| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 371900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05377789728421619, | |
| "grad_norm": 0.7084974050521851, | |
| "learning_rate": 9.375e-06, | |
| "loss": 6.8536, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.10755579456843238, | |
| "grad_norm": 0.8629822731018066, | |
| "learning_rate": 1.875e-05, | |
| "loss": 5.3656, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16133369185264856, | |
| "grad_norm": 0.9347710013389587, | |
| "learning_rate": 2.8125e-05, | |
| "loss": 5.0405, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.21511158913686476, | |
| "grad_norm": 0.9131508469581604, | |
| "learning_rate": 3.75e-05, | |
| "loss": 4.8207, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.26888948642108096, | |
| "grad_norm": 1.0589386224746704, | |
| "learning_rate": 4.6874999999999994e-05, | |
| "loss": 4.6468, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32266738370529713, | |
| "grad_norm": 0.9457437992095947, | |
| "learning_rate": 5.625e-05, | |
| "loss": 4.4985, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.3764452809895133, | |
| "grad_norm": 0.9490616917610168, | |
| "learning_rate": 6.5625e-05, | |
| "loss": 4.3901, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.4302231782737295, | |
| "grad_norm": 0.9046244621276855, | |
| "learning_rate": 7.5e-05, | |
| "loss": 4.2829, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.4840010755579457, | |
| "grad_norm": 0.9089182615280151, | |
| "learning_rate": 8.437499999999999e-05, | |
| "loss": 4.2055, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.5377789728421619, | |
| "grad_norm": 0.8736773729324341, | |
| "learning_rate": 9.374999999999999e-05, | |
| "loss": 4.1359, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.591556870126378, | |
| "grad_norm": 0.9014385342597961, | |
| "learning_rate": 0.000103115625, | |
| "loss": 4.07, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.6453347674105943, | |
| "grad_norm": 0.8683918118476868, | |
| "learning_rate": 0.000112490625, | |
| "loss": 4.0086, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.6991126646948105, | |
| "grad_norm": 0.818735659122467, | |
| "learning_rate": 0.00012185624999999998, | |
| "loss": 3.946, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.7528905619790266, | |
| "grad_norm": 0.8135210275650024, | |
| "learning_rate": 0.00013123125, | |
| "loss": 3.8931, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.8066684592632428, | |
| "grad_norm": 0.8030542731285095, | |
| "learning_rate": 0.000140596875, | |
| "loss": 3.8372, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.860446356547459, | |
| "grad_norm": 0.7653594613075256, | |
| "learning_rate": 0.0001499625, | |
| "loss": 3.8021, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.9142242538316752, | |
| "grad_norm": 0.749381422996521, | |
| "learning_rate": 0.00015933749999999996, | |
| "loss": 3.762, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.9680021511158914, | |
| "grad_norm": 0.7983600497245789, | |
| "learning_rate": 0.00016871249999999996, | |
| "loss": 3.7356, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.34747039389556295, | |
| "eval_loss": 3.884793758392334, | |
| "eval_runtime": 152.1058, | |
| "eval_samples_per_second": 380.794, | |
| "eval_steps_per_second": 5.956, | |
| "step": 18595 | |
| }, | |
| { | |
| "epoch": 1.0217800484001076, | |
| "grad_norm": 0.7763124704360962, | |
| "learning_rate": 0.000178078125, | |
| "loss": 3.6984, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.0755579456843238, | |
| "grad_norm": 0.7071164846420288, | |
| "learning_rate": 0.00018745312499999998, | |
| "loss": 3.6652, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.1293358429685398, | |
| "grad_norm": 0.6947665214538574, | |
| "learning_rate": 0.00019681874999999998, | |
| "loss": 3.6519, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.183113740252756, | |
| "grad_norm": 0.6985417008399963, | |
| "learning_rate": 0.00020619374999999998, | |
| "loss": 3.6337, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.2368916375369723, | |
| "grad_norm": 0.6635871529579163, | |
| "learning_rate": 0.00021556874999999998, | |
| "loss": 3.6167, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.2906695348211885, | |
| "grad_norm": 0.6359060406684875, | |
| "learning_rate": 0.00022493437499999998, | |
| "loss": 3.5919, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.3444474321054047, | |
| "grad_norm": 0.6042994260787964, | |
| "learning_rate": 0.00023430937499999997, | |
| "loss": 3.5719, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.398225329389621, | |
| "grad_norm": 0.6072087287902832, | |
| "learning_rate": 0.00024368437499999997, | |
| "loss": 3.563, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.452003226673837, | |
| "grad_norm": 0.6790162920951843, | |
| "learning_rate": 0.00025305, | |
| "loss": 3.5498, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.5057811239580532, | |
| "grad_norm": 0.5274273157119751, | |
| "learning_rate": 0.000262425, | |
| "loss": 3.5337, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.5595590212422694, | |
| "grad_norm": 0.5580631494522095, | |
| "learning_rate": 0.000271790625, | |
| "loss": 3.5206, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.6133369185264856, | |
| "grad_norm": 0.5512628555297852, | |
| "learning_rate": 0.000281165625, | |
| "loss": 3.5171, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.6671148158107019, | |
| "grad_norm": 0.5888254642486572, | |
| "learning_rate": 0.000290540625, | |
| "loss": 3.5055, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.7208927130949179, | |
| "grad_norm": 0.5367724895477295, | |
| "learning_rate": 0.000299915625, | |
| "loss": 3.4934, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.7746706103791343, | |
| "grad_norm": 0.512359619140625, | |
| "learning_rate": 0.00029912621359223297, | |
| "loss": 3.4814, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.8284485076633503, | |
| "grad_norm": 0.5001698732376099, | |
| "learning_rate": 0.00029824360105913504, | |
| "loss": 3.4694, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.8822264049475665, | |
| "grad_norm": 0.48298901319503784, | |
| "learning_rate": 0.00029736187113857014, | |
| "loss": 3.4576, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.9360043022317828, | |
| "grad_norm": 0.4670732021331787, | |
| "learning_rate": 0.00029647925860547215, | |
| "loss": 3.4434, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.9897821995159988, | |
| "grad_norm": 0.4532361924648285, | |
| "learning_rate": 0.0002955966460723742, | |
| "loss": 3.4297, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3762663582363661, | |
| "eval_loss": 3.621605634689331, | |
| "eval_runtime": 154.0921, | |
| "eval_samples_per_second": 375.886, | |
| "eval_steps_per_second": 5.88, | |
| "step": 37190 | |
| }, | |
| { | |
| "epoch": 2.043560096800215, | |
| "grad_norm": 0.45829683542251587, | |
| "learning_rate": 0.0002947149161518093, | |
| "loss": 3.3852, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.0973379940844312, | |
| "grad_norm": 0.4475788474082947, | |
| "learning_rate": 0.0002938323036187114, | |
| "loss": 3.3735, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.1511158913686477, | |
| "grad_norm": 0.45039886236190796, | |
| "learning_rate": 0.00029294969108561335, | |
| "loss": 3.3662, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.2048937886528637, | |
| "grad_norm": 0.43961483240127563, | |
| "learning_rate": 0.0002920679611650485, | |
| "loss": 3.3651, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.2586716859370797, | |
| "grad_norm": 0.4328303337097168, | |
| "learning_rate": 0.0002911853486319506, | |
| "loss": 3.3594, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.312449583221296, | |
| "grad_norm": 0.42095446586608887, | |
| "learning_rate": 0.00029030273609885254, | |
| "loss": 3.3508, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.366227480505512, | |
| "grad_norm": 0.41147682070732117, | |
| "learning_rate": 0.0002894201235657546, | |
| "loss": 3.3519, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.4200053777897286, | |
| "grad_norm": 0.506058394908905, | |
| "learning_rate": 0.00028853839364518976, | |
| "loss": 3.3417, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.4737832750739446, | |
| "grad_norm": 0.4321398138999939, | |
| "learning_rate": 0.0002876557811120918, | |
| "loss": 3.3402, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.527561172358161, | |
| "grad_norm": 0.3769163191318512, | |
| "learning_rate": 0.0002867731685789938, | |
| "loss": 3.3323, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.581339069642377, | |
| "grad_norm": 0.3871973156929016, | |
| "learning_rate": 0.0002858905560458958, | |
| "loss": 3.3219, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.635116966926593, | |
| "grad_norm": 0.3995376527309418, | |
| "learning_rate": 0.00028500970873786405, | |
| "loss": 3.3231, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.6888948642108095, | |
| "grad_norm": 0.4042655825614929, | |
| "learning_rate": 0.0002841270962047661, | |
| "loss": 3.3141, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.7426727614950255, | |
| "grad_norm": 0.38159438967704773, | |
| "learning_rate": 0.0002832453662842012, | |
| "loss": 3.3062, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.796450658779242, | |
| "grad_norm": 0.3883545994758606, | |
| "learning_rate": 0.00028236275375110323, | |
| "loss": 3.3035, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.850228556063458, | |
| "grad_norm": 0.382907509803772, | |
| "learning_rate": 0.00028148014121800525, | |
| "loss": 3.297, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.904006453347674, | |
| "grad_norm": 0.3734199106693268, | |
| "learning_rate": 0.0002805984112974404, | |
| "loss": 3.2939, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.9577843506318904, | |
| "grad_norm": 0.3749130964279175, | |
| "learning_rate": 0.0002797157987643424, | |
| "loss": 3.2928, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.3889436282358831, | |
| "eval_loss": 3.500473737716675, | |
| "eval_runtime": 153.9288, | |
| "eval_samples_per_second": 376.284, | |
| "eval_steps_per_second": 5.886, | |
| "step": 55785 | |
| }, | |
| { | |
| "epoch": 3.0115622479161064, | |
| "grad_norm": 0.3949376344680786, | |
| "learning_rate": 0.00027883318623124443, | |
| "loss": 3.2765, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 3.065340145200323, | |
| "grad_norm": 0.3761361241340637, | |
| "learning_rate": 0.0002779505736981465, | |
| "loss": 3.2289, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.119118042484539, | |
| "grad_norm": 0.3742624521255493, | |
| "learning_rate": 0.0002770688437775816, | |
| "loss": 3.2292, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.172895939768755, | |
| "grad_norm": 0.3935544490814209, | |
| "learning_rate": 0.0002761862312444836, | |
| "loss": 3.2269, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 3.2266738370529713, | |
| "grad_norm": 0.3780879080295563, | |
| "learning_rate": 0.00027530450132391877, | |
| "loss": 3.2311, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.2804517343371873, | |
| "grad_norm": 0.38651397824287415, | |
| "learning_rate": 0.0002744218887908208, | |
| "loss": 3.2284, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.3342296316214037, | |
| "grad_norm": 0.36211293935775757, | |
| "learning_rate": 0.000273541041482789, | |
| "loss": 3.2297, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.3880075289056197, | |
| "grad_norm": 0.35539260506629944, | |
| "learning_rate": 0.0002726584289496911, | |
| "loss": 3.2274, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.4417854261898357, | |
| "grad_norm": 0.3886893391609192, | |
| "learning_rate": 0.0002717758164165931, | |
| "loss": 3.2248, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.495563323474052, | |
| "grad_norm": 0.35611864924430847, | |
| "learning_rate": 0.0002708940864960282, | |
| "loss": 3.2225, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.549341220758268, | |
| "grad_norm": 0.36730748414993286, | |
| "learning_rate": 0.0002700114739629303, | |
| "loss": 3.2154, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.6031191180424846, | |
| "grad_norm": 0.3518161475658417, | |
| "learning_rate": 0.0002691297440423654, | |
| "loss": 3.2157, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.6568970153267006, | |
| "grad_norm": 0.35682082176208496, | |
| "learning_rate": 0.0002682471315092674, | |
| "loss": 3.2173, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.7106749126109166, | |
| "grad_norm": 0.35613182187080383, | |
| "learning_rate": 0.00026736451897616946, | |
| "loss": 3.2145, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.764452809895133, | |
| "grad_norm": 0.36303049325942993, | |
| "learning_rate": 0.0002664819064430715, | |
| "loss": 3.217, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.8182307071793495, | |
| "grad_norm": 0.3425980508327484, | |
| "learning_rate": 0.0002656010591350397, | |
| "loss": 3.2102, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.8720086044635655, | |
| "grad_norm": 0.3505886197090149, | |
| "learning_rate": 0.00026471844660194173, | |
| "loss": 3.2077, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.9257865017477815, | |
| "grad_norm": 0.3440117835998535, | |
| "learning_rate": 0.00026383583406884375, | |
| "loss": 3.2041, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.979564399031998, | |
| "grad_norm": 0.353041410446167, | |
| "learning_rate": 0.0002629532215357458, | |
| "loss": 3.2003, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3951829061145071, | |
| "eval_loss": 3.465585231781006, | |
| "eval_runtime": 154.6505, | |
| "eval_samples_per_second": 374.528, | |
| "eval_steps_per_second": 5.858, | |
| "step": 74380 | |
| }, | |
| { | |
| "epoch": 4.033342296316214, | |
| "grad_norm": 0.3720568120479584, | |
| "learning_rate": 0.0002620706090026478, | |
| "loss": 3.165, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 4.08712019360043, | |
| "grad_norm": 0.36939120292663574, | |
| "learning_rate": 0.00026118887908208293, | |
| "loss": 3.1394, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 4.140898090884646, | |
| "grad_norm": 0.37178149819374084, | |
| "learning_rate": 0.0002603071491615181, | |
| "loss": 3.1458, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 4.1946759881688624, | |
| "grad_norm": 0.35939717292785645, | |
| "learning_rate": 0.0002594245366284201, | |
| "loss": 3.1472, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 4.2484538854530784, | |
| "grad_norm": 0.39039725065231323, | |
| "learning_rate": 0.0002585419240953221, | |
| "loss": 3.1441, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 4.302231782737295, | |
| "grad_norm": 0.35796868801116943, | |
| "learning_rate": 0.00025765931156222413, | |
| "loss": 3.1484, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 4.356009680021511, | |
| "grad_norm": 0.3571794033050537, | |
| "learning_rate": 0.0002567775816416593, | |
| "loss": 3.1469, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 4.409787577305727, | |
| "grad_norm": 0.34421437978744507, | |
| "learning_rate": 0.00025589496910856136, | |
| "loss": 3.1543, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 4.463565474589943, | |
| "grad_norm": 0.3431869447231293, | |
| "learning_rate": 0.0002550123565754633, | |
| "loss": 3.151, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 4.517343371874159, | |
| "grad_norm": 0.3607708215713501, | |
| "learning_rate": 0.00025413062665489847, | |
| "loss": 3.1488, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.571121269158376, | |
| "grad_norm": 0.33604896068573, | |
| "learning_rate": 0.0002532480141218005, | |
| "loss": 3.1481, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 4.624899166442592, | |
| "grad_norm": 0.3487938940525055, | |
| "learning_rate": 0.00025236540158870256, | |
| "loss": 3.1539, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 4.678677063726808, | |
| "grad_norm": 0.33633244037628174, | |
| "learning_rate": 0.00025148367166813766, | |
| "loss": 3.1482, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 4.732454961011024, | |
| "grad_norm": 0.3653378188610077, | |
| "learning_rate": 0.0002506010591350397, | |
| "loss": 3.1469, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 4.786232858295241, | |
| "grad_norm": 0.34846368432044983, | |
| "learning_rate": 0.00024971844660194174, | |
| "loss": 3.1499, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 4.840010755579457, | |
| "grad_norm": 0.34447020292282104, | |
| "learning_rate": 0.00024883671668137684, | |
| "loss": 3.1463, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 4.893788652863673, | |
| "grad_norm": 0.34981223940849304, | |
| "learning_rate": 0.00024795410414827886, | |
| "loss": 3.1476, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 4.947566550147889, | |
| "grad_norm": 0.33254608511924744, | |
| "learning_rate": 0.00024707149161518093, | |
| "loss": 3.1447, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.39942771910512487, | |
| "eval_loss": 3.406761407852173, | |
| "eval_runtime": 154.9021, | |
| "eval_samples_per_second": 373.92, | |
| "eval_steps_per_second": 5.849, | |
| "step": 92975 | |
| }, | |
| { | |
| "epoch": 5.001344447432105, | |
| "grad_norm": 0.35289084911346436, | |
| "learning_rate": 0.00024618976169461603, | |
| "loss": 3.1467, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 5.055122344716321, | |
| "grad_norm": 0.3626822531223297, | |
| "learning_rate": 0.00024530714916151804, | |
| "loss": 3.0786, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 5.108900242000538, | |
| "grad_norm": 0.3887752592563629, | |
| "learning_rate": 0.0002444245366284201, | |
| "loss": 3.0804, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 5.162678139284754, | |
| "grad_norm": 0.3546692430973053, | |
| "learning_rate": 0.00024354280670785524, | |
| "loss": 3.0885, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 5.21645603656897, | |
| "grad_norm": 0.3464365601539612, | |
| "learning_rate": 0.00024266019417475728, | |
| "loss": 3.092, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 5.270233933853186, | |
| "grad_norm": 0.3478812873363495, | |
| "learning_rate": 0.00024177758164165927, | |
| "loss": 3.0923, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 5.324011831137403, | |
| "grad_norm": 0.331265389919281, | |
| "learning_rate": 0.00024089496910856132, | |
| "loss": 3.0949, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 5.377789728421619, | |
| "grad_norm": 0.343042254447937, | |
| "learning_rate": 0.00024001412180052955, | |
| "loss": 3.0939, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 5.431567625705835, | |
| "grad_norm": 0.34835925698280334, | |
| "learning_rate": 0.0002391315092674316, | |
| "loss": 3.0965, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 5.485345522990051, | |
| "grad_norm": 0.3566593825817108, | |
| "learning_rate": 0.00023824889673433358, | |
| "loss": 3.0956, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 5.539123420274267, | |
| "grad_norm": 0.37776514887809753, | |
| "learning_rate": 0.00023736716681376874, | |
| "loss": 3.0957, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 5.592901317558484, | |
| "grad_norm": 0.3815798759460449, | |
| "learning_rate": 0.00023648455428067078, | |
| "loss": 3.0974, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 5.6466792148427, | |
| "grad_norm": 0.344208687543869, | |
| "learning_rate": 0.0002356019417475728, | |
| "loss": 3.1012, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 5.700457112126916, | |
| "grad_norm": 0.3420184850692749, | |
| "learning_rate": 0.00023471932921447481, | |
| "loss": 3.1009, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 5.754235009411132, | |
| "grad_norm": 0.3633731007575989, | |
| "learning_rate": 0.00023383848190644305, | |
| "loss": 3.0986, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 5.808012906695348, | |
| "grad_norm": 0.3575678765773773, | |
| "learning_rate": 0.0002329558693733451, | |
| "loss": 3.1004, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 5.861790803979565, | |
| "grad_norm": 0.35180947184562683, | |
| "learning_rate": 0.00023207325684024714, | |
| "loss": 3.1004, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 5.915568701263781, | |
| "grad_norm": 0.33941227197647095, | |
| "learning_rate": 0.00023119064430714912, | |
| "loss": 3.0987, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 5.969346598547997, | |
| "grad_norm": 0.33375173807144165, | |
| "learning_rate": 0.00023030891438658428, | |
| "loss": 3.0991, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.4022032021745438, | |
| "eval_loss": 3.4298133850097656, | |
| "eval_runtime": 153.945, | |
| "eval_samples_per_second": 376.245, | |
| "eval_steps_per_second": 5.885, | |
| "step": 111570 | |
| }, | |
| { | |
| "epoch": 6.023124495832213, | |
| "grad_norm": 0.3521207869052887, | |
| "learning_rate": 0.0002294263018534863, | |
| "loss": 3.0669, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 6.076902393116429, | |
| "grad_norm": 0.3600497245788574, | |
| "learning_rate": 0.00022854457193292145, | |
| "loss": 3.0351, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 6.130680290400646, | |
| "grad_norm": 0.373585045337677, | |
| "learning_rate": 0.00022766195939982344, | |
| "loss": 3.034, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 6.184458187684862, | |
| "grad_norm": 0.37141576409339905, | |
| "learning_rate": 0.00022677934686672548, | |
| "loss": 3.0454, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 6.238236084969078, | |
| "grad_norm": 0.34734484553337097, | |
| "learning_rate": 0.0002258984995586937, | |
| "loss": 3.0489, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 6.292013982253294, | |
| "grad_norm": 0.34972816705703735, | |
| "learning_rate": 0.00022501588702559576, | |
| "loss": 3.0488, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 6.34579187953751, | |
| "grad_norm": 0.36381569504737854, | |
| "learning_rate": 0.00022413327449249777, | |
| "loss": 3.0513, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 6.399569776821727, | |
| "grad_norm": 0.3417690396308899, | |
| "learning_rate": 0.0002232515445719329, | |
| "loss": 3.0494, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 6.453347674105943, | |
| "grad_norm": 0.35012194514274597, | |
| "learning_rate": 0.00022236893203883494, | |
| "loss": 3.0557, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 6.507125571390159, | |
| "grad_norm": 0.35446369647979736, | |
| "learning_rate": 0.00022148631950573699, | |
| "loss": 3.0584, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 6.560903468674375, | |
| "grad_norm": 0.3539404273033142, | |
| "learning_rate": 0.00022060370697263897, | |
| "loss": 3.0593, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 6.614681365958591, | |
| "grad_norm": 0.3495212197303772, | |
| "learning_rate": 0.00021972197705207413, | |
| "loss": 3.0562, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 6.6684592632428075, | |
| "grad_norm": 0.34305086731910706, | |
| "learning_rate": 0.00021883936451897614, | |
| "loss": 3.0582, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 6.7222371605270235, | |
| "grad_norm": 0.34190261363983154, | |
| "learning_rate": 0.00021795675198587816, | |
| "loss": 3.0585, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 6.7760150578112395, | |
| "grad_norm": 0.3615049719810486, | |
| "learning_rate": 0.0002170741394527802, | |
| "loss": 3.06, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 6.8297929550954555, | |
| "grad_norm": 0.3419729769229889, | |
| "learning_rate": 0.00021619152691968225, | |
| "loss": 3.0571, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 6.8835708523796715, | |
| "grad_norm": 0.3504064381122589, | |
| "learning_rate": 0.00021531067961165048, | |
| "loss": 3.0587, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 6.937348749663888, | |
| "grad_norm": 0.3564949333667755, | |
| "learning_rate": 0.00021442806707855253, | |
| "loss": 3.0596, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 6.991126646948104, | |
| "grad_norm": 0.358786404132843, | |
| "learning_rate": 0.00021354633715798762, | |
| "loss": 3.0592, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.4040891064758338, | |
| "eval_loss": 3.398965358734131, | |
| "eval_runtime": 154.0201, | |
| "eval_samples_per_second": 376.061, | |
| "eval_steps_per_second": 5.882, | |
| "step": 130165 | |
| }, | |
| { | |
| "epoch": 7.04490454423232, | |
| "grad_norm": 0.36317363381385803, | |
| "learning_rate": 0.00021266372462488964, | |
| "loss": 3.0022, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 7.098682441516536, | |
| "grad_norm": 0.36218351125717163, | |
| "learning_rate": 0.0002117819947043248, | |
| "loss": 2.995, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 7.152460338800753, | |
| "grad_norm": 0.36327067017555237, | |
| "learning_rate": 0.00021089938217122684, | |
| "loss": 3.0054, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 7.206238236084969, | |
| "grad_norm": 0.33495405316352844, | |
| "learning_rate": 0.00021001676963812883, | |
| "loss": 3.0095, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 7.260016133369185, | |
| "grad_norm": 0.38024088740348816, | |
| "learning_rate": 0.00020913415710503087, | |
| "loss": 3.0094, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 7.313794030653401, | |
| "grad_norm": 0.3585461676120758, | |
| "learning_rate": 0.00020825242718446602, | |
| "loss": 3.0138, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 7.367571927937617, | |
| "grad_norm": 0.3688068687915802, | |
| "learning_rate": 0.000207369814651368, | |
| "loss": 3.0179, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 7.421349825221834, | |
| "grad_norm": 0.35052135586738586, | |
| "learning_rate": 0.00020648720211827005, | |
| "loss": 3.0208, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 7.47512772250605, | |
| "grad_norm": 0.35148343443870544, | |
| "learning_rate": 0.00020560547219770518, | |
| "loss": 3.0192, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 7.528905619790266, | |
| "grad_norm": 0.3611489236354828, | |
| "learning_rate": 0.00020472285966460722, | |
| "loss": 3.0199, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 7.582683517074482, | |
| "grad_norm": 0.3296034634113312, | |
| "learning_rate": 0.00020384024713150924, | |
| "loss": 3.0237, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 7.636461414358699, | |
| "grad_norm": 0.35607197880744934, | |
| "learning_rate": 0.00020295851721094437, | |
| "loss": 3.0238, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 7.690239311642915, | |
| "grad_norm": 0.35551297664642334, | |
| "learning_rate": 0.0002020759046778464, | |
| "loss": 3.0217, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 7.744017208927131, | |
| "grad_norm": 0.3628598153591156, | |
| "learning_rate": 0.00020119329214474845, | |
| "loss": 3.0262, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 7.797795106211347, | |
| "grad_norm": 0.350730836391449, | |
| "learning_rate": 0.00020031156222418355, | |
| "loss": 3.0256, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 7.851573003495563, | |
| "grad_norm": 0.3436521291732788, | |
| "learning_rate": 0.0001994289496910856, | |
| "loss": 3.0269, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 7.90535090077978, | |
| "grad_norm": 0.3377685248851776, | |
| "learning_rate": 0.00019854633715798764, | |
| "loss": 3.0278, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 7.959128798063996, | |
| "grad_norm": 0.3418731689453125, | |
| "learning_rate": 0.00019766372462488965, | |
| "loss": 3.0279, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.40574452106609454, | |
| "eval_loss": 3.3796093463897705, | |
| "eval_runtime": 154.5289, | |
| "eval_samples_per_second": 374.823, | |
| "eval_steps_per_second": 5.863, | |
| "step": 148760 | |
| }, | |
| { | |
| "epoch": 8.012906695348212, | |
| "grad_norm": 0.34725406765937805, | |
| "learning_rate": 0.00019678111209179167, | |
| "loss": 3.0056, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 8.066684592632429, | |
| "grad_norm": 0.34432175755500793, | |
| "learning_rate": 0.0001959002647837599, | |
| "loss": 2.962, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 8.120462489916644, | |
| "grad_norm": 0.3772197961807251, | |
| "learning_rate": 0.00019501765225066195, | |
| "loss": 2.9645, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 8.17424038720086, | |
| "grad_norm": 0.3569514751434326, | |
| "learning_rate": 0.00019413503971756397, | |
| "loss": 2.9748, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 8.228018284485076, | |
| "grad_norm": 0.3487160801887512, | |
| "learning_rate": 0.000193252427184466, | |
| "loss": 2.9757, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 8.281796181769293, | |
| "grad_norm": 0.35110118985176086, | |
| "learning_rate": 0.00019237069726390113, | |
| "loss": 2.9814, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 8.33557407905351, | |
| "grad_norm": 0.36479270458221436, | |
| "learning_rate": 0.00019148808473080318, | |
| "loss": 2.9834, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 8.389351976337725, | |
| "grad_norm": 0.36717069149017334, | |
| "learning_rate": 0.00019060547219770517, | |
| "loss": 2.9883, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 8.443129873621942, | |
| "grad_norm": 0.35104259848594666, | |
| "learning_rate": 0.00018972374227714032, | |
| "loss": 2.9878, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 8.496907770906157, | |
| "grad_norm": 0.3428269922733307, | |
| "learning_rate": 0.00018884112974404236, | |
| "loss": 2.9893, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 8.550685668190374, | |
| "grad_norm": 0.3437536358833313, | |
| "learning_rate": 0.0001879593998234775, | |
| "loss": 2.9924, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 8.60446356547459, | |
| "grad_norm": 0.34665447473526, | |
| "learning_rate": 0.0001870767872903795, | |
| "loss": 2.9911, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 8.658241462758806, | |
| "grad_norm": 0.36713042855262756, | |
| "learning_rate": 0.00018619417475728152, | |
| "loss": 2.9901, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 8.712019360043023, | |
| "grad_norm": 0.38735637068748474, | |
| "learning_rate": 0.00018531244483671667, | |
| "loss": 2.9959, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 8.765797257327238, | |
| "grad_norm": 0.3698832094669342, | |
| "learning_rate": 0.00018442983230361872, | |
| "loss": 2.9985, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 8.819575154611455, | |
| "grad_norm": 0.37609660625457764, | |
| "learning_rate": 0.0001835472197705207, | |
| "loss": 2.9956, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 8.873353051895672, | |
| "grad_norm": 0.3600751757621765, | |
| "learning_rate": 0.00018266460723742275, | |
| "loss": 2.9981, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 8.927130949179887, | |
| "grad_norm": 0.37786364555358887, | |
| "learning_rate": 0.00018178287731685788, | |
| "loss": 2.9975, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 8.980908846464104, | |
| "grad_norm": 0.3594890832901001, | |
| "learning_rate": 0.00018090114739629303, | |
| "loss": 2.9978, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.40631117239589115, | |
| "eval_loss": 3.3757009506225586, | |
| "eval_runtime": 153.8626, | |
| "eval_samples_per_second": 376.446, | |
| "eval_steps_per_second": 5.888, | |
| "step": 167355 | |
| }, | |
| { | |
| "epoch": 9.034686743748319, | |
| "grad_norm": 0.36975300312042236, | |
| "learning_rate": 0.00018001853486319502, | |
| "loss": 2.9532, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 9.088464641032536, | |
| "grad_norm": 0.4000973403453827, | |
| "learning_rate": 0.00017913592233009706, | |
| "loss": 2.9361, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 9.142242538316752, | |
| "grad_norm": 0.3800918161869049, | |
| "learning_rate": 0.0001782533097969991, | |
| "loss": 2.942, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 9.196020435600968, | |
| "grad_norm": 0.3795217275619507, | |
| "learning_rate": 0.00017737157987643426, | |
| "loss": 2.9452, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 9.249798332885184, | |
| "grad_norm": 0.37562939524650574, | |
| "learning_rate": 0.00017648984995586936, | |
| "loss": 2.9491, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 9.3035762301694, | |
| "grad_norm": 0.3602105379104614, | |
| "learning_rate": 0.00017560723742277137, | |
| "loss": 2.9543, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 9.357354127453616, | |
| "grad_norm": 0.3644072115421295, | |
| "learning_rate": 0.00017472550750220652, | |
| "loss": 2.9583, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 9.411132024737833, | |
| "grad_norm": 0.3795081675052643, | |
| "learning_rate": 0.00017384289496910857, | |
| "loss": 2.963, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 9.464909922022049, | |
| "grad_norm": 0.37337377667427063, | |
| "learning_rate": 0.00017296028243601056, | |
| "loss": 2.9585, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 9.518687819306265, | |
| "grad_norm": 0.35948848724365234, | |
| "learning_rate": 0.0001720785525154457, | |
| "loss": 2.9648, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 9.57246571659048, | |
| "grad_norm": 0.3625440001487732, | |
| "learning_rate": 0.00017119593998234775, | |
| "loss": 2.9573, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 9.626243613874697, | |
| "grad_norm": 0.35655486583709717, | |
| "learning_rate": 0.00017031332744924974, | |
| "loss": 2.9647, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 9.680021511158914, | |
| "grad_norm": 0.3656957745552063, | |
| "learning_rate": 0.0001694307149161518, | |
| "loss": 2.9667, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 9.73379940844313, | |
| "grad_norm": 0.3741768002510071, | |
| "learning_rate": 0.0001685489849955869, | |
| "loss": 2.9664, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 9.787577305727346, | |
| "grad_norm": 0.37431761622428894, | |
| "learning_rate": 0.00016766725507502206, | |
| "loss": 2.966, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 9.841355203011563, | |
| "grad_norm": 0.3461393117904663, | |
| "learning_rate": 0.0001667846425419241, | |
| "loss": 2.9733, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 9.895133100295778, | |
| "grad_norm": 0.3577495217323303, | |
| "learning_rate": 0.0001659029126213592, | |
| "loss": 2.9728, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 9.948910997579995, | |
| "grad_norm": 0.35284364223480225, | |
| "learning_rate": 0.00016502030008826125, | |
| "loss": 2.9761, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4067828332123111, | |
| "eval_loss": 3.390697717666626, | |
| "eval_runtime": 154.1665, | |
| "eval_samples_per_second": 375.704, | |
| "eval_steps_per_second": 5.877, | |
| "step": 185950 | |
| }, | |
| { | |
| "epoch": 10.00268889486421, | |
| "grad_norm": 0.3526718020439148, | |
| "learning_rate": 0.00016413768755516327, | |
| "loss": 2.9713, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 10.056466792148427, | |
| "grad_norm": 0.3849118649959564, | |
| "learning_rate": 0.00016325507502206528, | |
| "loss": 2.9052, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 10.110244689432642, | |
| "grad_norm": 0.3799303472042084, | |
| "learning_rate": 0.00016237246248896733, | |
| "loss": 2.9176, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 10.16402258671686, | |
| "grad_norm": 0.3599086105823517, | |
| "learning_rate": 0.00016148984995586937, | |
| "loss": 2.9221, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 10.217800484001076, | |
| "grad_norm": 0.3713199198246002, | |
| "learning_rate": 0.0001606081200353045, | |
| "loss": 2.9216, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 10.271578381285291, | |
| "grad_norm": 0.4002034664154053, | |
| "learning_rate": 0.0001597255075022065, | |
| "loss": 2.9235, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 10.325356278569508, | |
| "grad_norm": 0.39608556032180786, | |
| "learning_rate": 0.00015884466019417475, | |
| "loss": 2.9293, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 10.379134175853725, | |
| "grad_norm": 0.3750903904438019, | |
| "learning_rate": 0.00015796204766107676, | |
| "loss": 2.9305, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 10.43291207313794, | |
| "grad_norm": 0.36680716276168823, | |
| "learning_rate": 0.0001570794351279788, | |
| "loss": 2.9354, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 10.486689970422157, | |
| "grad_norm": 0.36688679456710815, | |
| "learning_rate": 0.00015619682259488082, | |
| "loss": 2.9362, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 10.540467867706372, | |
| "grad_norm": 0.3786678910255432, | |
| "learning_rate": 0.00015531509267431595, | |
| "loss": 2.9394, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 10.594245764990589, | |
| "grad_norm": 0.39312633872032166, | |
| "learning_rate": 0.0001544333627537511, | |
| "loss": 2.938, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 10.648023662274806, | |
| "grad_norm": 0.3482972979545593, | |
| "learning_rate": 0.00015355075022065312, | |
| "loss": 2.9451, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 10.701801559559021, | |
| "grad_norm": 0.3624250590801239, | |
| "learning_rate": 0.00015266813768755513, | |
| "loss": 2.9398, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 10.755579456843238, | |
| "grad_norm": 0.3837498128414154, | |
| "learning_rate": 0.00015178552515445718, | |
| "loss": 2.9443, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 10.809357354127453, | |
| "grad_norm": 0.37360501289367676, | |
| "learning_rate": 0.0001509037952338923, | |
| "loss": 2.949, | |
| "step": 201000 | |
| }, | |
| { | |
| "epoch": 10.86313525141167, | |
| "grad_norm": 0.37839803099632263, | |
| "learning_rate": 0.00015002206531332745, | |
| "loss": 2.9452, | |
| "step": 202000 | |
| }, | |
| { | |
| "epoch": 10.916913148695887, | |
| "grad_norm": 0.39486682415008545, | |
| "learning_rate": 0.00014913945278022947, | |
| "loss": 2.9498, | |
| "step": 203000 | |
| }, | |
| { | |
| "epoch": 10.970691045980102, | |
| "grad_norm": 0.3875749409198761, | |
| "learning_rate": 0.0001482568402471315, | |
| "loss": 2.9464, | |
| "step": 204000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.4072896956051168, | |
| "eval_loss": 3.3881287574768066, | |
| "eval_runtime": 154.0418, | |
| "eval_samples_per_second": 376.008, | |
| "eval_steps_per_second": 5.882, | |
| "step": 204545 | |
| }, | |
| { | |
| "epoch": 11.024468943264319, | |
| "grad_norm": 0.40221425890922546, | |
| "learning_rate": 0.00014737422771403353, | |
| "loss": 2.9172, | |
| "step": 205000 | |
| }, | |
| { | |
| "epoch": 11.078246840548534, | |
| "grad_norm": 0.384132981300354, | |
| "learning_rate": 0.00014649249779346866, | |
| "loss": 2.8888, | |
| "step": 206000 | |
| }, | |
| { | |
| "epoch": 11.13202473783275, | |
| "grad_norm": 0.40102720260620117, | |
| "learning_rate": 0.0001456098852603707, | |
| "loss": 2.8918, | |
| "step": 207000 | |
| }, | |
| { | |
| "epoch": 11.185802635116968, | |
| "grad_norm": 0.3817023038864136, | |
| "learning_rate": 0.00014472727272727272, | |
| "loss": 2.8962, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 11.239580532401183, | |
| "grad_norm": 0.384730726480484, | |
| "learning_rate": 0.00014384466019417473, | |
| "loss": 2.9001, | |
| "step": 209000 | |
| }, | |
| { | |
| "epoch": 11.2933584296854, | |
| "grad_norm": 0.41385701298713684, | |
| "learning_rate": 0.00014296293027360986, | |
| "loss": 2.9037, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 11.347136326969615, | |
| "grad_norm": 0.4016377329826355, | |
| "learning_rate": 0.000142081200353045, | |
| "loss": 2.9088, | |
| "step": 211000 | |
| }, | |
| { | |
| "epoch": 11.400914224253832, | |
| "grad_norm": 0.39662501215934753, | |
| "learning_rate": 0.00014119858781994703, | |
| "loss": 2.9112, | |
| "step": 212000 | |
| }, | |
| { | |
| "epoch": 11.454692121538049, | |
| "grad_norm": 0.3822558522224426, | |
| "learning_rate": 0.00014031597528684904, | |
| "loss": 2.914, | |
| "step": 213000 | |
| }, | |
| { | |
| "epoch": 11.508470018822264, | |
| "grad_norm": 0.3875982463359833, | |
| "learning_rate": 0.0001394333627537511, | |
| "loss": 2.917, | |
| "step": 214000 | |
| }, | |
| { | |
| "epoch": 11.56224791610648, | |
| "grad_norm": 0.38756969571113586, | |
| "learning_rate": 0.0001385507502206531, | |
| "loss": 2.9182, | |
| "step": 215000 | |
| }, | |
| { | |
| "epoch": 11.616025813390696, | |
| "grad_norm": 0.3970773220062256, | |
| "learning_rate": 0.00013766902030008826, | |
| "loss": 2.9191, | |
| "step": 216000 | |
| }, | |
| { | |
| "epoch": 11.669803710674913, | |
| "grad_norm": 0.36975619196891785, | |
| "learning_rate": 0.00013678640776699027, | |
| "loss": 2.9223, | |
| "step": 217000 | |
| }, | |
| { | |
| "epoch": 11.72358160795913, | |
| "grad_norm": 0.3606310188770294, | |
| "learning_rate": 0.0001359046778464254, | |
| "loss": 2.9189, | |
| "step": 218000 | |
| }, | |
| { | |
| "epoch": 11.777359505243345, | |
| "grad_norm": 0.38579317927360535, | |
| "learning_rate": 0.00013502206531332744, | |
| "loss": 2.9166, | |
| "step": 219000 | |
| }, | |
| { | |
| "epoch": 11.831137402527562, | |
| "grad_norm": 0.38102084398269653, | |
| "learning_rate": 0.00013413945278022949, | |
| "loss": 2.923, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 11.884915299811777, | |
| "grad_norm": 0.39203277230262756, | |
| "learning_rate": 0.0001332568402471315, | |
| "loss": 2.9235, | |
| "step": 221000 | |
| }, | |
| { | |
| "epoch": 11.938693197095994, | |
| "grad_norm": 0.40768030285835266, | |
| "learning_rate": 0.00013237511032656663, | |
| "loss": 2.9268, | |
| "step": 222000 | |
| }, | |
| { | |
| "epoch": 11.99247109438021, | |
| "grad_norm": 0.3832620084285736, | |
| "learning_rate": 0.00013149249779346864, | |
| "loss": 2.9236, | |
| "step": 223000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4080087079563691, | |
| "eval_loss": 3.3905434608459473, | |
| "eval_runtime": 153.9882, | |
| "eval_samples_per_second": 376.139, | |
| "eval_steps_per_second": 5.884, | |
| "step": 223140 | |
| }, | |
| { | |
| "epoch": 12.046248991664426, | |
| "grad_norm": 0.39426013827323914, | |
| "learning_rate": 0.0001306098852603707, | |
| "loss": 2.867, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 12.100026888948642, | |
| "grad_norm": 0.3989400267601013, | |
| "learning_rate": 0.00012972727272727273, | |
| "loss": 2.8675, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 12.153804786232858, | |
| "grad_norm": 0.4114680290222168, | |
| "learning_rate": 0.00012884554280670786, | |
| "loss": 2.8738, | |
| "step": 226000 | |
| }, | |
| { | |
| "epoch": 12.207582683517074, | |
| "grad_norm": 0.3789440393447876, | |
| "learning_rate": 0.00012796293027360987, | |
| "loss": 2.8802, | |
| "step": 227000 | |
| }, | |
| { | |
| "epoch": 12.261360580801291, | |
| "grad_norm": 0.4167686700820923, | |
| "learning_rate": 0.0001270803177405119, | |
| "loss": 2.8806, | |
| "step": 228000 | |
| }, | |
| { | |
| "epoch": 12.315138478085506, | |
| "grad_norm": 0.3888895511627197, | |
| "learning_rate": 0.00012619770520741393, | |
| "loss": 2.8876, | |
| "step": 229000 | |
| }, | |
| { | |
| "epoch": 12.368916375369723, | |
| "grad_norm": 0.40582403540611267, | |
| "learning_rate": 0.00012531597528684906, | |
| "loss": 2.8857, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 12.422694272653938, | |
| "grad_norm": 0.3923300504684448, | |
| "learning_rate": 0.0001244333627537511, | |
| "loss": 2.8883, | |
| "step": 231000 | |
| }, | |
| { | |
| "epoch": 12.476472169938155, | |
| "grad_norm": 0.3777754604816437, | |
| "learning_rate": 0.00012355163283318623, | |
| "loss": 2.893, | |
| "step": 232000 | |
| }, | |
| { | |
| "epoch": 12.530250067222372, | |
| "grad_norm": 0.4126594066619873, | |
| "learning_rate": 0.00012266990291262135, | |
| "loss": 2.8934, | |
| "step": 233000 | |
| }, | |
| { | |
| "epoch": 12.584027964506587, | |
| "grad_norm": 0.3741176426410675, | |
| "learning_rate": 0.00012178729037952337, | |
| "loss": 2.8971, | |
| "step": 234000 | |
| }, | |
| { | |
| "epoch": 12.637805861790804, | |
| "grad_norm": 0.41800355911254883, | |
| "learning_rate": 0.00012090467784642541, | |
| "loss": 2.895, | |
| "step": 235000 | |
| }, | |
| { | |
| "epoch": 12.69158375907502, | |
| "grad_norm": 0.3904781937599182, | |
| "learning_rate": 0.00012002294792586052, | |
| "loss": 2.8958, | |
| "step": 236000 | |
| }, | |
| { | |
| "epoch": 12.745361656359236, | |
| "grad_norm": 0.3890039324760437, | |
| "learning_rate": 0.00011914033539276257, | |
| "loss": 2.8997, | |
| "step": 237000 | |
| }, | |
| { | |
| "epoch": 12.799139553643453, | |
| "grad_norm": 0.3933662474155426, | |
| "learning_rate": 0.0001182577228596646, | |
| "loss": 2.9018, | |
| "step": 238000 | |
| }, | |
| { | |
| "epoch": 12.852917450927668, | |
| "grad_norm": 0.3928120732307434, | |
| "learning_rate": 0.00011737511032656663, | |
| "loss": 2.902, | |
| "step": 239000 | |
| }, | |
| { | |
| "epoch": 12.906695348211885, | |
| "grad_norm": 0.37820199131965637, | |
| "learning_rate": 0.00011649338040600175, | |
| "loss": 2.905, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 12.9604732454961, | |
| "grad_norm": 0.3890197277069092, | |
| "learning_rate": 0.0001156107678729038, | |
| "loss": 2.907, | |
| "step": 241000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.4086836511585584, | |
| "eval_loss": 3.3879897594451904, | |
| "eval_runtime": 154.2919, | |
| "eval_samples_per_second": 375.399, | |
| "eval_steps_per_second": 5.872, | |
| "step": 241735 | |
| }, | |
| { | |
| "epoch": 13.014251142780317, | |
| "grad_norm": 0.3926098942756653, | |
| "learning_rate": 0.00011472903795233891, | |
| "loss": 2.8894, | |
| "step": 242000 | |
| }, | |
| { | |
| "epoch": 13.068029040064534, | |
| "grad_norm": 0.4247579872608185, | |
| "learning_rate": 0.00011384642541924095, | |
| "loss": 2.8512, | |
| "step": 243000 | |
| }, | |
| { | |
| "epoch": 13.12180693734875, | |
| "grad_norm": 0.4137355387210846, | |
| "learning_rate": 0.00011296469549867606, | |
| "loss": 2.8545, | |
| "step": 244000 | |
| }, | |
| { | |
| "epoch": 13.175584834632966, | |
| "grad_norm": 0.3952256739139557, | |
| "learning_rate": 0.0001120829655781112, | |
| "loss": 2.8564, | |
| "step": 245000 | |
| }, | |
| { | |
| "epoch": 13.229362731917181, | |
| "grad_norm": 0.3921244740486145, | |
| "learning_rate": 0.00011120035304501322, | |
| "loss": 2.8581, | |
| "step": 246000 | |
| }, | |
| { | |
| "epoch": 13.283140629201398, | |
| "grad_norm": 0.3818395137786865, | |
| "learning_rate": 0.00011031774051191526, | |
| "loss": 2.8626, | |
| "step": 247000 | |
| }, | |
| { | |
| "epoch": 13.336918526485615, | |
| "grad_norm": 0.40112724900245667, | |
| "learning_rate": 0.00010943601059135037, | |
| "loss": 2.8613, | |
| "step": 248000 | |
| }, | |
| { | |
| "epoch": 13.39069642376983, | |
| "grad_norm": 0.3878421187400818, | |
| "learning_rate": 0.00010855339805825242, | |
| "loss": 2.8637, | |
| "step": 249000 | |
| }, | |
| { | |
| "epoch": 13.444474321054047, | |
| "grad_norm": 0.4171620011329651, | |
| "learning_rate": 0.00010767078552515445, | |
| "loss": 2.8673, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 13.498252218338264, | |
| "grad_norm": 0.39591026306152344, | |
| "learning_rate": 0.00010678905560458957, | |
| "loss": 2.8703, | |
| "step": 251000 | |
| }, | |
| { | |
| "epoch": 13.552030115622479, | |
| "grad_norm": 0.39251312613487244, | |
| "learning_rate": 0.0001059064430714916, | |
| "loss": 2.878, | |
| "step": 252000 | |
| }, | |
| { | |
| "epoch": 13.605808012906696, | |
| "grad_norm": 0.3924385607242584, | |
| "learning_rate": 0.00010502471315092674, | |
| "loss": 2.8778, | |
| "step": 253000 | |
| }, | |
| { | |
| "epoch": 13.659585910190911, | |
| "grad_norm": 0.40747836232185364, | |
| "learning_rate": 0.00010414210061782876, | |
| "loss": 2.8754, | |
| "step": 254000 | |
| }, | |
| { | |
| "epoch": 13.713363807475128, | |
| "grad_norm": 0.3943057954311371, | |
| "learning_rate": 0.0001032594880847308, | |
| "loss": 2.8814, | |
| "step": 255000 | |
| }, | |
| { | |
| "epoch": 13.767141704759345, | |
| "grad_norm": 0.39504125714302063, | |
| "learning_rate": 0.00010237775816416591, | |
| "loss": 2.8816, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 13.82091960204356, | |
| "grad_norm": 0.3863300085067749, | |
| "learning_rate": 0.00010149514563106796, | |
| "loss": 2.885, | |
| "step": 257000 | |
| }, | |
| { | |
| "epoch": 13.874697499327777, | |
| "grad_norm": 0.406428724527359, | |
| "learning_rate": 0.00010061341571050307, | |
| "loss": 2.8806, | |
| "step": 258000 | |
| }, | |
| { | |
| "epoch": 13.928475396611992, | |
| "grad_norm": 0.4188634157180786, | |
| "learning_rate": 9.973080317740511e-05, | |
| "loss": 2.8808, | |
| "step": 259000 | |
| }, | |
| { | |
| "epoch": 13.982253293896209, | |
| "grad_norm": 0.4093535542488098, | |
| "learning_rate": 9.884819064430714e-05, | |
| "loss": 2.8803, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.40880511003280534, | |
| "eval_loss": 3.3926610946655273, | |
| "eval_runtime": 154.1123, | |
| "eval_samples_per_second": 375.836, | |
| "eval_steps_per_second": 5.879, | |
| "step": 260330 | |
| }, | |
| { | |
| "epoch": 14.036031191180426, | |
| "grad_norm": 0.41151463985443115, | |
| "learning_rate": 9.796557811120916e-05, | |
| "loss": 2.8484, | |
| "step": 261000 | |
| }, | |
| { | |
| "epoch": 14.08980908846464, | |
| "grad_norm": 0.4244749844074249, | |
| "learning_rate": 9.70838481906443e-05, | |
| "loss": 2.8302, | |
| "step": 262000 | |
| }, | |
| { | |
| "epoch": 14.143586985748858, | |
| "grad_norm": 0.43255192041397095, | |
| "learning_rate": 9.620123565754632e-05, | |
| "loss": 2.8321, | |
| "step": 263000 | |
| }, | |
| { | |
| "epoch": 14.197364883033073, | |
| "grad_norm": 0.40345045924186707, | |
| "learning_rate": 9.531862312444836e-05, | |
| "loss": 2.8367, | |
| "step": 264000 | |
| }, | |
| { | |
| "epoch": 14.25114278031729, | |
| "grad_norm": 0.4035230875015259, | |
| "learning_rate": 9.443601059135039e-05, | |
| "loss": 2.8442, | |
| "step": 265000 | |
| }, | |
| { | |
| "epoch": 14.304920677601507, | |
| "grad_norm": 0.40154018998146057, | |
| "learning_rate": 9.355516328331861e-05, | |
| "loss": 2.845, | |
| "step": 266000 | |
| }, | |
| { | |
| "epoch": 14.358698574885722, | |
| "grad_norm": 0.39079976081848145, | |
| "learning_rate": 9.267255075022065e-05, | |
| "loss": 2.8454, | |
| "step": 267000 | |
| }, | |
| { | |
| "epoch": 14.412476472169939, | |
| "grad_norm": 0.39859312772750854, | |
| "learning_rate": 9.178993821712268e-05, | |
| "loss": 2.8517, | |
| "step": 268000 | |
| }, | |
| { | |
| "epoch": 14.466254369454154, | |
| "grad_norm": 0.42548927664756775, | |
| "learning_rate": 9.09073256840247e-05, | |
| "loss": 2.8499, | |
| "step": 269000 | |
| }, | |
| { | |
| "epoch": 14.52003226673837, | |
| "grad_norm": 0.4101540148258209, | |
| "learning_rate": 9.002647837599293e-05, | |
| "loss": 2.8538, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 14.573810164022587, | |
| "grad_norm": 0.4425499737262726, | |
| "learning_rate": 8.914386584289496e-05, | |
| "loss": 2.8565, | |
| "step": 271000 | |
| }, | |
| { | |
| "epoch": 14.627588061306803, | |
| "grad_norm": 0.40012863278388977, | |
| "learning_rate": 8.8261253309797e-05, | |
| "loss": 2.8565, | |
| "step": 272000 | |
| }, | |
| { | |
| "epoch": 14.68136595859102, | |
| "grad_norm": 0.39511892199516296, | |
| "learning_rate": 8.737952338923212e-05, | |
| "loss": 2.8588, | |
| "step": 273000 | |
| }, | |
| { | |
| "epoch": 14.735143855875235, | |
| "grad_norm": 0.42598825693130493, | |
| "learning_rate": 8.649691085613415e-05, | |
| "loss": 2.8593, | |
| "step": 274000 | |
| }, | |
| { | |
| "epoch": 14.788921753159451, | |
| "grad_norm": 0.4081316292285919, | |
| "learning_rate": 8.561429832303618e-05, | |
| "loss": 2.861, | |
| "step": 275000 | |
| }, | |
| { | |
| "epoch": 14.842699650443668, | |
| "grad_norm": 0.4154646098613739, | |
| "learning_rate": 8.473168578993821e-05, | |
| "loss": 2.8624, | |
| "step": 276000 | |
| }, | |
| { | |
| "epoch": 14.896477547727883, | |
| "grad_norm": 0.4205508828163147, | |
| "learning_rate": 8.384995586937333e-05, | |
| "loss": 2.8649, | |
| "step": 277000 | |
| }, | |
| { | |
| "epoch": 14.9502554450121, | |
| "grad_norm": 0.4075837731361389, | |
| "learning_rate": 8.296734333627537e-05, | |
| "loss": 2.8614, | |
| "step": 278000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.4090819529646479, | |
| "eval_loss": 3.3923890590667725, | |
| "eval_runtime": 153.9312, | |
| "eval_samples_per_second": 376.279, | |
| "eval_steps_per_second": 5.886, | |
| "step": 278925 | |
| }, | |
| { | |
| "epoch": 15.004033342296315, | |
| "grad_norm": 0.42821255326271057, | |
| "learning_rate": 8.20847308031774e-05, | |
| "loss": 2.8598, | |
| "step": 279000 | |
| }, | |
| { | |
| "epoch": 15.057811239580532, | |
| "grad_norm": 0.41713452339172363, | |
| "learning_rate": 8.120300088261253e-05, | |
| "loss": 2.8112, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 15.11158913686475, | |
| "grad_norm": 0.4310871660709381, | |
| "learning_rate": 8.032127096204765e-05, | |
| "loss": 2.8187, | |
| "step": 281000 | |
| }, | |
| { | |
| "epoch": 15.165367034148964, | |
| "grad_norm": 0.42941373586654663, | |
| "learning_rate": 7.943865842894969e-05, | |
| "loss": 2.82, | |
| "step": 282000 | |
| }, | |
| { | |
| "epoch": 15.219144931433181, | |
| "grad_norm": 0.4279724657535553, | |
| "learning_rate": 7.85560458958517e-05, | |
| "loss": 2.824, | |
| "step": 283000 | |
| }, | |
| { | |
| "epoch": 15.272922828717396, | |
| "grad_norm": 0.4205762445926666, | |
| "learning_rate": 7.767343336275375e-05, | |
| "loss": 2.8251, | |
| "step": 284000 | |
| }, | |
| { | |
| "epoch": 15.326700726001613, | |
| "grad_norm": 0.40928196907043457, | |
| "learning_rate": 7.679082082965578e-05, | |
| "loss": 2.8295, | |
| "step": 285000 | |
| }, | |
| { | |
| "epoch": 15.38047862328583, | |
| "grad_norm": 0.4192693829536438, | |
| "learning_rate": 7.59090909090909e-05, | |
| "loss": 2.8333, | |
| "step": 286000 | |
| }, | |
| { | |
| "epoch": 15.434256520570045, | |
| "grad_norm": 0.40103474259376526, | |
| "learning_rate": 7.502647837599294e-05, | |
| "loss": 2.8327, | |
| "step": 287000 | |
| }, | |
| { | |
| "epoch": 15.488034417854262, | |
| "grad_norm": 0.42553824186325073, | |
| "learning_rate": 7.414386584289497e-05, | |
| "loss": 2.8343, | |
| "step": 288000 | |
| }, | |
| { | |
| "epoch": 15.541812315138477, | |
| "grad_norm": 0.4238561987876892, | |
| "learning_rate": 7.3261253309797e-05, | |
| "loss": 2.8346, | |
| "step": 289000 | |
| }, | |
| { | |
| "epoch": 15.595590212422694, | |
| "grad_norm": 0.44222867488861084, | |
| "learning_rate": 7.237952338923212e-05, | |
| "loss": 2.837, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 15.649368109706911, | |
| "grad_norm": 0.4435216784477234, | |
| "learning_rate": 7.149691085613415e-05, | |
| "loss": 2.8348, | |
| "step": 291000 | |
| }, | |
| { | |
| "epoch": 15.703146006991126, | |
| "grad_norm": 0.434276282787323, | |
| "learning_rate": 7.061518093556928e-05, | |
| "loss": 2.8376, | |
| "step": 292000 | |
| }, | |
| { | |
| "epoch": 15.756923904275343, | |
| "grad_norm": 0.4052608013153076, | |
| "learning_rate": 6.97325684024713e-05, | |
| "loss": 2.8421, | |
| "step": 293000 | |
| }, | |
| { | |
| "epoch": 15.81070180155956, | |
| "grad_norm": 0.4162592589855194, | |
| "learning_rate": 6.885083848190643e-05, | |
| "loss": 2.8413, | |
| "step": 294000 | |
| }, | |
| { | |
| "epoch": 15.864479698843775, | |
| "grad_norm": 0.4163593649864197, | |
| "learning_rate": 6.796822594880847e-05, | |
| "loss": 2.843, | |
| "step": 295000 | |
| }, | |
| { | |
| "epoch": 15.918257596127992, | |
| "grad_norm": 0.4112990200519562, | |
| "learning_rate": 6.70856134157105e-05, | |
| "loss": 2.8427, | |
| "step": 296000 | |
| }, | |
| { | |
| "epoch": 15.972035493412207, | |
| "grad_norm": 0.3960702121257782, | |
| "learning_rate": 6.620300088261252e-05, | |
| "loss": 2.8414, | |
| "step": 297000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.4098178943182842, | |
| "eval_loss": 3.3970065116882324, | |
| "eval_runtime": 154.1119, | |
| "eval_samples_per_second": 375.837, | |
| "eval_steps_per_second": 5.879, | |
| "step": 297520 | |
| }, | |
| { | |
| "epoch": 16.025813390696424, | |
| "grad_norm": 0.4249773919582367, | |
| "learning_rate": 6.532127096204766e-05, | |
| "loss": 2.8211, | |
| "step": 298000 | |
| }, | |
| { | |
| "epoch": 16.07959128798064, | |
| "grad_norm": 0.4380571246147156, | |
| "learning_rate": 6.443865842894968e-05, | |
| "loss": 2.8058, | |
| "step": 299000 | |
| }, | |
| { | |
| "epoch": 16.133369185264858, | |
| "grad_norm": 0.43307915329933167, | |
| "learning_rate": 6.355604589585172e-05, | |
| "loss": 2.8082, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 16.18714708254907, | |
| "grad_norm": 0.44826626777648926, | |
| "learning_rate": 6.267431597528685e-05, | |
| "loss": 2.8046, | |
| "step": 301000 | |
| }, | |
| { | |
| "epoch": 16.240924979833288, | |
| "grad_norm": 0.4037844240665436, | |
| "learning_rate": 6.179170344218888e-05, | |
| "loss": 2.8096, | |
| "step": 302000 | |
| }, | |
| { | |
| "epoch": 16.294702877117505, | |
| "grad_norm": 0.4280061721801758, | |
| "learning_rate": 6.0909973521623994e-05, | |
| "loss": 2.8112, | |
| "step": 303000 | |
| }, | |
| { | |
| "epoch": 16.34848077440172, | |
| "grad_norm": 0.43137866258621216, | |
| "learning_rate": 6.002736098852603e-05, | |
| "loss": 2.8104, | |
| "step": 304000 | |
| }, | |
| { | |
| "epoch": 16.40225867168594, | |
| "grad_norm": 0.41685307025909424, | |
| "learning_rate": 5.914563106796116e-05, | |
| "loss": 2.811, | |
| "step": 305000 | |
| }, | |
| { | |
| "epoch": 16.456036568970152, | |
| "grad_norm": 0.4230118989944458, | |
| "learning_rate": 5.8263018534863187e-05, | |
| "loss": 2.8155, | |
| "step": 306000 | |
| }, | |
| { | |
| "epoch": 16.50981446625437, | |
| "grad_norm": 0.4346941113471985, | |
| "learning_rate": 5.738040600176522e-05, | |
| "loss": 2.8177, | |
| "step": 307000 | |
| }, | |
| { | |
| "epoch": 16.563592363538586, | |
| "grad_norm": 0.46773964166641235, | |
| "learning_rate": 5.6497793468667254e-05, | |
| "loss": 2.8171, | |
| "step": 308000 | |
| }, | |
| { | |
| "epoch": 16.617370260822803, | |
| "grad_norm": 0.42793235182762146, | |
| "learning_rate": 5.561606354810238e-05, | |
| "loss": 2.815, | |
| "step": 309000 | |
| }, | |
| { | |
| "epoch": 16.67114815810702, | |
| "grad_norm": 0.43302974104881287, | |
| "learning_rate": 5.4734333627537504e-05, | |
| "loss": 2.8221, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 16.724926055391233, | |
| "grad_norm": 0.4403083622455597, | |
| "learning_rate": 5.3851721094439534e-05, | |
| "loss": 2.8215, | |
| "step": 311000 | |
| }, | |
| { | |
| "epoch": 16.77870395267545, | |
| "grad_norm": 0.4399563670158386, | |
| "learning_rate": 5.2969108561341564e-05, | |
| "loss": 2.8208, | |
| "step": 312000 | |
| }, | |
| { | |
| "epoch": 16.832481849959667, | |
| "grad_norm": 0.4378258287906647, | |
| "learning_rate": 5.208737864077669e-05, | |
| "loss": 2.8231, | |
| "step": 313000 | |
| }, | |
| { | |
| "epoch": 16.886259747243884, | |
| "grad_norm": 0.41973891854286194, | |
| "learning_rate": 5.1204766107678726e-05, | |
| "loss": 2.8237, | |
| "step": 314000 | |
| }, | |
| { | |
| "epoch": 16.9400376445281, | |
| "grad_norm": 0.4335421323776245, | |
| "learning_rate": 5.0322153574580757e-05, | |
| "loss": 2.8232, | |
| "step": 315000 | |
| }, | |
| { | |
| "epoch": 16.993815541812314, | |
| "grad_norm": 0.4244571924209595, | |
| "learning_rate": 4.944042365401588e-05, | |
| "loss": 2.8226, | |
| "step": 316000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.40916397801191307, | |
| "eval_loss": 3.411121129989624, | |
| "eval_runtime": 154.5975, | |
| "eval_samples_per_second": 374.657, | |
| "eval_steps_per_second": 5.86, | |
| "step": 316115 | |
| }, | |
| { | |
| "epoch": 17.04759343909653, | |
| "grad_norm": 0.4443401098251343, | |
| "learning_rate": 4.855781112091791e-05, | |
| "loss": 2.789, | |
| "step": 317000 | |
| }, | |
| { | |
| "epoch": 17.101371336380748, | |
| "grad_norm": 0.40948519110679626, | |
| "learning_rate": 4.767519858781995e-05, | |
| "loss": 2.7875, | |
| "step": 318000 | |
| }, | |
| { | |
| "epoch": 17.155149233664964, | |
| "grad_norm": 0.44507384300231934, | |
| "learning_rate": 4.6793468667255074e-05, | |
| "loss": 2.7909, | |
| "step": 319000 | |
| }, | |
| { | |
| "epoch": 17.20892713094918, | |
| "grad_norm": 0.44478124380111694, | |
| "learning_rate": 4.5910856134157104e-05, | |
| "loss": 2.7936, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 17.262705028233395, | |
| "grad_norm": 0.44282448291778564, | |
| "learning_rate": 4.502824360105913e-05, | |
| "loss": 2.7925, | |
| "step": 321000 | |
| }, | |
| { | |
| "epoch": 17.31648292551761, | |
| "grad_norm": 0.45248299837112427, | |
| "learning_rate": 4.414651368049426e-05, | |
| "loss": 2.7958, | |
| "step": 322000 | |
| }, | |
| { | |
| "epoch": 17.37026082280183, | |
| "grad_norm": 0.4552863836288452, | |
| "learning_rate": 4.3263901147396296e-05, | |
| "loss": 2.7993, | |
| "step": 323000 | |
| }, | |
| { | |
| "epoch": 17.424038720086045, | |
| "grad_norm": 0.46229588985443115, | |
| "learning_rate": 4.238128861429832e-05, | |
| "loss": 2.7969, | |
| "step": 324000 | |
| }, | |
| { | |
| "epoch": 17.477816617370262, | |
| "grad_norm": 0.4318771958351135, | |
| "learning_rate": 4.149955869373345e-05, | |
| "loss": 2.7997, | |
| "step": 325000 | |
| }, | |
| { | |
| "epoch": 17.531594514654476, | |
| "grad_norm": 0.45026683807373047, | |
| "learning_rate": 4.0616946160635475e-05, | |
| "loss": 2.7975, | |
| "step": 326000 | |
| }, | |
| { | |
| "epoch": 17.585372411938692, | |
| "grad_norm": 0.4539460837841034, | |
| "learning_rate": 3.9734333627537505e-05, | |
| "loss": 2.8004, | |
| "step": 327000 | |
| }, | |
| { | |
| "epoch": 17.63915030922291, | |
| "grad_norm": 0.44773465394973755, | |
| "learning_rate": 3.885260370697263e-05, | |
| "loss": 2.8014, | |
| "step": 328000 | |
| }, | |
| { | |
| "epoch": 17.692928206507126, | |
| "grad_norm": 0.443488210439682, | |
| "learning_rate": 3.796999117387467e-05, | |
| "loss": 2.804, | |
| "step": 329000 | |
| }, | |
| { | |
| "epoch": 17.746706103791343, | |
| "grad_norm": 0.4228520691394806, | |
| "learning_rate": 3.70873786407767e-05, | |
| "loss": 2.8023, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 17.800484001075557, | |
| "grad_norm": 0.43788018822669983, | |
| "learning_rate": 3.620564872021182e-05, | |
| "loss": 2.8027, | |
| "step": 331000 | |
| }, | |
| { | |
| "epoch": 17.854261898359773, | |
| "grad_norm": 0.4629572629928589, | |
| "learning_rate": 3.532303618711385e-05, | |
| "loss": 2.8044, | |
| "step": 332000 | |
| }, | |
| { | |
| "epoch": 17.90803979564399, | |
| "grad_norm": 0.4508598744869232, | |
| "learning_rate": 3.444042365401588e-05, | |
| "loss": 2.8069, | |
| "step": 333000 | |
| }, | |
| { | |
| "epoch": 17.961817692928207, | |
| "grad_norm": 0.4376344084739685, | |
| "learning_rate": 3.355869373345101e-05, | |
| "loss": 2.8063, | |
| "step": 334000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.4091488628312131, | |
| "eval_loss": 3.419936418533325, | |
| "eval_runtime": 154.2734, | |
| "eval_samples_per_second": 375.444, | |
| "eval_steps_per_second": 5.873, | |
| "step": 334710 | |
| }, | |
| { | |
| "epoch": 18.015595590212424, | |
| "grad_norm": 0.44690123200416565, | |
| "learning_rate": 3.2676081200353045e-05, | |
| "loss": 2.7994, | |
| "step": 335000 | |
| }, | |
| { | |
| "epoch": 18.069373487496637, | |
| "grad_norm": 0.4568174183368683, | |
| "learning_rate": 3.179435127978817e-05, | |
| "loss": 2.7778, | |
| "step": 336000 | |
| }, | |
| { | |
| "epoch": 18.123151384780854, | |
| "grad_norm": 0.4270166754722595, | |
| "learning_rate": 3.09117387466902e-05, | |
| "loss": 2.7778, | |
| "step": 337000 | |
| }, | |
| { | |
| "epoch": 18.17692928206507, | |
| "grad_norm": 0.4290498197078705, | |
| "learning_rate": 3.0030008826125325e-05, | |
| "loss": 2.7754, | |
| "step": 338000 | |
| }, | |
| { | |
| "epoch": 18.230707179349288, | |
| "grad_norm": 0.44628649950027466, | |
| "learning_rate": 2.914739629302736e-05, | |
| "loss": 2.7749, | |
| "step": 339000 | |
| }, | |
| { | |
| "epoch": 18.284485076633505, | |
| "grad_norm": 0.4463174641132355, | |
| "learning_rate": 2.8264783759929386e-05, | |
| "loss": 2.7784, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 18.33826297391772, | |
| "grad_norm": 0.46662789583206177, | |
| "learning_rate": 2.7383936451897613e-05, | |
| "loss": 2.7852, | |
| "step": 341000 | |
| }, | |
| { | |
| "epoch": 18.392040871201935, | |
| "grad_norm": 0.4375401437282562, | |
| "learning_rate": 2.6501323918799646e-05, | |
| "loss": 2.7807, | |
| "step": 342000 | |
| }, | |
| { | |
| "epoch": 18.445818768486152, | |
| "grad_norm": 0.4658895432949066, | |
| "learning_rate": 2.5618711385701673e-05, | |
| "loss": 2.7848, | |
| "step": 343000 | |
| }, | |
| { | |
| "epoch": 18.49959666577037, | |
| "grad_norm": 0.47445955872535706, | |
| "learning_rate": 2.47369814651368e-05, | |
| "loss": 2.7869, | |
| "step": 344000 | |
| }, | |
| { | |
| "epoch": 18.553374563054586, | |
| "grad_norm": 0.45969846844673157, | |
| "learning_rate": 2.385436893203883e-05, | |
| "loss": 2.7861, | |
| "step": 345000 | |
| }, | |
| { | |
| "epoch": 18.6071524603388, | |
| "grad_norm": 0.4380541741847992, | |
| "learning_rate": 2.2971756398940862e-05, | |
| "loss": 2.7824, | |
| "step": 346000 | |
| }, | |
| { | |
| "epoch": 18.660930357623016, | |
| "grad_norm": 0.45412492752075195, | |
| "learning_rate": 2.2089143865842892e-05, | |
| "loss": 2.7861, | |
| "step": 347000 | |
| }, | |
| { | |
| "epoch": 18.714708254907233, | |
| "grad_norm": 0.47223806381225586, | |
| "learning_rate": 2.120741394527802e-05, | |
| "loss": 2.7849, | |
| "step": 348000 | |
| }, | |
| { | |
| "epoch": 18.76848615219145, | |
| "grad_norm": 0.4475175440311432, | |
| "learning_rate": 2.032568402471315e-05, | |
| "loss": 2.7855, | |
| "step": 349000 | |
| }, | |
| { | |
| "epoch": 18.822264049475667, | |
| "grad_norm": 0.44210994243621826, | |
| "learning_rate": 1.944307149161518e-05, | |
| "loss": 2.7863, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 18.87604194675988, | |
| "grad_norm": 0.4564194977283478, | |
| "learning_rate": 1.856045895851721e-05, | |
| "loss": 2.7858, | |
| "step": 351000 | |
| }, | |
| { | |
| "epoch": 18.929819844044097, | |
| "grad_norm": 0.43993762135505676, | |
| "learning_rate": 1.7678729037952338e-05, | |
| "loss": 2.7868, | |
| "step": 352000 | |
| }, | |
| { | |
| "epoch": 18.983597741328314, | |
| "grad_norm": 0.42271167039871216, | |
| "learning_rate": 1.6796116504854368e-05, | |
| "loss": 2.7905, | |
| "step": 353000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.4092819435999537, | |
| "eval_loss": 3.423438310623169, | |
| "eval_runtime": 154.0851, | |
| "eval_samples_per_second": 375.903, | |
| "eval_steps_per_second": 5.88, | |
| "step": 353305 | |
| }, | |
| { | |
| "epoch": 19.03737563861253, | |
| "grad_norm": 0.4467901289463043, | |
| "learning_rate": 1.5914386584289496e-05, | |
| "loss": 2.7736, | |
| "step": 354000 | |
| }, | |
| { | |
| "epoch": 19.091153535896748, | |
| "grad_norm": 0.45330286026000977, | |
| "learning_rate": 1.5031774051191525e-05, | |
| "loss": 2.7671, | |
| "step": 355000 | |
| }, | |
| { | |
| "epoch": 19.14493143318096, | |
| "grad_norm": 0.451779305934906, | |
| "learning_rate": 1.4149161518093555e-05, | |
| "loss": 2.7674, | |
| "step": 356000 | |
| }, | |
| { | |
| "epoch": 19.198709330465178, | |
| "grad_norm": 0.462492436170578, | |
| "learning_rate": 1.3266548984995585e-05, | |
| "loss": 2.7679, | |
| "step": 357000 | |
| }, | |
| { | |
| "epoch": 19.252487227749395, | |
| "grad_norm": 0.47610774636268616, | |
| "learning_rate": 1.2383936451897616e-05, | |
| "loss": 2.7677, | |
| "step": 358000 | |
| }, | |
| { | |
| "epoch": 19.30626512503361, | |
| "grad_norm": 0.47243231534957886, | |
| "learning_rate": 1.1501323918799646e-05, | |
| "loss": 2.7695, | |
| "step": 359000 | |
| }, | |
| { | |
| "epoch": 19.36004302231783, | |
| "grad_norm": 0.47811663150787354, | |
| "learning_rate": 1.0618711385701676e-05, | |
| "loss": 2.7679, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 19.413820919602042, | |
| "grad_norm": 0.4842175543308258, | |
| "learning_rate": 9.737864077669903e-06, | |
| "loss": 2.7683, | |
| "step": 361000 | |
| }, | |
| { | |
| "epoch": 19.46759881688626, | |
| "grad_norm": 0.4452952444553375, | |
| "learning_rate": 8.855251544571931e-06, | |
| "loss": 2.7647, | |
| "step": 362000 | |
| }, | |
| { | |
| "epoch": 19.521376714170476, | |
| "grad_norm": 0.47025254368782043, | |
| "learning_rate": 7.972639011473961e-06, | |
| "loss": 2.7692, | |
| "step": 363000 | |
| }, | |
| { | |
| "epoch": 19.575154611454693, | |
| "grad_norm": 0.446979820728302, | |
| "learning_rate": 7.09090909090909e-06, | |
| "loss": 2.7708, | |
| "step": 364000 | |
| }, | |
| { | |
| "epoch": 19.62893250873891, | |
| "grad_norm": 0.4704710841178894, | |
| "learning_rate": 6.20829655781112e-06, | |
| "loss": 2.7699, | |
| "step": 365000 | |
| }, | |
| { | |
| "epoch": 19.682710406023123, | |
| "grad_norm": 0.4596230685710907, | |
| "learning_rate": 5.32568402471315e-06, | |
| "loss": 2.7718, | |
| "step": 366000 | |
| }, | |
| { | |
| "epoch": 19.73648830330734, | |
| "grad_norm": 0.4677480459213257, | |
| "learning_rate": 4.443954104148278e-06, | |
| "loss": 2.7665, | |
| "step": 367000 | |
| }, | |
| { | |
| "epoch": 19.790266200591557, | |
| "grad_norm": 0.45196905732154846, | |
| "learning_rate": 3.5613415710503085e-06, | |
| "loss": 2.7717, | |
| "step": 368000 | |
| }, | |
| { | |
| "epoch": 19.844044097875774, | |
| "grad_norm": 0.44339102506637573, | |
| "learning_rate": 2.6787290379523387e-06, | |
| "loss": 2.7711, | |
| "step": 369000 | |
| }, | |
| { | |
| "epoch": 19.89782199515999, | |
| "grad_norm": 0.46412837505340576, | |
| "learning_rate": 1.7961165048543689e-06, | |
| "loss": 2.7658, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 19.951599892444204, | |
| "grad_norm": 0.4551531970500946, | |
| "learning_rate": 9.143865842894968e-07, | |
| "loss": 2.7753, | |
| "step": 371000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.40922034084127873, | |
| "eval_loss": 3.4287688732147217, | |
| "eval_runtime": 154.0794, | |
| "eval_samples_per_second": 375.917, | |
| "eval_steps_per_second": 5.88, | |
| "step": 371900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 371900, | |
| "total_flos": 1.5669257538816e+18, | |
| "train_loss": 3.0594762622313207, | |
| "train_runtime": 80951.5095, | |
| "train_samples_per_second": 147.01, | |
| "train_steps_per_second": 4.594 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 371900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 5000, | |
| "total_flos": 1.5669257538816e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |