,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 5.7641,
|
|
"mean_token_accuracy": 0.16096356213092805,
|
|
"num_tokens": 3568047.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.014238119125366,
|
|
"epoch": 0.09612293448209348,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 5.7819,
|
|
"mean_token_accuracy": 0.1614854723215103,
|
|
"num_tokens": 3579864.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.090922594070435,
|
|
"epoch": 0.09644441586832123,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 5.8161,
|
|
"mean_token_accuracy": 0.15966751128435136,
|
|
"num_tokens": 3591602.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.911532735824585,
|
|
"epoch": 0.09676589725454895,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.7131,
|
|
"mean_token_accuracy": 0.16157886236906052,
|
|
"num_tokens": 3603802.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.096415281295776,
|
|
"epoch": 0.0970873786407767,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 5.8251,
|
|
"mean_token_accuracy": 0.16491867303848268,
|
|
"num_tokens": 3614958.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.926008939743042,
|
|
"epoch": 0.09740886002700444,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 5.7093,
|
|
"mean_token_accuracy": 0.16871902197599412,
|
|
"num_tokens": 3626175.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.9772974967956545,
|
|
"epoch": 0.09773034141323217,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 5.6843,
|
|
"mean_token_accuracy": 0.16257247924804688,
|
|
"num_tokens": 3638818.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.0240332126617435,
|
|
"epoch": 0.09805182279945991,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 5.7146,
|
|
"mean_token_accuracy": 0.16350910812616348,
|
|
"num_tokens": 3650838.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.000803804397583,
|
|
"epoch": 0.09837330418568765,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 5.7236,
|
|
"mean_token_accuracy": 0.16991809606552125,
|
|
"num_tokens": 3661879.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.000612354278564,
|
|
"epoch": 0.09869478557191538,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 5.7162,
|
|
"mean_token_accuracy": 0.17283950746059418,
|
|
"num_tokens": 3674464.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.956651830673218,
|
|
"epoch": 0.09901626695814313,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 5.6744,
|
|
"mean_token_accuracy": 0.17085929065942765,
|
|
"num_tokens": 3686065.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.961250972747803,
|
|
"epoch": 0.09933774834437085,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 5.7411,
|
|
"mean_token_accuracy": 0.17531964033842087,
|
|
"num_tokens": 3697946.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.9872818946838375,
|
|
"epoch": 0.0996592297305986,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 5.7024,
|
|
"mean_token_accuracy": 0.16942446082830429,
|
|
"num_tokens": 3710695.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.950470924377441,
|
|
"epoch": 0.09998071111682634,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 5.6825,
|
|
"mean_token_accuracy": 0.17383817434310914,
|
|
"num_tokens": 3721288.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.928533840179443,
|
|
"epoch": 0.10030219250305407,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 5.6365,
|
|
"mean_token_accuracy": 0.16649701148271562,
|
|
"num_tokens": 3733611.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.955244255065918,
|
|
"epoch": 0.10062367388928181,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 5.8012,
|
|
"mean_token_accuracy": 0.16040457487106324,
|
|
"num_tokens": 3745018.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.074891996383667,
|
|
"epoch": 0.10094515527550955,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 5.7903,
|
|
"mean_token_accuracy": 0.16442708522081376,
|
|
"num_tokens": 3755414.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.872460556030274,
|
|
"epoch": 0.10126663666173728,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 5.5877,
|
|
"mean_token_accuracy": 0.17960427105426788,
|
|
"num_tokens": 3767398.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.056284761428833,
|
|
"epoch": 0.10158811804796503,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 5.7858,
|
|
"mean_token_accuracy": 0.16207575798034668,
|
|
"num_tokens": 3779344.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.901925182342529,
|
|
"epoch": 0.10190959943419275,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 5.6982,
|
|
"mean_token_accuracy": 0.16722951531410218,
|
|
"num_tokens": 3791642.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.919206190109253,
|
|
"epoch": 0.1022310808204205,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 5.6483,
|
|
"mean_token_accuracy": 0.16981685012578965,
|
|
"num_tokens": 3803604.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.09901065826416,
|
|
"epoch": 0.10255256220664824,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 5.7711,
|
|
"mean_token_accuracy": 0.16022978574037552,
|
|
"num_tokens": 3815472.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.008738374710083,
|
|
"epoch": 0.10287404359287597,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 5.8329,
|
|
"mean_token_accuracy": 0.1592625081539154,
|
|
"num_tokens": 3827468.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.05591311454773,
|
|
"epoch": 0.10319552497910371,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 5.814,
|
|
"mean_token_accuracy": 0.16187592148780822,
|
|
"num_tokens": 3839532.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.920851564407348,
|
|
"epoch": 0.10351700636533144,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 5.6763,
|
|
"mean_token_accuracy": 0.1719309151172638,
|
|
"num_tokens": 3851907.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.0737604141235355,
|
|
"epoch": 0.10383848775155918,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 5.8084,
|
|
"mean_token_accuracy": 0.1594211921095848,
|
|
"num_tokens": 3864624.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.935598373413086,
|
|
"epoch": 0.10415996913778693,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 5.6913,
|
|
"mean_token_accuracy": 0.1719366803765297,
|
|
"num_tokens": 3876832.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.900327444076538,
|
|
"epoch": 0.10448145052401465,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 5.6533,
|
|
"mean_token_accuracy": 0.1725366458296776,
|
|
"num_tokens": 3888969.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.919269466400147,
|
|
"epoch": 0.1048029319102424,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 5.692,
|
|
"mean_token_accuracy": 0.16689868569374083,
|
|
"num_tokens": 3900634.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.978420782089233,
|
|
"epoch": 0.10512441329647014,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 5.6596,
|
|
"mean_token_accuracy": 0.17244009226560592,
|
|
"num_tokens": 3911993.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.944800853729248,
|
|
"epoch": 0.10544589468269787,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 5.7075,
|
|
"mean_token_accuracy": 0.16923267394304276,
|
|
"num_tokens": 3923959.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.934698915481567,
|
|
"epoch": 0.10576737606892561,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 5.6799,
|
|
"mean_token_accuracy": 0.17092464715242386,
|
|
"num_tokens": 3936868.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.951383352279663,
|
|
"epoch": 0.10608885745515334,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 5.6691,
|
|
"mean_token_accuracy": 0.17145125269889833,
|
|
"num_tokens": 3949777.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.009766578674316,
|
|
"epoch": 0.10641033884138108,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 5.7863,
|
|
"mean_token_accuracy": 0.16202808246016503,
|
|
"num_tokens": 3963913.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.909622049331665,
|
|
"epoch": 0.10673182022760883,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 5.6703,
|
|
"mean_token_accuracy": 0.1712466612458229,
|
|
"num_tokens": 3976187.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.883133316040039,
|
|
"epoch": 0.10705330161383655,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 5.6042,
|
|
"mean_token_accuracy": 0.17793208807706834,
|
|
"num_tokens": 3986324.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.935062217712402,
|
|
"epoch": 0.1073747830000643,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 5.7279,
|
|
"mean_token_accuracy": 0.17077707052230834,
|
|
"num_tokens": 3998347.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.9468677043914795,
|
|
"epoch": 0.10769626438629204,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 5.7379,
|
|
"mean_token_accuracy": 0.1692286878824234,
|
|
"num_tokens": 4010640.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.965192079544067,
|
|
"epoch": 0.10801774577251977,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 5.7021,
|
|
"mean_token_accuracy": 0.16421150118112565,
|
|
"num_tokens": 4023544.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.997580337524414,
|
|
"epoch": 0.10833922715874751,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 5.7932,
|
|
"mean_token_accuracy": 0.16685343831777572,
|
|
"num_tokens": 4035359.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.005310583114624,
|
|
"epoch": 0.10866070854497524,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 5.7503,
|
|
"mean_token_accuracy": 0.16817554086446762,
|
|
"num_tokens": 4047055.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.008270597457885,
|
|
"epoch": 0.10898218993120298,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 5.7602,
|
|
"mean_token_accuracy": 0.16708459556102753,
|
|
"num_tokens": 4058231.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.912751865386963,
|
|
"epoch": 0.10930367131743073,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 5.6617,
|
|
"mean_token_accuracy": 0.17167637795209884,
|
|
"num_tokens": 4071138.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.909320926666259,
|
|
"epoch": 0.10962515270365845,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 5.6095,
|
|
"mean_token_accuracy": 0.1787335529923439,
|
|
"num_tokens": 4084144.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.895288896560669,
|
|
"epoch": 0.1099466340898862,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 5.6396,
|
|
"mean_token_accuracy": 0.1743350476026535,
|
|
"num_tokens": 4095828.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.958630466461182,
|
|
"epoch": 0.11026811547611394,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 5.7617,
|
|
"mean_token_accuracy": 0.1587551474571228,
|
|
"num_tokens": 4108463.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.90827522277832,
|
|
"epoch": 0.11058959686234167,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 5.6133,
|
|
"mean_token_accuracy": 0.17577288746833802,
|
|
"num_tokens": 4120153.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.952684164047241,
|
|
"epoch": 0.11091107824856941,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 5.6548,
|
|
"mean_token_accuracy": 0.17637971192598342,
|
|
"num_tokens": 4131684.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.95789852142334,
|
|
"epoch": 0.11123255963479714,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 5.7584,
|
|
"mean_token_accuracy": 0.16837924271821975,
|
|
"num_tokens": 4142757.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.962955284118652,
|
|
"epoch": 0.11155404102102488,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 5.7158,
|
|
"mean_token_accuracy": 0.16554963290691377,
|
|
"num_tokens": 4153892.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.933376312255859,
|
|
"epoch": 0.11187552240725263,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 5.7288,
|
|
"mean_token_accuracy": 0.16116232872009278,
|
|
"num_tokens": 4166534.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.9422413349151615,
|
|
"epoch": 0.11219700379348035,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 5.6579,
|
|
"mean_token_accuracy": 0.1683451145887375,
|
|
"num_tokens": 4178648.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.848151159286499,
|
|
"epoch": 0.1125184851797081,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 5.6261,
|
|
"mean_token_accuracy": 0.17486759424209594,
|
|
"num_tokens": 4191522.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.953406143188476,
|
|
"epoch": 0.11283996656593583,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 5.6226,
|
|
"mean_token_accuracy": 0.17257331013679506,
|
|
"num_tokens": 4203900.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.861113119125366,
|
|
"epoch": 0.11316144795216357,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 5.5864,
|
|
"mean_token_accuracy": 0.17675792276859284,
|
|
"num_tokens": 4214861.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.804804039001465,
|
|
"epoch": 0.11348292933839131,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 5.5351,
|
|
"mean_token_accuracy": 0.1776354894042015,
|
|
"num_tokens": 4227137.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.933787488937378,
|
|
"epoch": 0.11380441072461904,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 5.7155,
|
|
"mean_token_accuracy": 0.1676717832684517,
|
|
"num_tokens": 4239997.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.958622455596924,
|
|
"epoch": 0.11412589211084678,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 5.661,
|
|
"mean_token_accuracy": 0.16713007688522338,
|
|
"num_tokens": 4252556.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.792713975906372,
|
|
"epoch": 0.11444737349707453,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 5.639,
|
|
"mean_token_accuracy": 0.1729072615504265,
|
|
"num_tokens": 4265702.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.8895658493042,
|
|
"epoch": 0.11476885488330225,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 5.6602,
|
|
"mean_token_accuracy": 0.17446050494909288,
|
|
"num_tokens": 4278244.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.912885761260986,
|
|
"epoch": 0.11509033626953,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 5.6157,
|
|
"mean_token_accuracy": 0.17406022995710374,
|
|
"num_tokens": 4290242.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.925296115875244,
|
|
"epoch": 0.11541181765575773,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 5.8008,
|
|
"mean_token_accuracy": 0.16324599087238312,
|
|
"num_tokens": 4303198.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.972964572906494,
|
|
"epoch": 0.11573329904198547,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 5.7539,
|
|
"mean_token_accuracy": 0.16906532049179077,
|
|
"num_tokens": 4316389.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.867070102691651,
|
|
"epoch": 0.11605478042821321,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 5.6284,
|
|
"mean_token_accuracy": 0.17795921117067337,
|
|
"num_tokens": 4329829.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.836501359939575,
|
|
"epoch": 0.11637626181444094,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 5.5708,
|
|
"mean_token_accuracy": 0.1804552987217903,
|
|
"num_tokens": 4342024.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.882567024230957,
|
|
"epoch": 0.11669774320066868,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 5.6703,
|
|
"mean_token_accuracy": 0.16930529624223709,
|
|
"num_tokens": 4355575.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.800036668777466,
|
|
"epoch": 0.11701922458689643,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 5.5661,
|
|
"mean_token_accuracy": 0.17972640544176102,
|
|
"num_tokens": 4368461.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.918715858459473,
|
|
"epoch": 0.11734070597312415,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 5.6427,
|
|
"mean_token_accuracy": 0.17834917455911636,
|
|
"num_tokens": 4380483.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.868259286880493,
|
|
"epoch": 0.1176621873593519,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 5.6284,
|
|
"mean_token_accuracy": 0.16914028078317642,
|
|
"num_tokens": 4393823.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.887679815292358,
|
|
"epoch": 0.11798366874557963,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 5.6887,
|
|
"mean_token_accuracy": 0.1705163061618805,
|
|
"num_tokens": 4405230.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.95110125541687,
|
|
"epoch": 0.11830515013180737,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 5.626,
|
|
"mean_token_accuracy": 0.17905376702547074,
|
|
"num_tokens": 4416638.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.80819149017334,
|
|
"epoch": 0.11862663151803511,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 5.5459,
|
|
"mean_token_accuracy": 0.1780714675784111,
|
|
"num_tokens": 4428778.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.8609541893005375,
|
|
"epoch": 0.11894811290426284,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 5.5975,
|
|
"mean_token_accuracy": 0.17223136574029924,
|
|
"num_tokens": 4440698.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.850246858596802,
|
|
"epoch": 0.11926959429049058,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 5.6452,
|
|
"mean_token_accuracy": 0.17951953709125518,
|
|
"num_tokens": 4451867.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.898460865020752,
|
|
"epoch": 0.11959107567671831,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 5.6622,
|
|
"mean_token_accuracy": 0.16997012495994568,
|
|
"num_tokens": 4464111.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.0535829067230225,
|
|
"epoch": 0.11991255706294605,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 5.8647,
|
|
"mean_token_accuracy": 0.1605095535516739,
|
|
"num_tokens": 4476378.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.918209362030029,
|
|
"epoch": 0.1202340384491738,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 5.6126,
|
|
"mean_token_accuracy": 0.17721273005008698,
|
|
"num_tokens": 4488714.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.869247961044311,
|
|
"epoch": 0.12055551983540153,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 5.6413,
|
|
"mean_token_accuracy": 0.17850209176540374,
|
|
"num_tokens": 4500168.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.9244630336761475,
|
|
"epoch": 0.12087700122162927,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 5.6474,
|
|
"mean_token_accuracy": 0.17328109741210937,
|
|
"num_tokens": 4511792.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.839860105514527,
|
|
"epoch": 0.12119848260785701,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 5.5932,
|
|
"mean_token_accuracy": 0.18388805836439132,
|
|
"num_tokens": 4522314.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.965823554992676,
|
|
"epoch": 0.12151996399408474,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 5.5762,
|
|
"mean_token_accuracy": 0.16557498574256896,
|
|
"num_tokens": 4534827.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.761017513275147,
|
|
"epoch": 0.12184144538031248,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 5.4789,
|
|
"mean_token_accuracy": 0.17830830216407775,
|
|
"num_tokens": 4546730.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.70470962524414,
|
|
"epoch": 0.12216292676654021,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 5.5623,
|
|
"mean_token_accuracy": 0.1807377517223358,
|
|
"num_tokens": 4557573.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.9006062030792235,
|
|
"epoch": 0.12248440815276795,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 5.6343,
|
|
"mean_token_accuracy": 0.174290831387043,
|
|
"num_tokens": 4570238.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.864686441421509,
|
|
"epoch": 0.1228058895389957,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 5.6814,
|
|
"mean_token_accuracy": 0.1700124755501747,
|
|
"num_tokens": 4583007.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.936568975448608,
|
|
"epoch": 0.12312737092522343,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 5.6154,
|
|
"mean_token_accuracy": 0.16709319800138472,
|
|
"num_tokens": 4594224.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.852287435531617,
|
|
"epoch": 0.12344885231145117,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 5.6582,
|
|
"mean_token_accuracy": 0.17214078456163406,
|
|
"num_tokens": 4606092.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.904173707962036,
|
|
"epoch": 0.12377033369767891,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 5.6247,
|
|
"mean_token_accuracy": 0.17527548372745513,
|
|
"num_tokens": 4616751.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.864779424667359,
|
|
"epoch": 0.12409181508390664,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 5.647,
|
|
"mean_token_accuracy": 0.17503723949193956,
|
|
"num_tokens": 4629497.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.880205869674683,
|
|
"epoch": 0.12441329647013438,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 5.625,
|
|
"mean_token_accuracy": 0.17394295632839202,
|
|
"num_tokens": 4641070.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.814072561264038,
|
|
"epoch": 0.12473477785636211,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 5.602,
|
|
"mean_token_accuracy": 0.1692856341600418,
|
|
"num_tokens": 4653813.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.759505414962769,
|
|
"epoch": 0.12505625924258987,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 5.56,
|
|
"mean_token_accuracy": 0.1789846494793892,
|
|
"num_tokens": 4665865.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.860892677307129,
|
|
"epoch": 0.1253777406288176,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 5.6134,
|
|
"mean_token_accuracy": 0.17644744664430617,
|
|
"num_tokens": 4678841.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.835351991653442,
|
|
"epoch": 0.12569922201504533,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 5.569,
|
|
"mean_token_accuracy": 0.17764328867197038,
|
|
"num_tokens": 4691204.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.906078720092774,
|
|
"epoch": 0.12602070340127305,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 5.5548,
|
|
"mean_token_accuracy": 0.18006090223789215,
|
|
"num_tokens": 4702379.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.812206029891968,
|
|
"epoch": 0.1263421847875008,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 5.5982,
|
|
"mean_token_accuracy": 0.1708923801779747,
|
|
"num_tokens": 4714737.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.948781299591064,
|
|
"epoch": 0.12666366617372854,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.6617,
|
|
"mean_token_accuracy": 0.16960146874189377,
|
|
"num_tokens": 4726653.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.826849174499512,
|
|
"epoch": 0.12698514755995627,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.1792762890458107,
|
|
"num_tokens": 4738989.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.8345729351043705,
|
|
"epoch": 0.12730662894618403,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 5.4978,
|
|
"mean_token_accuracy": 0.18163758963346482,
|
|
"num_tokens": 4750493.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.797660684585571,
|
|
"epoch": 0.12762811033241175,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.5225,
|
|
"mean_token_accuracy": 0.17844658195972443,
|
|
"num_tokens": 4763240.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.834128379821777,
|
|
"epoch": 0.12794959171863948,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 5.5719,
|
|
"mean_token_accuracy": 0.18070205599069594,
|
|
"num_tokens": 4774804.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.8326366424560545,
|
|
"epoch": 0.12827107310486724,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 5.6495,
|
|
"mean_token_accuracy": 0.17115117609500885,
|
|
"num_tokens": 4788684.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.905576086044311,
|
|
"epoch": 0.12859255449109497,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 5.6195,
|
|
"mean_token_accuracy": 0.17407953292131423,
|
|
"num_tokens": 4799819.0,
|
|
"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": 1078644754022400.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|