| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 371840, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.125e-05, | |
| "loss": 6.2273, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 6.25e-05, | |
| "loss": 5.0064, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.375e-05, | |
| "loss": 4.6786, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.000125, | |
| "loss": 4.4497, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00015625, | |
| "loss": 4.2995, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0001875, | |
| "loss": 4.1799, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00021875, | |
| "loss": 4.0734, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00025, | |
| "loss": 3.9794, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00028121875, | |
| "loss": 3.9125, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00031246875000000003, | |
| "loss": 3.8491, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00034368749999999997, | |
| "loss": 3.8003, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00037490625, | |
| "loss": 3.7623, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00040615625, | |
| "loss": 3.7286, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00043740625, | |
| "loss": 3.697, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00046865625, | |
| "loss": 3.6659, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.00049984375, | |
| "loss": 3.6414, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00053109375, | |
| "loss": 3.6203, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0005623125, | |
| "loss": 3.6046, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3593838920220195, | |
| "eval_loss": 3.783267021179199, | |
| "eval_runtime": 153.4166, | |
| "eval_samples_per_second": 377.528, | |
| "eval_steps_per_second": 5.899, | |
| "step": 18592 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00059353125, | |
| "loss": 3.5711, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00062478125, | |
| "loss": 3.5432, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.0006560312499999999, | |
| "loss": 3.5383, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00068728125, | |
| "loss": 3.5273, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.00071846875, | |
| "loss": 3.5186, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00074971875, | |
| "loss": 3.5029, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00078096875, | |
| "loss": 3.4884, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00081221875, | |
| "loss": 3.483, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00084346875, | |
| "loss": 3.4799, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00087471875, | |
| "loss": 3.4665, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0009059375, | |
| "loss": 3.4599, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.0009371875, | |
| "loss": 3.4558, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0009684062500000001, | |
| "loss": 3.4422, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.0009996562500000001, | |
| "loss": 3.4367, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.0009970927495291902, | |
| "loss": 3.426, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.000994150188323917, | |
| "loss": 3.4158, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.0009912076271186441, | |
| "loss": 3.4045, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.0009882680084745763, | |
| "loss": 3.3986, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.0009853254472693032, | |
| "loss": 3.3837, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3805308271991013, | |
| "eval_loss": 3.585395336151123, | |
| "eval_runtime": 153.8737, | |
| "eval_samples_per_second": 376.406, | |
| "eval_steps_per_second": 5.881, | |
| "step": 37184 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0009823858286252354, | |
| "loss": 3.3356, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0009794432674199625, | |
| "loss": 3.3244, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.0009765007062146893, | |
| "loss": 3.3239, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.0009735581450094162, | |
| "loss": 3.3237, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0009706155838041431, | |
| "loss": 3.3158, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0009676759651600754, | |
| "loss": 3.3155, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0009647334039548023, | |
| "loss": 3.3056, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.0009617967278719397, | |
| "loss": 3.2987, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.0009588541666666667, | |
| "loss": 3.2986, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0009559116054613936, | |
| "loss": 3.2939, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.0009529719868173258, | |
| "loss": 3.2838, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.0009500294256120528, | |
| "loss": 3.2825, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.000947089806967985, | |
| "loss": 3.2764, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.0009441472457627119, | |
| "loss": 3.2789, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0009412046845574388, | |
| "loss": 3.2675, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.000938265065913371, | |
| "loss": 3.2626, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.000935322504708098, | |
| "loss": 3.2634, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.0009323799435028248, | |
| "loss": 3.26, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.3930830410348331, | |
| "eval_loss": 3.4487695693969727, | |
| "eval_runtime": 154.1089, | |
| "eval_samples_per_second": 375.832, | |
| "eval_steps_per_second": 5.872, | |
| "step": 55776 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.0009294403248587571, | |
| "loss": 3.2383, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.0009265007062146893, | |
| "loss": 3.1912, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.0009235581450094163, | |
| "loss": 3.1918, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.0009206155838041431, | |
| "loss": 3.194, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.0009176759651600753, | |
| "loss": 3.1957, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.0009147334039548023, | |
| "loss": 3.197, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0009117908427495291, | |
| "loss": 3.1975, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.0009088482815442561, | |
| "loss": 3.1927, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.0009059057203389832, | |
| "loss": 3.1992, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.00090296315913371, | |
| "loss": 3.1941, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.000900020597928437, | |
| "loss": 3.1912, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0008970839218455744, | |
| "loss": 3.1912, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 0.0008941413606403014, | |
| "loss": 3.1841, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.0008911987994350282, | |
| "loss": 3.1878, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 0.0008882562382297552, | |
| "loss": 3.1869, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.0008853136770244821, | |
| "loss": 3.1847, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.0008823740583804143, | |
| "loss": 3.1828, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.0008794314971751413, | |
| "loss": 3.1792, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 0.0008764918785310735, | |
| "loss": 3.1824, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.398565875878046, | |
| "eval_loss": 3.4153025150299072, | |
| "eval_runtime": 154.2791, | |
| "eval_samples_per_second": 375.417, | |
| "eval_steps_per_second": 5.866, | |
| "step": 74368 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.0008735493173258003, | |
| "loss": 3.1395, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.0008706096986817325, | |
| "loss": 3.1148, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 0.0008676671374764595, | |
| "loss": 3.1217, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 0.0008647275188323918, | |
| "loss": 3.1234, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.0008617849576271186, | |
| "loss": 3.125, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 0.0008588423964218456, | |
| "loss": 3.1258, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.0008559027777777778, | |
| "loss": 3.1259, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 0.0008529602165725048, | |
| "loss": 3.1217, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 0.0008500205979284368, | |
| "loss": 3.1289, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.0008470809792843691, | |
| "loss": 3.1277, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.0008441384180790961, | |
| "loss": 3.1229, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.000841195856873823, | |
| "loss": 3.126, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 0.0008382532956685499, | |
| "loss": 3.1276, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.0008353136770244821, | |
| "loss": 3.1249, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.0008323740583804143, | |
| "loss": 3.1257, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.0008294314971751412, | |
| "loss": 3.1257, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.0008264889359698681, | |
| "loss": 3.1228, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.0008235463747645952, | |
| "loss": 3.1238, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.40275097726376624, | |
| "eval_loss": 3.3852765560150146, | |
| "eval_runtime": 153.7791, | |
| "eval_samples_per_second": 376.638, | |
| "eval_steps_per_second": 5.885, | |
| "step": 92960 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0008206067561205274, | |
| "loss": 3.1162, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 0.0008176671374764595, | |
| "loss": 3.0546, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.0008147245762711864, | |
| "loss": 3.063, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.0008117820150659134, | |
| "loss": 3.0632, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.0008088394538606403, | |
| "loss": 3.069, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.0008058998352165726, | |
| "loss": 3.0673, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 0.0008029602165725047, | |
| "loss": 3.0733, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.0008000176553672317, | |
| "loss": 3.0762, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.0007970750941619586, | |
| "loss": 3.0755, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 0.0007941325329566855, | |
| "loss": 3.0736, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 0.0007911929143126177, | |
| "loss": 3.0788, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 0.0007882503531073446, | |
| "loss": 3.0782, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 0.0007853077919020715, | |
| "loss": 3.0806, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.0007823681732580038, | |
| "loss": 3.0796, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.0007794256120527307, | |
| "loss": 3.0806, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.0007764830508474577, | |
| "loss": 3.0824, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.0007735434322033898, | |
| "loss": 3.0785, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 0.0007706008709981168, | |
| "loss": 3.0839, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 0.0007676612523540489, | |
| "loss": 3.0837, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.4060002487036941, | |
| "eval_loss": 3.3511743545532227, | |
| "eval_runtime": 154.3771, | |
| "eval_samples_per_second": 375.179, | |
| "eval_steps_per_second": 5.862, | |
| "step": 111552 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 0.0007647186911487759, | |
| "loss": 3.0507, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.0007617761299435028, | |
| "loss": 3.0175, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 0.0007588365112994351, | |
| "loss": 3.0242, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 0.000755893950094162, | |
| "loss": 3.022, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 0.0007529513888888889, | |
| "loss": 3.0227, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 0.0007500088276836158, | |
| "loss": 3.032, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 0.0007470662664783428, | |
| "loss": 3.0324, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.000744126647834275, | |
| "loss": 3.0343, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.0007411870291902072, | |
| "loss": 3.0381, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 0.0007382444679849341, | |
| "loss": 3.0422, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.0007353019067796611, | |
| "loss": 3.0362, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.000732359345574388, | |
| "loss": 3.0415, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.0007294167843691149, | |
| "loss": 3.0433, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.0007264801082862523, | |
| "loss": 3.0398, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0007235375470809793, | |
| "loss": 3.0429, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.0007205979284369116, | |
| "loss": 3.0465, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 0.0007176553672316384, | |
| "loss": 3.0435, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.0007147128060263654, | |
| "loss": 3.0413, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 0.0007117702448210924, | |
| "loss": 3.0442, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.40654595915952124, | |
| "eval_loss": 3.3564212322235107, | |
| "eval_runtime": 154.4221, | |
| "eval_samples_per_second": 375.069, | |
| "eval_steps_per_second": 5.861, | |
| "step": 130144 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 0.0007088306261770245, | |
| "loss": 2.9796, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.0007058910075329566, | |
| "loss": 2.9806, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "learning_rate": 0.0007029484463276836, | |
| "loss": 2.9855, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 0.0007000058851224106, | |
| "loss": 2.9876, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.0006970633239171374, | |
| "loss": 3.0, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 0.0006941237052730697, | |
| "loss": 2.9999, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.0006911811440677967, | |
| "loss": 2.9999, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 0.0006882385828625236, | |
| "loss": 2.9978, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 0.0006852960216572505, | |
| "loss": 3.0077, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 0.0006823564030131827, | |
| "loss": 3.0006, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 0.0006794138418079096, | |
| "loss": 3.0076, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 0.0006764771657250472, | |
| "loss": 3.0092, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 0.000673534604519774, | |
| "loss": 3.0084, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "learning_rate": 0.000670592043314501, | |
| "loss": 3.0095, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 0.0006676524246704332, | |
| "loss": 3.0106, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 0.00066470986346516, | |
| "loss": 3.012, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 0.000661767302259887, | |
| "loss": 3.015, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 0.000658824741054614, | |
| "loss": 3.0168, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4083005227413059, | |
| "eval_loss": 3.343769073486328, | |
| "eval_runtime": 154.0462, | |
| "eval_samples_per_second": 375.985, | |
| "eval_steps_per_second": 5.875, | |
| "step": 148736 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 0.0006558851224105461, | |
| "loss": 2.9953, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.0006529425612052731, | |
| "loss": 2.9443, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 0.0006500000000000001, | |
| "loss": 2.9507, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 0.0006470603813559323, | |
| "loss": 2.9529, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 0.0006441178201506591, | |
| "loss": 2.9574, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 0.0006411782015065913, | |
| "loss": 2.9675, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 0.0006382356403013183, | |
| "loss": 2.9733, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 0.0006352930790960451, | |
| "loss": 2.9748, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 0.0006323505178907721, | |
| "loss": 2.974, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 0.0006294138418079096, | |
| "loss": 2.9764, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 0.0006264712806026366, | |
| "loss": 2.9803, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 0.0006235287193973634, | |
| "loss": 2.9803, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 0.0006205891007532956, | |
| "loss": 2.9798, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 0.0006176465395480226, | |
| "loss": 2.9829, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 0.0006147039783427495, | |
| "loss": 2.9814, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 0.0006117614171374764, | |
| "loss": 2.9844, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 0.0006088217984934087, | |
| "loss": 2.9853, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 0.0006058821798493409, | |
| "loss": 2.9841, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 0.0006029396186440678, | |
| "loss": 2.9792, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.4090171414083387, | |
| "eval_loss": 3.3494999408721924, | |
| "eval_runtime": 153.9334, | |
| "eval_samples_per_second": 376.26, | |
| "eval_steps_per_second": 5.879, | |
| "step": 167328 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 0.0006, | |
| "loss": 2.9453, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 0.0005970574387947269, | |
| "loss": 2.9245, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "learning_rate": 0.0005941148775894539, | |
| "loss": 2.927, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 0.0005911752589453861, | |
| "loss": 2.9364, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 0.000588232697740113, | |
| "loss": 2.9349, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 0.00058529013653484, | |
| "loss": 2.9388, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "learning_rate": 0.0005823475753295669, | |
| "loss": 2.947, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 0.000579407956685499, | |
| "loss": 2.9452, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 0.000576465395480226, | |
| "loss": 2.9474, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 0.0005735228342749529, | |
| "loss": 2.9493, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 0.0005705802730696798, | |
| "loss": 2.9519, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "learning_rate": 0.0005676406544256121, | |
| "loss": 2.9557, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 0.0005646980932203391, | |
| "loss": 2.9564, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 0.0005617614171374764, | |
| "loss": 2.9516, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 0.0005588188559322034, | |
| "loss": 2.9541, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "learning_rate": 0.0005558762947269303, | |
| "loss": 2.9574, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 0.0005529337335216572, | |
| "loss": 2.9585, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 0.0005499911723163841, | |
| "loss": 2.9607, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4090612120791528, | |
| "eval_loss": 3.357858896255493, | |
| "eval_runtime": 153.9438, | |
| "eval_samples_per_second": 376.235, | |
| "eval_steps_per_second": 5.879, | |
| "step": 185920 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0005470515536723164, | |
| "loss": 2.9524, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 0.0005441089924670434, | |
| "loss": 2.892, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 0.0005411664312617703, | |
| "loss": 2.9011, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.0005382297551789077, | |
| "loss": 2.9026, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 0.0005352871939736346, | |
| "loss": 2.9121, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "learning_rate": 0.0005323446327683616, | |
| "loss": 2.9159, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 10.33, | |
| "learning_rate": 0.0005294020715630885, | |
| "loss": 2.9202, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 0.0005264624529190208, | |
| "loss": 2.918, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "learning_rate": 0.000523522834274953, | |
| "loss": 2.9229, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 0.0005205802730696798, | |
| "loss": 2.9255, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "learning_rate": 0.0005176377118644068, | |
| "loss": 2.9251, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 10.6, | |
| "learning_rate": 0.0005146951506591337, | |
| "loss": 2.929, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "learning_rate": 0.0005117555320150659, | |
| "loss": 2.9292, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 10.7, | |
| "learning_rate": 0.0005088129708097928, | |
| "loss": 2.9324, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 0.0005058704096045199, | |
| "loss": 2.9327, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 0.0005029278483992468, | |
| "loss": 2.9359, | |
| "step": 201000 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "learning_rate": 0.0004999852871939737, | |
| "loss": 2.9319, | |
| "step": 202000 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "learning_rate": 0.0004970456685499059, | |
| "loss": 2.9368, | |
| "step": 203000 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 0.0004941031073446327, | |
| "loss": 2.9363, | |
| "step": 204000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.41160011267580654, | |
| "eval_loss": 3.3419740200042725, | |
| "eval_runtime": 153.6415, | |
| "eval_samples_per_second": 376.975, | |
| "eval_steps_per_second": 5.89, | |
| "step": 204512 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 0.000491163488700565, | |
| "loss": 2.9059, | |
| "step": 205000 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 0.0004882238700564972, | |
| "loss": 2.8766, | |
| "step": 206000 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 0.0004852813088512241, | |
| "loss": 2.8806, | |
| "step": 207000 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "learning_rate": 0.000482338747645951, | |
| "loss": 2.8831, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 0.00047939618644067793, | |
| "loss": 2.8891, | |
| "step": 209000 | |
| }, | |
| { | |
| "epoch": 11.3, | |
| "learning_rate": 0.0004764565677966102, | |
| "loss": 2.8895, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 0.0004735140065913371, | |
| "loss": 2.9, | |
| "step": 211000 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "learning_rate": 0.00047057144538606405, | |
| "loss": 2.8961, | |
| "step": 212000 | |
| }, | |
| { | |
| "epoch": 11.46, | |
| "learning_rate": 0.00046763182674199624, | |
| "loss": 2.9009, | |
| "step": 213000 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 0.0004646892655367232, | |
| "loss": 2.9049, | |
| "step": 214000 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 0.00046174964689265534, | |
| "loss": 2.9065, | |
| "step": 215000 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 0.0004588070856873823, | |
| "loss": 2.9058, | |
| "step": 216000 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 0.0004558674670433145, | |
| "loss": 2.9093, | |
| "step": 217000 | |
| }, | |
| { | |
| "epoch": 11.73, | |
| "learning_rate": 0.00045292490583804146, | |
| "loss": 2.9111, | |
| "step": 218000 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 0.00044998234463276837, | |
| "loss": 2.9121, | |
| "step": 219000 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "learning_rate": 0.00044704272598870056, | |
| "loss": 2.9088, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "learning_rate": 0.0004441001647834275, | |
| "loss": 2.9156, | |
| "step": 221000 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 0.00044115760357815443, | |
| "loss": 2.9111, | |
| "step": 222000 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 0.0004382179849340867, | |
| "loss": 2.9148, | |
| "step": 223000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4105683484039128, | |
| "eval_loss": 3.363103151321411, | |
| "eval_runtime": 153.7306, | |
| "eval_samples_per_second": 376.756, | |
| "eval_steps_per_second": 5.887, | |
| "step": 223104 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 0.0004352754237288136, | |
| "loss": 2.8595, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 0.0004323358050847458, | |
| "loss": 2.8587, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 0.0004293932438794727, | |
| "loss": 2.8609, | |
| "step": 226000 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 0.00042645362523540493, | |
| "loss": 2.8685, | |
| "step": 227000 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 0.00042351106403013184, | |
| "loss": 2.8705, | |
| "step": 228000 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 0.0004205714453860641, | |
| "loss": 2.8726, | |
| "step": 229000 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "learning_rate": 0.000417628884180791, | |
| "loss": 2.8792, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 12.42, | |
| "learning_rate": 0.0004146863229755179, | |
| "loss": 2.8785, | |
| "step": 231000 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 0.0004117437617702448, | |
| "loss": 2.8824, | |
| "step": 232000 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "learning_rate": 0.000408804143126177, | |
| "loss": 2.8818, | |
| "step": 233000 | |
| }, | |
| { | |
| "epoch": 12.59, | |
| "learning_rate": 0.00040586158192090397, | |
| "loss": 2.8892, | |
| "step": 234000 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "learning_rate": 0.00040292490583804144, | |
| "loss": 2.8847, | |
| "step": 235000 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 0.0003999823446327684, | |
| "loss": 2.885, | |
| "step": 236000 | |
| }, | |
| { | |
| "epoch": 12.75, | |
| "learning_rate": 0.00039704272598870054, | |
| "loss": 2.8872, | |
| "step": 237000 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.0003941001647834275, | |
| "loss": 2.8898, | |
| "step": 238000 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 0.00039115760357815447, | |
| "loss": 2.8938, | |
| "step": 239000 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "learning_rate": 0.0003882179849340866, | |
| "loss": 2.8883, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "learning_rate": 0.00038527542372881357, | |
| "loss": 2.893, | |
| "step": 241000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.4106221603053184, | |
| "eval_loss": 3.3609366416931152, | |
| "eval_runtime": 154.0852, | |
| "eval_samples_per_second": 375.889, | |
| "eval_steps_per_second": 5.873, | |
| "step": 241696 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 0.0003823328625235405, | |
| "loss": 2.8725, | |
| "step": 242000 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 0.0003793903013182674, | |
| "loss": 2.8384, | |
| "step": 243000 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 0.00037644774011299435, | |
| "loss": 2.8434, | |
| "step": 244000 | |
| }, | |
| { | |
| "epoch": 13.18, | |
| "learning_rate": 0.00037350812146892654, | |
| "loss": 2.8456, | |
| "step": 245000 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 0.0003705655602636535, | |
| "loss": 2.8473, | |
| "step": 246000 | |
| }, | |
| { | |
| "epoch": 13.29, | |
| "learning_rate": 0.0003676259416195857, | |
| "loss": 2.8551, | |
| "step": 247000 | |
| }, | |
| { | |
| "epoch": 13.34, | |
| "learning_rate": 0.00036468338041431266, | |
| "loss": 2.8529, | |
| "step": 248000 | |
| }, | |
| { | |
| "epoch": 13.39, | |
| "learning_rate": 0.00036174081920903957, | |
| "loss": 2.8597, | |
| "step": 249000 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 0.0003588012005649718, | |
| "loss": 2.8603, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 0.0003558586393596987, | |
| "loss": 2.8582, | |
| "step": 251000 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "learning_rate": 0.00035291607815442563, | |
| "loss": 2.8672, | |
| "step": 252000 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "learning_rate": 0.0003499764595103578, | |
| "loss": 2.8639, | |
| "step": 253000 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 0.00034703389830508473, | |
| "loss": 2.8686, | |
| "step": 254000 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 0.00034409722222222225, | |
| "loss": 2.8681, | |
| "step": 255000 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 0.00034115466101694916, | |
| "loss": 2.8689, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "learning_rate": 0.00033821209981167613, | |
| "loss": 2.8687, | |
| "step": 257000 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "learning_rate": 0.00033526953860640304, | |
| "loss": 2.8727, | |
| "step": 258000 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 0.00033232697740112995, | |
| "loss": 2.8695, | |
| "step": 259000 | |
| }, | |
| { | |
| "epoch": 13.98, | |
| "learning_rate": 0.0003293903013182674, | |
| "loss": 2.8729, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.4100738969752673, | |
| "eval_loss": 3.3806495666503906, | |
| "eval_runtime": 153.899, | |
| "eval_samples_per_second": 376.344, | |
| "eval_steps_per_second": 5.88, | |
| "step": 260288 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 0.0003264477401129944, | |
| "loss": 2.8286, | |
| "step": 261000 | |
| }, | |
| { | |
| "epoch": 14.09, | |
| "learning_rate": 0.0003235051789077213, | |
| "loss": 2.8289, | |
| "step": 262000 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 0.0003205626177024482, | |
| "loss": 2.8278, | |
| "step": 263000 | |
| }, | |
| { | |
| "epoch": 14.2, | |
| "learning_rate": 0.0003176200564971751, | |
| "loss": 2.8306, | |
| "step": 264000 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 0.00031468043785310736, | |
| "loss": 2.8314, | |
| "step": 265000 | |
| }, | |
| { | |
| "epoch": 14.31, | |
| "learning_rate": 0.00031173787664783427, | |
| "loss": 2.8352, | |
| "step": 266000 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "learning_rate": 0.0003087982580037665, | |
| "loss": 2.8342, | |
| "step": 267000 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 0.0003058556967984934, | |
| "loss": 2.839, | |
| "step": 268000 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 0.0003029160781544256, | |
| "loss": 2.8452, | |
| "step": 269000 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "learning_rate": 0.0002999735169491525, | |
| "loss": 2.8406, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "learning_rate": 0.0002970309557438795, | |
| "loss": 2.8424, | |
| "step": 271000 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 0.0002940883945386064, | |
| "loss": 2.8431, | |
| "step": 272000 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 0.0002911517184557439, | |
| "loss": 2.8441, | |
| "step": 273000 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "learning_rate": 0.00028820915725047083, | |
| "loss": 2.8519, | |
| "step": 274000 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "learning_rate": 0.00028526659604519774, | |
| "loss": 2.8516, | |
| "step": 275000 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "learning_rate": 0.00028232697740112993, | |
| "loss": 2.8511, | |
| "step": 276000 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "learning_rate": 0.00027938441619585684, | |
| "loss": 2.8489, | |
| "step": 277000 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "learning_rate": 0.0002764418549905838, | |
| "loss": 2.8543, | |
| "step": 278000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.4111850690717071, | |
| "eval_loss": 3.3685128688812256, | |
| "eval_runtime": 153.9585, | |
| "eval_samples_per_second": 376.199, | |
| "eval_steps_per_second": 5.878, | |
| "step": 278880 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 0.00027349929378531077, | |
| "loss": 2.8455, | |
| "step": 279000 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "learning_rate": 0.00027055967514124296, | |
| "loss": 2.8025, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "learning_rate": 0.00026762005649717515, | |
| "loss": 2.8063, | |
| "step": 281000 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 0.0002646774952919021, | |
| "loss": 2.8132, | |
| "step": 282000 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "learning_rate": 0.000261734934086629, | |
| "loss": 2.817, | |
| "step": 283000 | |
| }, | |
| { | |
| "epoch": 15.28, | |
| "learning_rate": 0.00025879237288135593, | |
| "loss": 2.8158, | |
| "step": 284000 | |
| }, | |
| { | |
| "epoch": 15.33, | |
| "learning_rate": 0.0002558527542372882, | |
| "loss": 2.8136, | |
| "step": 285000 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 0.0002529101930320151, | |
| "loss": 2.8203, | |
| "step": 286000 | |
| }, | |
| { | |
| "epoch": 15.44, | |
| "learning_rate": 0.000249967631826742, | |
| "loss": 2.8201, | |
| "step": 287000 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 0.0002470250706214689, | |
| "loss": 2.8247, | |
| "step": 288000 | |
| }, | |
| { | |
| "epoch": 15.54, | |
| "learning_rate": 0.00024408545197740115, | |
| "loss": 2.8271, | |
| "step": 289000 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 0.00024114583333333334, | |
| "loss": 2.8281, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "learning_rate": 0.00023820327212806025, | |
| "loss": 2.8238, | |
| "step": 291000 | |
| }, | |
| { | |
| "epoch": 15.71, | |
| "learning_rate": 0.00023526071092278719, | |
| "loss": 2.8279, | |
| "step": 292000 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "learning_rate": 0.00023231814971751415, | |
| "loss": 2.8282, | |
| "step": 293000 | |
| }, | |
| { | |
| "epoch": 15.81, | |
| "learning_rate": 0.00022937558851224106, | |
| "loss": 2.829, | |
| "step": 294000 | |
| }, | |
| { | |
| "epoch": 15.87, | |
| "learning_rate": 0.00022643596986817328, | |
| "loss": 2.834, | |
| "step": 295000 | |
| }, | |
| { | |
| "epoch": 15.92, | |
| "learning_rate": 0.0002234934086629002, | |
| "loss": 2.8335, | |
| "step": 296000 | |
| }, | |
| { | |
| "epoch": 15.97, | |
| "learning_rate": 0.00022055379001883238, | |
| "loss": 2.8352, | |
| "step": 297000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.4118599012185473, | |
| "eval_loss": 3.3733930587768555, | |
| "eval_runtime": 153.6522, | |
| "eval_samples_per_second": 376.949, | |
| "eval_steps_per_second": 5.89, | |
| "step": 297472 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "learning_rate": 0.00021761122881355931, | |
| "loss": 2.8101, | |
| "step": 298000 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "learning_rate": 0.00021467161016949153, | |
| "loss": 2.7924, | |
| "step": 299000 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 0.00021172904896421847, | |
| "loss": 2.7933, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "learning_rate": 0.00020878943032015066, | |
| "loss": 2.797, | |
| "step": 301000 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "learning_rate": 0.0002058468691148776, | |
| "loss": 2.8004, | |
| "step": 302000 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "learning_rate": 0.0002029043079096045, | |
| "loss": 2.8006, | |
| "step": 303000 | |
| }, | |
| { | |
| "epoch": 16.35, | |
| "learning_rate": 0.00019996468926553675, | |
| "loss": 2.8037, | |
| "step": 304000 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "learning_rate": 0.00019702212806026366, | |
| "loss": 2.8037, | |
| "step": 305000 | |
| }, | |
| { | |
| "epoch": 16.46, | |
| "learning_rate": 0.0001940795668549906, | |
| "loss": 2.7998, | |
| "step": 306000 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "learning_rate": 0.00019113994821092279, | |
| "loss": 2.8083, | |
| "step": 307000 | |
| }, | |
| { | |
| "epoch": 16.57, | |
| "learning_rate": 0.00018819738700564972, | |
| "loss": 2.811, | |
| "step": 308000 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 0.00018525482580037663, | |
| "loss": 2.8054, | |
| "step": 309000 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 0.00018231226459510357, | |
| "loss": 2.8115, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 16.73, | |
| "learning_rate": 0.0001793726459510358, | |
| "loss": 2.8121, | |
| "step": 311000 | |
| }, | |
| { | |
| "epoch": 16.78, | |
| "learning_rate": 0.00017643008474576272, | |
| "loss": 2.8128, | |
| "step": 312000 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "learning_rate": 0.0001734904661016949, | |
| "loss": 2.8079, | |
| "step": 313000 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "learning_rate": 0.00017054790489642185, | |
| "loss": 2.8118, | |
| "step": 314000 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "learning_rate": 0.00016760828625235404, | |
| "loss": 2.8115, | |
| "step": 315000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 0.000164665725047081, | |
| "loss": 2.8131, | |
| "step": 316000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.4114817399288944, | |
| "eval_loss": 3.3759450912475586, | |
| "eval_runtime": 153.9374, | |
| "eval_samples_per_second": 376.25, | |
| "eval_steps_per_second": 5.879, | |
| "step": 316064 | |
| }, | |
| { | |
| "epoch": 17.05, | |
| "learning_rate": 0.0001617261064030132, | |
| "loss": 2.7776, | |
| "step": 317000 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "learning_rate": 0.0001587835451977401, | |
| "loss": 2.7771, | |
| "step": 318000 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "learning_rate": 0.00015584392655367232, | |
| "loss": 2.781, | |
| "step": 319000 | |
| }, | |
| { | |
| "epoch": 17.21, | |
| "learning_rate": 0.00015290136534839923, | |
| "loss": 2.7831, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 17.27, | |
| "learning_rate": 0.00014995880414312617, | |
| "loss": 2.7817, | |
| "step": 321000 | |
| }, | |
| { | |
| "epoch": 17.32, | |
| "learning_rate": 0.00014701624293785313, | |
| "loss": 2.7884, | |
| "step": 322000 | |
| }, | |
| { | |
| "epoch": 17.37, | |
| "learning_rate": 0.00014407662429378532, | |
| "loss": 2.7868, | |
| "step": 323000 | |
| }, | |
| { | |
| "epoch": 17.43, | |
| "learning_rate": 0.00014113406308851223, | |
| "loss": 2.791, | |
| "step": 324000 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "learning_rate": 0.00013819444444444445, | |
| "loss": 2.7901, | |
| "step": 325000 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "learning_rate": 0.00013525482580037666, | |
| "loss": 2.7897, | |
| "step": 326000 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 0.0001323122645951036, | |
| "loss": 2.7896, | |
| "step": 327000 | |
| }, | |
| { | |
| "epoch": 17.64, | |
| "learning_rate": 0.0001293697033898305, | |
| "loss": 2.791, | |
| "step": 328000 | |
| }, | |
| { | |
| "epoch": 17.7, | |
| "learning_rate": 0.0001264300847457627, | |
| "loss": 2.7875, | |
| "step": 329000 | |
| }, | |
| { | |
| "epoch": 17.75, | |
| "learning_rate": 0.00012348752354048964, | |
| "loss": 2.7925, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 17.8, | |
| "learning_rate": 0.00012054496233521658, | |
| "loss": 2.7942, | |
| "step": 331000 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 0.00011760240112994351, | |
| "loss": 2.7953, | |
| "step": 332000 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "learning_rate": 0.0001146627824858757, | |
| "loss": 2.7957, | |
| "step": 333000 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "learning_rate": 0.00011172022128060264, | |
| "loss": 2.7949, | |
| "step": 334000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.41107556420367825, | |
| "eval_loss": 3.3841803073883057, | |
| "eval_runtime": 153.7647, | |
| "eval_samples_per_second": 376.673, | |
| "eval_steps_per_second": 5.886, | |
| "step": 334656 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 0.00010877766007532956, | |
| "loss": 2.7843, | |
| "step": 335000 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 0.0001058350988700565, | |
| "loss": 2.7685, | |
| "step": 336000 | |
| }, | |
| { | |
| "epoch": 18.13, | |
| "learning_rate": 0.00010289842278719398, | |
| "loss": 2.7677, | |
| "step": 337000 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 9.995586158192091e-05, | |
| "loss": 2.7716, | |
| "step": 338000 | |
| }, | |
| { | |
| "epoch": 18.23, | |
| "learning_rate": 9.701330037664783e-05, | |
| "loss": 2.766, | |
| "step": 339000 | |
| }, | |
| { | |
| "epoch": 18.29, | |
| "learning_rate": 9.407073917137477e-05, | |
| "loss": 2.7697, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "learning_rate": 9.112817796610169e-05, | |
| "loss": 2.7682, | |
| "step": 341000 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 8.81885593220339e-05, | |
| "loss": 2.7741, | |
| "step": 342000 | |
| }, | |
| { | |
| "epoch": 18.45, | |
| "learning_rate": 8.524599811676083e-05, | |
| "loss": 2.7741, | |
| "step": 343000 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "learning_rate": 8.230637947269304e-05, | |
| "loss": 2.7731, | |
| "step": 344000 | |
| }, | |
| { | |
| "epoch": 18.56, | |
| "learning_rate": 7.936381826741996e-05, | |
| "loss": 2.7749, | |
| "step": 345000 | |
| }, | |
| { | |
| "epoch": 18.61, | |
| "learning_rate": 7.642419962335216e-05, | |
| "loss": 2.7719, | |
| "step": 346000 | |
| }, | |
| { | |
| "epoch": 18.66, | |
| "learning_rate": 7.34816384180791e-05, | |
| "loss": 2.7739, | |
| "step": 347000 | |
| }, | |
| { | |
| "epoch": 18.72, | |
| "learning_rate": 7.053907721280602e-05, | |
| "loss": 2.7731, | |
| "step": 348000 | |
| }, | |
| { | |
| "epoch": 18.77, | |
| "learning_rate": 6.759945856873823e-05, | |
| "loss": 2.7745, | |
| "step": 349000 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "learning_rate": 6.465689736346516e-05, | |
| "loss": 2.7739, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 18.88, | |
| "learning_rate": 6.171727871939737e-05, | |
| "loss": 2.7757, | |
| "step": 351000 | |
| }, | |
| { | |
| "epoch": 18.93, | |
| "learning_rate": 5.87747175141243e-05, | |
| "loss": 2.778, | |
| "step": 352000 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 5.583215630885122e-05, | |
| "loss": 2.7756, | |
| "step": 353000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.411504581435109, | |
| "eval_loss": 3.3892807960510254, | |
| "eval_runtime": 153.8847, | |
| "eval_samples_per_second": 376.379, | |
| "eval_steps_per_second": 5.881, | |
| "step": 353248 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 5.2892537664783424e-05, | |
| "loss": 2.7622, | |
| "step": 354000 | |
| }, | |
| { | |
| "epoch": 19.09, | |
| "learning_rate": 4.995291902071563e-05, | |
| "loss": 2.7549, | |
| "step": 355000 | |
| }, | |
| { | |
| "epoch": 19.15, | |
| "learning_rate": 4.7010357815442564e-05, | |
| "loss": 2.7563, | |
| "step": 356000 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 4.406779661016949e-05, | |
| "loss": 2.7579, | |
| "step": 357000 | |
| }, | |
| { | |
| "epoch": 19.26, | |
| "learning_rate": 4.11281779661017e-05, | |
| "loss": 2.7567, | |
| "step": 358000 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "learning_rate": 3.818561676082863e-05, | |
| "loss": 2.7588, | |
| "step": 359000 | |
| }, | |
| { | |
| "epoch": 19.36, | |
| "learning_rate": 3.524305555555556e-05, | |
| "loss": 2.7607, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 19.42, | |
| "learning_rate": 3.230049435028248e-05, | |
| "loss": 2.7583, | |
| "step": 361000 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "learning_rate": 2.936087570621469e-05, | |
| "loss": 2.7563, | |
| "step": 362000 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "learning_rate": 2.641831450094162e-05, | |
| "loss": 2.7584, | |
| "step": 363000 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "learning_rate": 2.347575329566855e-05, | |
| "loss": 2.7595, | |
| "step": 364000 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "learning_rate": 2.0536134651600753e-05, | |
| "loss": 2.7561, | |
| "step": 365000 | |
| }, | |
| { | |
| "epoch": 19.69, | |
| "learning_rate": 1.7593573446327684e-05, | |
| "loss": 2.7576, | |
| "step": 366000 | |
| }, | |
| { | |
| "epoch": 19.74, | |
| "learning_rate": 1.4653954802259888e-05, | |
| "loss": 2.7586, | |
| "step": 367000 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "learning_rate": 1.1711393596986819e-05, | |
| "loss": 2.762, | |
| "step": 368000 | |
| }, | |
| { | |
| "epoch": 19.85, | |
| "learning_rate": 8.768832391713748e-06, | |
| "loss": 2.7588, | |
| "step": 369000 | |
| }, | |
| { | |
| "epoch": 19.9, | |
| "learning_rate": 5.832156308851224e-06, | |
| "loss": 2.7588, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 19.95, | |
| "learning_rate": 2.8895951035781546e-06, | |
| "loss": 2.7607, | |
| "step": 371000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.4112973955375624, | |
| "eval_loss": 3.397616147994995, | |
| "eval_runtime": 154.21, | |
| "eval_samples_per_second": 375.585, | |
| "eval_steps_per_second": 5.869, | |
| "step": 371840 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 371840, | |
| "total_flos": 1.56667295384064e+18, | |
| "train_loss": 3.0277852818395514, | |
| "train_runtime": 81849.4924, | |
| "train_samples_per_second": 145.374, | |
| "train_steps_per_second": 4.543 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 371840, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 5000, | |
| "total_flos": 1.56667295384064e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |