| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 371880, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.8108707666397095, | |
| "learning_rate": 3.125e-05, | |
| "loss": 6.2176, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.8937825560569763, | |
| "learning_rate": 6.25e-05, | |
| "loss": 5.014, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.8364607691764832, | |
| "learning_rate": 9.375e-05, | |
| "loss": 4.6797, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.80474454164505, | |
| "learning_rate": 0.000125, | |
| "loss": 4.4554, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.8272607922554016, | |
| "learning_rate": 0.00015625, | |
| "loss": 4.2966, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.7503081560134888, | |
| "learning_rate": 0.0001875, | |
| "loss": 4.1799, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.6826441884040833, | |
| "learning_rate": 0.00021875, | |
| "loss": 4.0754, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.675743579864502, | |
| "learning_rate": 0.00025, | |
| "loss": 3.9852, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.6586818099021912, | |
| "learning_rate": 0.00028121875, | |
| "loss": 3.905, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.6130994558334351, | |
| "learning_rate": 0.00031246875000000003, | |
| "loss": 3.8441, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.5862373113632202, | |
| "learning_rate": 0.00034368749999999997, | |
| "loss": 3.8003, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.5662557482719421, | |
| "learning_rate": 0.0003749375, | |
| "loss": 3.7638, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.5084989666938782, | |
| "learning_rate": 0.0004061875, | |
| "loss": 3.7227, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.4641186594963074, | |
| "learning_rate": 0.0004374375, | |
| "loss": 3.6876, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.4646807610988617, | |
| "learning_rate": 0.00046865625, | |
| "loss": 3.6618, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.4126532971858978, | |
| "learning_rate": 0.00049990625, | |
| "loss": 3.6414, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.4246925711631775, | |
| "learning_rate": 0.00053109375, | |
| "loss": 3.6218, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.38441503047943115, | |
| "learning_rate": 0.00056234375, | |
| "loss": 3.6021, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.35802428577743683, | |
| "eval_loss": 3.802968978881836, | |
| "eval_runtime": 153.3684, | |
| "eval_samples_per_second": 377.64, | |
| "eval_steps_per_second": 5.901, | |
| "step": 18594 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 0.3549180030822754, | |
| "learning_rate": 0.0005935625, | |
| "loss": 3.5728, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.32603904604911804, | |
| "learning_rate": 0.00062478125, | |
| "loss": 3.5442, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 0.31714433431625366, | |
| "learning_rate": 0.0006560312499999999, | |
| "loss": 3.5347, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 0.3043628931045532, | |
| "learning_rate": 0.00068728125, | |
| "loss": 3.524, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.29280972480773926, | |
| "learning_rate": 0.00071853125, | |
| "loss": 3.5164, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 0.306801974773407, | |
| "learning_rate": 0.0007497500000000001, | |
| "loss": 3.5009, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 0.27948838472366333, | |
| "learning_rate": 0.000781, | |
| "loss": 3.4938, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.3129396438598633, | |
| "learning_rate": 0.00081225, | |
| "loss": 3.4841, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.24496008455753326, | |
| "learning_rate": 0.00084346875, | |
| "loss": 3.4793, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 0.23840215802192688, | |
| "learning_rate": 0.00087471875, | |
| "loss": 3.4676, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.2514866292476654, | |
| "learning_rate": 0.0009059375, | |
| "loss": 3.4596, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 0.24790681898593903, | |
| "learning_rate": 0.0009371875, | |
| "loss": 3.4503, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 0.23787342011928558, | |
| "learning_rate": 0.0009684062500000001, | |
| "loss": 3.4442, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.25654569268226624, | |
| "learning_rate": 0.0009996562500000001, | |
| "loss": 3.4397, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 0.2547774016857147, | |
| "learning_rate": 0.0009970901494645169, | |
| "loss": 3.4263, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 0.2178904265165329, | |
| "learning_rate": 0.00099415087678004, | |
| "loss": 3.4174, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.2203870415687561, | |
| "learning_rate": 0.000991208661880664, | |
| "loss": 3.4045, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 0.23818820714950562, | |
| "learning_rate": 0.000988269389196187, | |
| "loss": 3.3961, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 0.21215203404426575, | |
| "learning_rate": 0.0009853271742968105, | |
| "loss": 3.3886, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3807623229030497, | |
| "eval_loss": 3.6018991470336914, | |
| "eval_runtime": 154.0265, | |
| "eval_samples_per_second": 376.026, | |
| "eval_steps_per_second": 5.876, | |
| "step": 37188 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.21038022637367249, | |
| "learning_rate": 0.0009823849593974345, | |
| "loss": 3.3355, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.2191259115934372, | |
| "learning_rate": 0.0009794456867129576, | |
| "loss": 3.3288, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.19959229230880737, | |
| "learning_rate": 0.0009765064140284806, | |
| "loss": 3.3227, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 0.2156379371881485, | |
| "learning_rate": 0.0009735641991291045, | |
| "loss": 3.3187, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 0.2161584496498108, | |
| "learning_rate": 0.0009706219842297282, | |
| "loss": 3.3161, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 0.1998431235551834, | |
| "learning_rate": 0.000967679769330352, | |
| "loss": 3.3137, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 0.19886909425258636, | |
| "learning_rate": 0.0009647375544309756, | |
| "loss": 3.3056, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 0.196481391787529, | |
| "learning_rate": 0.0009617953395315993, | |
| "loss": 3.304, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 0.1846245974302292, | |
| "learning_rate": 0.0009588590090620219, | |
| "loss": 3.2982, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 0.26932549476623535, | |
| "learning_rate": 0.0009559167941626457, | |
| "loss": 3.2934, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 0.19994695484638214, | |
| "learning_rate": 0.0009529775214781687, | |
| "loss": 3.2856, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.1982845515012741, | |
| "learning_rate": 0.0009500353065787925, | |
| "loss": 3.2853, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "grad_norm": 0.22861963510513306, | |
| "learning_rate": 0.0009470930916794163, | |
| "loss": 3.2773, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 0.21855035424232483, | |
| "learning_rate": 0.0009441538189949395, | |
| "loss": 3.2775, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.1985599547624588, | |
| "learning_rate": 0.0009412116040955631, | |
| "loss": 3.2705, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.21117864549160004, | |
| "learning_rate": 0.0009382723314110863, | |
| "loss": 3.2669, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 0.20696263015270233, | |
| "learning_rate": 0.0009353301165117101, | |
| "loss": 3.2612, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.19961431622505188, | |
| "learning_rate": 0.0009323879016123337, | |
| "loss": 3.2595, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.39251212938512425, | |
| "eval_loss": 3.45461368560791, | |
| "eval_runtime": 154.2675, | |
| "eval_samples_per_second": 375.439, | |
| "eval_steps_per_second": 5.866, | |
| "step": 55782 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 0.2054821103811264, | |
| "learning_rate": 0.0009294456867129574, | |
| "loss": 3.2412, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 0.24094636738300323, | |
| "learning_rate": 0.0009265064140284807, | |
| "loss": 3.1894, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.20816025137901306, | |
| "learning_rate": 0.0009235671413440038, | |
| "loss": 3.1928, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 0.19106166064739227, | |
| "learning_rate": 0.0009206278686595268, | |
| "loss": 3.1975, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 0.19138582050800323, | |
| "learning_rate": 0.0009176856537601506, | |
| "loss": 3.1966, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "grad_norm": 0.22713367640972137, | |
| "learning_rate": 0.0009147434388607744, | |
| "loss": 3.1969, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "grad_norm": 0.21247854828834534, | |
| "learning_rate": 0.0009118012239613982, | |
| "loss": 3.1941, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 0.21524560451507568, | |
| "learning_rate": 0.0009088590090620219, | |
| "loss": 3.198, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 0.19328753650188446, | |
| "learning_rate": 0.000905919736377545, | |
| "loss": 3.1909, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 0.20355555415153503, | |
| "learning_rate": 0.0009029775214781688, | |
| "loss": 3.193, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "grad_norm": 0.19756613671779633, | |
| "learning_rate": 0.0009000353065787925, | |
| "loss": 3.1905, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.19862300157546997, | |
| "learning_rate": 0.0008970960338943157, | |
| "loss": 3.194, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "grad_norm": 0.22642965614795685, | |
| "learning_rate": 0.0008941538189949394, | |
| "loss": 3.1927, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "grad_norm": 0.23159223794937134, | |
| "learning_rate": 0.0008912116040955631, | |
| "loss": 3.1915, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 0.19946345686912537, | |
| "learning_rate": 0.0008882693891961869, | |
| "loss": 3.1871, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "grad_norm": 0.2760438621044159, | |
| "learning_rate": 0.00088533011651171, | |
| "loss": 3.1881, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "grad_norm": 0.24984723329544067, | |
| "learning_rate": 0.0008823879016123337, | |
| "loss": 3.1819, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "grad_norm": 0.20177915692329407, | |
| "learning_rate": 0.000879448628927857, | |
| "loss": 3.1835, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "grad_norm": 0.20022615790367126, | |
| "learning_rate": 0.0008765064140284806, | |
| "loss": 3.1812, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3980135339604644, | |
| "eval_loss": 3.4454047679901123, | |
| "eval_runtime": 153.6394, | |
| "eval_samples_per_second": 376.974, | |
| "eval_steps_per_second": 5.89, | |
| "step": 74376 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 0.19454769790172577, | |
| "learning_rate": 0.0008735641991291044, | |
| "loss": 3.133, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 0.20843255519866943, | |
| "learning_rate": 0.0008706249264446276, | |
| "loss": 3.116, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "grad_norm": 0.21121448278427124, | |
| "learning_rate": 0.0008676827115452513, | |
| "loss": 3.1183, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "grad_norm": 0.2094685435295105, | |
| "learning_rate": 0.0008647434388607744, | |
| "loss": 3.118, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 0.2301424890756607, | |
| "learning_rate": 0.0008618012239613982, | |
| "loss": 3.1265, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "grad_norm": 0.24608786404132843, | |
| "learning_rate": 0.0008588590090620219, | |
| "loss": 3.1284, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "grad_norm": 0.21532559394836426, | |
| "learning_rate": 0.000855919736377545, | |
| "loss": 3.1279, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "grad_norm": 0.18981580436229706, | |
| "learning_rate": 0.0008529775214781688, | |
| "loss": 3.1293, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "grad_norm": 0.2737584412097931, | |
| "learning_rate": 0.0008500353065787926, | |
| "loss": 3.1304, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "grad_norm": 0.1946648210287094, | |
| "learning_rate": 0.0008470930916794163, | |
| "loss": 3.1229, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "grad_norm": 0.20237942039966583, | |
| "learning_rate": 0.0008441538189949394, | |
| "loss": 3.1282, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "grad_norm": 0.20490285754203796, | |
| "learning_rate": 0.0008412116040955632, | |
| "loss": 3.1285, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "grad_norm": 0.20395901799201965, | |
| "learning_rate": 0.0008382723314110862, | |
| "loss": 3.124, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "grad_norm": 0.22166697680950165, | |
| "learning_rate": 0.00083533011651171, | |
| "loss": 3.1295, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "grad_norm": 0.21369241178035736, | |
| "learning_rate": 0.0008323879016123338, | |
| "loss": 3.1255, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "grad_norm": 0.21132268011569977, | |
| "learning_rate": 0.0008294486289278568, | |
| "loss": 3.1248, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "grad_norm": 0.2022734135389328, | |
| "learning_rate": 0.0008265064140284807, | |
| "loss": 3.1268, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "grad_norm": 0.25799113512039185, | |
| "learning_rate": 0.0008235641991291044, | |
| "loss": 3.1282, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4015370314907847, | |
| "eval_loss": 3.4115054607391357, | |
| "eval_runtime": 153.7951, | |
| "eval_samples_per_second": 376.592, | |
| "eval_steps_per_second": 5.884, | |
| "step": 92970 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.21077696979045868, | |
| "learning_rate": 0.0008206219842297281, | |
| "loss": 3.1186, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "grad_norm": 0.21865037083625793, | |
| "learning_rate": 0.0008176827115452513, | |
| "loss": 3.055, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "grad_norm": 0.23599696159362793, | |
| "learning_rate": 0.0008147434388607744, | |
| "loss": 3.0637, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "grad_norm": 0.21938414871692657, | |
| "learning_rate": 0.000811801223961398, | |
| "loss": 3.0664, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "grad_norm": 0.23077841103076935, | |
| "learning_rate": 0.000808859009062022, | |
| "loss": 3.0707, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "grad_norm": 0.21069471538066864, | |
| "learning_rate": 0.000805919736377545, | |
| "loss": 3.0729, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "grad_norm": 0.23917998373508453, | |
| "learning_rate": 0.0008029775214781688, | |
| "loss": 3.0714, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "grad_norm": 0.18625333905220032, | |
| "learning_rate": 0.0008000353065787926, | |
| "loss": 3.0735, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "grad_norm": 0.2624528408050537, | |
| "learning_rate": 0.0007970960338943156, | |
| "loss": 3.0791, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "grad_norm": 0.20559582114219666, | |
| "learning_rate": 0.0007941538189949394, | |
| "loss": 3.0794, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "grad_norm": 0.20613546669483185, | |
| "learning_rate": 0.0007912116040955632, | |
| "loss": 3.0787, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "grad_norm": 0.22262121737003326, | |
| "learning_rate": 0.000788269389196187, | |
| "loss": 3.0778, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "grad_norm": 0.21528062224388123, | |
| "learning_rate": 0.00078533011651171, | |
| "loss": 3.0796, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "grad_norm": 0.20560766756534576, | |
| "learning_rate": 0.0007823879016123338, | |
| "loss": 3.0832, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "grad_norm": 0.22400015592575073, | |
| "learning_rate": 0.0007794486289278569, | |
| "loss": 3.0857, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "grad_norm": 0.22037924826145172, | |
| "learning_rate": 0.0007765064140284806, | |
| "loss": 3.0776, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "grad_norm": 0.20716436207294464, | |
| "learning_rate": 0.0007735671413440037, | |
| "loss": 3.0792, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "grad_norm": 0.2654527723789215, | |
| "learning_rate": 0.0007706249264446275, | |
| "loss": 3.0851, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "grad_norm": 0.20225553214550018, | |
| "learning_rate": 0.0007676856537601507, | |
| "loss": 3.0804, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.40455787284397654, | |
| "eval_loss": 3.3764584064483643, | |
| "eval_runtime": 154.2451, | |
| "eval_samples_per_second": 375.493, | |
| "eval_steps_per_second": 5.867, | |
| "step": 111564 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 0.2055094987154007, | |
| "learning_rate": 0.0007647434388607743, | |
| "loss": 3.0519, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "grad_norm": 0.2185896337032318, | |
| "learning_rate": 0.0007618041661762976, | |
| "loss": 3.0174, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "grad_norm": 0.20856788754463196, | |
| "learning_rate": 0.0007588619512769213, | |
| "loss": 3.0198, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "grad_norm": 0.22500406205654144, | |
| "learning_rate": 0.0007559197363775451, | |
| "loss": 3.0234, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.21073341369628906, | |
| "learning_rate": 0.0007529775214781687, | |
| "loss": 3.0292, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "grad_norm": 0.22741083800792694, | |
| "learning_rate": 0.0007500353065787924, | |
| "loss": 3.0311, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "grad_norm": 0.23282620310783386, | |
| "learning_rate": 0.0007470930916794163, | |
| "loss": 3.036, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 0.2192760556936264, | |
| "learning_rate": 0.0007441538189949394, | |
| "loss": 3.0341, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "grad_norm": 0.2083265483379364, | |
| "learning_rate": 0.0007412116040955633, | |
| "loss": 3.0376, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "grad_norm": 0.2194899469614029, | |
| "learning_rate": 0.0007382723314110863, | |
| "loss": 3.0377, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "grad_norm": 0.22579005360603333, | |
| "learning_rate": 0.00073533011651171, | |
| "loss": 3.0399, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "grad_norm": 0.21945177018642426, | |
| "learning_rate": 0.0007323879016123339, | |
| "loss": 3.038, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "grad_norm": 0.23311348259449005, | |
| "learning_rate": 0.0007294515711427562, | |
| "loss": 3.0418, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.20455820858478546, | |
| "learning_rate": 0.00072650935624338, | |
| "loss": 3.0488, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "grad_norm": 0.21514442563056946, | |
| "learning_rate": 0.0007235671413440038, | |
| "loss": 3.0442, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "grad_norm": 0.256694495677948, | |
| "learning_rate": 0.0007206249264446275, | |
| "loss": 3.0428, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "grad_norm": 0.2359844148159027, | |
| "learning_rate": 0.0007176827115452513, | |
| "loss": 3.0464, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "grad_norm": 0.22172461450099945, | |
| "learning_rate": 0.0007147434388607744, | |
| "loss": 3.0473, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "grad_norm": 0.32295501232147217, | |
| "learning_rate": 0.0007118012239613981, | |
| "loss": 3.0457, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.4062039005098178, | |
| "eval_loss": 3.3680343627929688, | |
| "eval_runtime": 153.628, | |
| "eval_samples_per_second": 377.002, | |
| "eval_steps_per_second": 5.891, | |
| "step": 130158 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "grad_norm": 0.22584497928619385, | |
| "learning_rate": 0.0007088590090620219, | |
| "loss": 2.9831, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "grad_norm": 0.20827385783195496, | |
| "learning_rate": 0.000705919736377545, | |
| "loss": 2.9797, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "grad_norm": 0.2678986191749573, | |
| "learning_rate": 0.0007029775214781687, | |
| "loss": 2.9887, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "grad_norm": 0.25479185581207275, | |
| "learning_rate": 0.0007000353065787925, | |
| "loss": 2.9954, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "grad_norm": 0.23116907477378845, | |
| "learning_rate": 0.0006970930916794163, | |
| "loss": 2.9968, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "grad_norm": 0.25440552830696106, | |
| "learning_rate": 0.0006941508767800401, | |
| "loss": 2.9965, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "grad_norm": 0.23387254774570465, | |
| "learning_rate": 0.0006912116040955631, | |
| "loss": 3.0031, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "grad_norm": 0.23628811538219452, | |
| "learning_rate": 0.0006882693891961869, | |
| "loss": 3.0035, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "grad_norm": 0.21515734493732452, | |
| "learning_rate": 0.0006853301165117101, | |
| "loss": 3.0011, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "grad_norm": 0.28068262338638306, | |
| "learning_rate": 0.0006823879016123337, | |
| "loss": 3.0051, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "grad_norm": 0.2298477292060852, | |
| "learning_rate": 0.0006794486289278569, | |
| "loss": 3.0091, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "grad_norm": 0.23959632217884064, | |
| "learning_rate": 0.0006765064140284807, | |
| "loss": 3.0079, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "grad_norm": 0.21614809334278107, | |
| "learning_rate": 0.0006735700835589031, | |
| "loss": 3.0101, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "grad_norm": 0.3193395435810089, | |
| "learning_rate": 0.0006706278686595268, | |
| "loss": 3.0085, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "grad_norm": 0.2261291742324829, | |
| "learning_rate": 0.0006676856537601506, | |
| "loss": 3.0156, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "grad_norm": 0.31309646368026733, | |
| "learning_rate": 0.0006647463810756738, | |
| "loss": 3.0141, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "grad_norm": 0.2506345808506012, | |
| "learning_rate": 0.0006618041661762975, | |
| "loss": 3.0153, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "grad_norm": 0.24304701387882233, | |
| "learning_rate": 0.0006588648934918207, | |
| "loss": 3.0127, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4069256373864405, | |
| "eval_loss": 3.3718349933624268, | |
| "eval_runtime": 153.6896, | |
| "eval_samples_per_second": 376.851, | |
| "eval_steps_per_second": 5.888, | |
| "step": 148752 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "grad_norm": 0.2484602928161621, | |
| "learning_rate": 0.0006559226785924444, | |
| "loss": 2.9984, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 0.31583237648010254, | |
| "learning_rate": 0.0006529804636930682, | |
| "loss": 2.9455, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "grad_norm": 0.24394702911376953, | |
| "learning_rate": 0.0006500382487936918, | |
| "loss": 2.9544, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "grad_norm": 0.29708966612815857, | |
| "learning_rate": 0.000647098976109215, | |
| "loss": 2.9598, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "grad_norm": 0.23569166660308838, | |
| "learning_rate": 0.0006441567612098388, | |
| "loss": 2.9661, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "grad_norm": 0.26388081908226013, | |
| "learning_rate": 0.0006412145463104624, | |
| "loss": 2.9691, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "grad_norm": 0.22147905826568604, | |
| "learning_rate": 0.0006382752736259857, | |
| "loss": 2.9663, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "grad_norm": 0.24741463363170624, | |
| "learning_rate": 0.0006353330587266094, | |
| "loss": 2.9733, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "grad_norm": 0.25748059153556824, | |
| "learning_rate": 0.0006323937860421325, | |
| "loss": 2.9751, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 0.2481698840856552, | |
| "learning_rate": 0.0006294515711427563, | |
| "loss": 2.9752, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "grad_norm": 0.23890063166618347, | |
| "learning_rate": 0.00062650935624338, | |
| "loss": 2.9759, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "grad_norm": 0.2355976551771164, | |
| "learning_rate": 0.0006235700835589031, | |
| "loss": 2.9747, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "grad_norm": 0.25961464643478394, | |
| "learning_rate": 0.0006206278686595269, | |
| "loss": 2.9835, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "grad_norm": 0.2471887469291687, | |
| "learning_rate": 0.00061768859597505, | |
| "loss": 2.9821, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "grad_norm": 0.2203267514705658, | |
| "learning_rate": 0.0006147463810756738, | |
| "loss": 2.984, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "grad_norm": 0.23866264522075653, | |
| "learning_rate": 0.0006118041661762975, | |
| "loss": 2.9875, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "grad_norm": 0.22918641567230225, | |
| "learning_rate": 0.0006088648934918207, | |
| "loss": 2.984, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "grad_norm": 0.23822970688343048, | |
| "learning_rate": 0.0006059226785924444, | |
| "loss": 2.9854, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "grad_norm": 0.22895199060440063, | |
| "learning_rate": 0.0006029804636930681, | |
| "loss": 2.9897, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.4080178161450047, | |
| "eval_loss": 3.3722941875457764, | |
| "eval_runtime": 153.9821, | |
| "eval_samples_per_second": 376.135, | |
| "eval_steps_per_second": 5.877, | |
| "step": 167346 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "grad_norm": 0.245958149433136, | |
| "learning_rate": 0.0006000411910085913, | |
| "loss": 2.9388, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "grad_norm": 0.22162730991840363, | |
| "learning_rate": 0.0005970989761092151, | |
| "loss": 2.9246, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "grad_norm": 0.37472179532051086, | |
| "learning_rate": 0.0005941597034247381, | |
| "loss": 2.9315, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "grad_norm": 0.22953909635543823, | |
| "learning_rate": 0.000591217488525362, | |
| "loss": 2.9302, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 0.2361230105161667, | |
| "learning_rate": 0.0005882752736259857, | |
| "loss": 2.941, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "grad_norm": 0.25345179438591003, | |
| "learning_rate": 0.0005853330587266094, | |
| "loss": 2.9453, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "grad_norm": 0.23210594058036804, | |
| "learning_rate": 0.0005823937860421326, | |
| "loss": 2.9429, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "grad_norm": 0.22280991077423096, | |
| "learning_rate": 0.0005794515711427563, | |
| "loss": 2.9464, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "grad_norm": 0.24767878651618958, | |
| "learning_rate": 0.0005765122984582793, | |
| "loss": 2.9467, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "grad_norm": 0.2509325444698334, | |
| "learning_rate": 0.0005735700835589032, | |
| "loss": 2.9484, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "grad_norm": 0.22943289577960968, | |
| "learning_rate": 0.0005706308108744262, | |
| "loss": 2.9523, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "grad_norm": 0.2891969084739685, | |
| "learning_rate": 0.00056768859597505, | |
| "loss": 2.9557, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "grad_norm": 0.242570161819458, | |
| "learning_rate": 0.0005647493232905732, | |
| "loss": 2.9549, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "grad_norm": 0.26979508996009827, | |
| "learning_rate": 0.0005618071083911968, | |
| "loss": 2.9595, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "grad_norm": 0.24241873621940613, | |
| "learning_rate": 0.0005588678357067201, | |
| "loss": 2.958, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "grad_norm": 0.24058844149112701, | |
| "learning_rate": 0.0005559256208073438, | |
| "loss": 2.9619, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "grad_norm": 0.23146815598011017, | |
| "learning_rate": 0.0005529834059079676, | |
| "loss": 2.9608, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "grad_norm": 0.22514961659908295, | |
| "learning_rate": 0.0005500441332234907, | |
| "loss": 2.9602, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4082696816900534, | |
| "eval_loss": 3.3669023513793945, | |
| "eval_runtime": 154.1716, | |
| "eval_samples_per_second": 375.672, | |
| "eval_steps_per_second": 5.87, | |
| "step": 185940 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.29674816131591797, | |
| "learning_rate": 0.0005471019183241144, | |
| "loss": 2.9548, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "grad_norm": 0.2413276582956314, | |
| "learning_rate": 0.0005441597034247382, | |
| "loss": 2.8963, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "grad_norm": 0.2437901645898819, | |
| "learning_rate": 0.0005412174885253619, | |
| "loss": 2.9023, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 10.16, | |
| "grad_norm": 0.27865538001060486, | |
| "learning_rate": 0.0005382811580557843, | |
| "loss": 2.9052, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "grad_norm": 0.26435813307762146, | |
| "learning_rate": 0.0005353389431564082, | |
| "loss": 2.9114, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "grad_norm": 0.3438023328781128, | |
| "learning_rate": 0.0005323996704719313, | |
| "loss": 2.915, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 10.33, | |
| "grad_norm": 0.2424338161945343, | |
| "learning_rate": 0.0005294574555725549, | |
| "loss": 2.9198, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "grad_norm": 0.2525182366371155, | |
| "learning_rate": 0.0005265152406731788, | |
| "loss": 2.9189, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "grad_norm": 0.23211458325386047, | |
| "learning_rate": 0.0005235730257738025, | |
| "loss": 2.9227, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "grad_norm": 0.2459993213415146, | |
| "learning_rate": 0.0005206308108744263, | |
| "loss": 2.9274, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "grad_norm": 0.26037323474884033, | |
| "learning_rate": 0.0005176915381899494, | |
| "loss": 2.9266, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "grad_norm": 0.23404568433761597, | |
| "learning_rate": 0.0005147522655054725, | |
| "loss": 2.9263, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "grad_norm": 0.2657169997692108, | |
| "learning_rate": 0.0005118100506060963, | |
| "loss": 2.9336, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 10.7, | |
| "grad_norm": 0.2722354233264923, | |
| "learning_rate": 0.0005088678357067201, | |
| "loss": 2.9352, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "grad_norm": 0.2347659170627594, | |
| "learning_rate": 0.0005059256208073439, | |
| "loss": 2.9349, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "grad_norm": 0.2561778724193573, | |
| "learning_rate": 0.0005029892903377663, | |
| "loss": 2.9333, | |
| "step": 201000 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "grad_norm": 0.24672134220600128, | |
| "learning_rate": 0.00050004707543839, | |
| "loss": 2.9346, | |
| "step": 202000 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "grad_norm": 0.24908240139484406, | |
| "learning_rate": 0.0004971048605390138, | |
| "loss": 2.9392, | |
| "step": 203000 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "grad_norm": 0.24685357511043549, | |
| "learning_rate": 0.0004941655878545369, | |
| "loss": 2.9377, | |
| "step": 204000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.4095683109207261, | |
| "eval_loss": 3.3622021675109863, | |
| "eval_runtime": 153.8148, | |
| "eval_samples_per_second": 376.544, | |
| "eval_steps_per_second": 5.884, | |
| "step": 204534 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "grad_norm": 0.24828964471817017, | |
| "learning_rate": 0.0004912233729551607, | |
| "loss": 2.9099, | |
| "step": 205000 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "grad_norm": 0.2552799880504608, | |
| "learning_rate": 0.00048828115805578436, | |
| "loss": 2.877, | |
| "step": 206000 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "grad_norm": 0.26694855093955994, | |
| "learning_rate": 0.00048533894315640815, | |
| "loss": 2.8848, | |
| "step": 207000 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "grad_norm": 0.2933233678340912, | |
| "learning_rate": 0.0004823996704719313, | |
| "loss": 2.8884, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "grad_norm": 0.3140556812286377, | |
| "learning_rate": 0.000479457455572555, | |
| "loss": 2.8935, | |
| "step": 209000 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "grad_norm": 0.2629554867744446, | |
| "learning_rate": 0.00047651818288807816, | |
| "loss": 2.8933, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "grad_norm": 0.2608068585395813, | |
| "learning_rate": 0.0004735759679887019, | |
| "loss": 2.8964, | |
| "step": 211000 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "grad_norm": 0.28370392322540283, | |
| "learning_rate": 0.00047063375308932563, | |
| "loss": 2.8947, | |
| "step": 212000 | |
| }, | |
| { | |
| "epoch": 11.46, | |
| "grad_norm": 0.242896169424057, | |
| "learning_rate": 0.0004676915381899494, | |
| "loss": 2.9034, | |
| "step": 213000 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "grad_norm": 0.2785256505012512, | |
| "learning_rate": 0.0004647522655054725, | |
| "loss": 2.9076, | |
| "step": 214000 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "grad_norm": 0.3482002019882202, | |
| "learning_rate": 0.00046181005060609624, | |
| "loss": 2.9073, | |
| "step": 215000 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "grad_norm": 0.23567426204681396, | |
| "learning_rate": 0.00045887077792161943, | |
| "loss": 2.9085, | |
| "step": 216000 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "grad_norm": 0.261733740568161, | |
| "learning_rate": 0.00045592856302224317, | |
| "loss": 2.9073, | |
| "step": 217000 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "grad_norm": 0.26738300919532776, | |
| "learning_rate": 0.00045298929033776625, | |
| "loss": 2.9069, | |
| "step": 218000 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "grad_norm": 0.23981693387031555, | |
| "learning_rate": 0.00045004707543839004, | |
| "loss": 2.9113, | |
| "step": 219000 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "grad_norm": 0.3659937381744385, | |
| "learning_rate": 0.0004471048605390138, | |
| "loss": 2.9166, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "grad_norm": 0.24524465203285217, | |
| "learning_rate": 0.0004441626456396375, | |
| "loss": 2.9113, | |
| "step": 221000 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "grad_norm": 0.24333979189395905, | |
| "learning_rate": 0.00044122337295516065, | |
| "loss": 2.9147, | |
| "step": 222000 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "grad_norm": 0.3160707652568817, | |
| "learning_rate": 0.0004382811580557844, | |
| "loss": 2.9136, | |
| "step": 223000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4097532631334546, | |
| "eval_loss": 3.37412428855896, | |
| "eval_runtime": 153.7573, | |
| "eval_samples_per_second": 376.685, | |
| "eval_steps_per_second": 5.886, | |
| "step": 223128 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "grad_norm": 0.2786504924297333, | |
| "learning_rate": 0.0004353418853713075, | |
| "loss": 2.8632, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "grad_norm": 0.2501066327095032, | |
| "learning_rate": 0.00043240261268683066, | |
| "loss": 2.8621, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "grad_norm": 0.28876030445098877, | |
| "learning_rate": 0.0004294603977874544, | |
| "loss": 2.8624, | |
| "step": 226000 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "grad_norm": 0.26900097727775574, | |
| "learning_rate": 0.00042651818288807813, | |
| "loss": 2.8696, | |
| "step": 227000 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "grad_norm": 0.32585257291793823, | |
| "learning_rate": 0.0004235759679887019, | |
| "loss": 2.869, | |
| "step": 228000 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "grad_norm": 0.2984183728694916, | |
| "learning_rate": 0.00042063669530422506, | |
| "loss": 2.8723, | |
| "step": 229000 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "grad_norm": 0.25952082872390747, | |
| "learning_rate": 0.00041769742261974814, | |
| "loss": 2.8813, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 12.42, | |
| "grad_norm": 0.272550493478775, | |
| "learning_rate": 0.0004147552077203719, | |
| "loss": 2.8817, | |
| "step": 231000 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "grad_norm": 0.3032265305519104, | |
| "learning_rate": 0.00041181299282099567, | |
| "loss": 2.8825, | |
| "step": 232000 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "grad_norm": 0.26728367805480957, | |
| "learning_rate": 0.0004088707779216194, | |
| "loss": 2.8847, | |
| "step": 233000 | |
| }, | |
| { | |
| "epoch": 12.58, | |
| "grad_norm": 0.2473539412021637, | |
| "learning_rate": 0.0004059315052371425, | |
| "loss": 2.8873, | |
| "step": 234000 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "grad_norm": 0.2527707517147064, | |
| "learning_rate": 0.0004029892903377663, | |
| "loss": 2.888, | |
| "step": 235000 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "grad_norm": 0.25239551067352295, | |
| "learning_rate": 0.0004000500176532894, | |
| "loss": 2.8873, | |
| "step": 236000 | |
| }, | |
| { | |
| "epoch": 12.75, | |
| "grad_norm": 0.2829276919364929, | |
| "learning_rate": 0.00039711074496881255, | |
| "loss": 2.8898, | |
| "step": 237000 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "grad_norm": 0.279778391122818, | |
| "learning_rate": 0.0003941685300694363, | |
| "loss": 2.8884, | |
| "step": 238000 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "grad_norm": 0.259166419506073, | |
| "learning_rate": 0.00039122631517006, | |
| "loss": 2.8938, | |
| "step": 239000 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "grad_norm": 0.25353771448135376, | |
| "learning_rate": 0.00038828410027068376, | |
| "loss": 2.8903, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "grad_norm": 0.28928855061531067, | |
| "learning_rate": 0.00038534188537130755, | |
| "loss": 2.8939, | |
| "step": 241000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.4091496995013613, | |
| "eval_loss": 3.382020950317383, | |
| "eval_runtime": 154.4792, | |
| "eval_samples_per_second": 374.924, | |
| "eval_steps_per_second": 5.858, | |
| "step": 241722 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "grad_norm": 0.2669352889060974, | |
| "learning_rate": 0.00038240261268683063, | |
| "loss": 2.8786, | |
| "step": 242000 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "grad_norm": 0.34619081020355225, | |
| "learning_rate": 0.00037946334000235377, | |
| "loss": 2.8411, | |
| "step": 243000 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "grad_norm": 0.2849779725074768, | |
| "learning_rate": 0.00037652112510297756, | |
| "loss": 2.8399, | |
| "step": 244000 | |
| }, | |
| { | |
| "epoch": 13.18, | |
| "grad_norm": 0.2858317494392395, | |
| "learning_rate": 0.0003735818524185007, | |
| "loss": 2.8474, | |
| "step": 245000 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "grad_norm": 0.3325158953666687, | |
| "learning_rate": 0.0003706396375191244, | |
| "loss": 2.8495, | |
| "step": 246000 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "grad_norm": 0.2560212016105652, | |
| "learning_rate": 0.00036769742261974817, | |
| "loss": 2.8569, | |
| "step": 247000 | |
| }, | |
| { | |
| "epoch": 13.34, | |
| "grad_norm": 0.2676179111003876, | |
| "learning_rate": 0.0003647552077203719, | |
| "loss": 2.8553, | |
| "step": 248000 | |
| }, | |
| { | |
| "epoch": 13.39, | |
| "grad_norm": 0.2544156610965729, | |
| "learning_rate": 0.00036181593503589504, | |
| "loss": 2.8598, | |
| "step": 249000 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "grad_norm": 0.25966116786003113, | |
| "learning_rate": 0.0003588737201365188, | |
| "loss": 2.8591, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "grad_norm": 0.30520108342170715, | |
| "learning_rate": 0.0003559315052371425, | |
| "loss": 2.8619, | |
| "step": 251000 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "grad_norm": 0.24815978109836578, | |
| "learning_rate": 0.00035298929033776626, | |
| "loss": 2.8629, | |
| "step": 252000 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "grad_norm": 0.32576534152030945, | |
| "learning_rate": 0.00035005001765328945, | |
| "loss": 2.8695, | |
| "step": 253000 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "grad_norm": 0.28603044152259827, | |
| "learning_rate": 0.00034710780275391313, | |
| "loss": 2.8704, | |
| "step": 254000 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "grad_norm": 0.25637853145599365, | |
| "learning_rate": 0.00034416558785453687, | |
| "loss": 2.8683, | |
| "step": 255000 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "grad_norm": 0.2801949977874756, | |
| "learning_rate": 0.00034122631517006006, | |
| "loss": 2.8667, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "grad_norm": 0.2888367176055908, | |
| "learning_rate": 0.0003382870424855832, | |
| "loss": 2.8714, | |
| "step": 257000 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "grad_norm": 0.25710824131965637, | |
| "learning_rate": 0.0003353448275862069, | |
| "loss": 2.8685, | |
| "step": 258000 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "grad_norm": 0.2829296886920929, | |
| "learning_rate": 0.00033240261268683067, | |
| "loss": 2.8743, | |
| "step": 259000 | |
| }, | |
| { | |
| "epoch": 13.98, | |
| "grad_norm": 0.30661359429359436, | |
| "learning_rate": 0.0003294603977874544, | |
| "loss": 2.8736, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.40986666645168407, | |
| "eval_loss": 3.3767247200012207, | |
| "eval_runtime": 153.89, | |
| "eval_samples_per_second": 376.36, | |
| "eval_steps_per_second": 5.881, | |
| "step": 260316 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "grad_norm": 0.2881269156932831, | |
| "learning_rate": 0.00032651818288807814, | |
| "loss": 2.8388, | |
| "step": 261000 | |
| }, | |
| { | |
| "epoch": 14.09, | |
| "grad_norm": 0.28384995460510254, | |
| "learning_rate": 0.0003235789102036013, | |
| "loss": 2.8223, | |
| "step": 262000 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "grad_norm": 0.2660735249519348, | |
| "learning_rate": 0.000320636695304225, | |
| "loss": 2.8256, | |
| "step": 263000 | |
| }, | |
| { | |
| "epoch": 14.2, | |
| "grad_norm": 0.25953084230422974, | |
| "learning_rate": 0.00031769742261974815, | |
| "loss": 2.8299, | |
| "step": 264000 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "grad_norm": 0.29592570662498474, | |
| "learning_rate": 0.0003147581499352713, | |
| "loss": 2.8349, | |
| "step": 265000 | |
| }, | |
| { | |
| "epoch": 14.31, | |
| "grad_norm": 0.3023082911968231, | |
| "learning_rate": 0.000311815935035895, | |
| "loss": 2.8355, | |
| "step": 266000 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "grad_norm": 0.2720430791378021, | |
| "learning_rate": 0.00030887372013651876, | |
| "loss": 2.8371, | |
| "step": 267000 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "grad_norm": 0.33669525384902954, | |
| "learning_rate": 0.00030593150523714255, | |
| "loss": 2.8416, | |
| "step": 268000 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "grad_norm": 0.27477729320526123, | |
| "learning_rate": 0.0003029922325526657, | |
| "loss": 2.8407, | |
| "step": 269000 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "grad_norm": 0.2937850058078766, | |
| "learning_rate": 0.00030005001765328937, | |
| "loss": 2.8447, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 14.57, | |
| "grad_norm": 0.2728506028652191, | |
| "learning_rate": 0.00029710780275391316, | |
| "loss": 2.8477, | |
| "step": 271000 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "grad_norm": 0.29167652130126953, | |
| "learning_rate": 0.0002941685300694363, | |
| "loss": 2.8472, | |
| "step": 272000 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "grad_norm": 0.2727662920951843, | |
| "learning_rate": 0.00029122631517006003, | |
| "loss": 2.8476, | |
| "step": 273000 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "grad_norm": 0.28426212072372437, | |
| "learning_rate": 0.0002882841002706838, | |
| "loss": 2.8478, | |
| "step": 274000 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "grad_norm": 0.28404590487480164, | |
| "learning_rate": 0.0002853418853713075, | |
| "loss": 2.8519, | |
| "step": 275000 | |
| }, | |
| { | |
| "epoch": 14.84, | |
| "grad_norm": 0.3085533082485199, | |
| "learning_rate": 0.00028240261268683064, | |
| "loss": 2.8547, | |
| "step": 276000 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "grad_norm": 0.27339011430740356, | |
| "learning_rate": 0.00027946039778745444, | |
| "loss": 2.8573, | |
| "step": 277000 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "grad_norm": 0.2646634876728058, | |
| "learning_rate": 0.0002765211251029775, | |
| "loss": 2.8493, | |
| "step": 278000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.4100954213678993, | |
| "eval_loss": 3.3848283290863037, | |
| "eval_runtime": 154.01, | |
| "eval_samples_per_second": 376.067, | |
| "eval_steps_per_second": 5.876, | |
| "step": 278910 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.2736656665802002, | |
| "learning_rate": 0.00027357891020360125, | |
| "loss": 2.8501, | |
| "step": 279000 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "grad_norm": 0.3071829378604889, | |
| "learning_rate": 0.0002706396375191244, | |
| "loss": 2.8049, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "grad_norm": 0.26746442914009094, | |
| "learning_rate": 0.0002676974226197482, | |
| "loss": 2.8131, | |
| "step": 281000 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "grad_norm": 0.2784716486930847, | |
| "learning_rate": 0.0002647552077203719, | |
| "loss": 2.8127, | |
| "step": 282000 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "grad_norm": 0.2832111120223999, | |
| "learning_rate": 0.00026181299282099566, | |
| "loss": 2.8129, | |
| "step": 283000 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "grad_norm": 0.26944035291671753, | |
| "learning_rate": 0.0002588737201365188, | |
| "loss": 2.8152, | |
| "step": 284000 | |
| }, | |
| { | |
| "epoch": 15.33, | |
| "grad_norm": 0.2603321075439453, | |
| "learning_rate": 0.00025593150523714253, | |
| "loss": 2.8193, | |
| "step": 285000 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "grad_norm": 0.35142990946769714, | |
| "learning_rate": 0.0002529892903377663, | |
| "loss": 2.8178, | |
| "step": 286000 | |
| }, | |
| { | |
| "epoch": 15.44, | |
| "grad_norm": 0.2828776240348816, | |
| "learning_rate": 0.0002500500176532894, | |
| "loss": 2.8211, | |
| "step": 287000 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "grad_norm": 0.2981540560722351, | |
| "learning_rate": 0.00024711074496881253, | |
| "loss": 2.826, | |
| "step": 288000 | |
| }, | |
| { | |
| "epoch": 15.54, | |
| "grad_norm": 0.256558358669281, | |
| "learning_rate": 0.00024416853006943627, | |
| "loss": 2.8269, | |
| "step": 289000 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "grad_norm": 0.28373146057128906, | |
| "learning_rate": 0.0002412292573849594, | |
| "loss": 2.8261, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "grad_norm": 0.26360180974006653, | |
| "learning_rate": 0.00023828704248558314, | |
| "loss": 2.8315, | |
| "step": 291000 | |
| }, | |
| { | |
| "epoch": 15.7, | |
| "grad_norm": 0.3487566113471985, | |
| "learning_rate": 0.0002353448275862069, | |
| "loss": 2.8313, | |
| "step": 292000 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "grad_norm": 0.35876941680908203, | |
| "learning_rate": 0.00023240261268683067, | |
| "loss": 2.8315, | |
| "step": 293000 | |
| }, | |
| { | |
| "epoch": 15.81, | |
| "grad_norm": 0.3193157911300659, | |
| "learning_rate": 0.00022946334000235378, | |
| "loss": 2.835, | |
| "step": 294000 | |
| }, | |
| { | |
| "epoch": 15.87, | |
| "grad_norm": 0.2774643301963806, | |
| "learning_rate": 0.00022652112510297755, | |
| "loss": 2.836, | |
| "step": 295000 | |
| }, | |
| { | |
| "epoch": 15.92, | |
| "grad_norm": 0.25371402502059937, | |
| "learning_rate": 0.00022357891020360126, | |
| "loss": 2.8341, | |
| "step": 296000 | |
| }, | |
| { | |
| "epoch": 15.97, | |
| "grad_norm": 0.3146109879016876, | |
| "learning_rate": 0.00022063669530422502, | |
| "loss": 2.831, | |
| "step": 297000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.40995870587465466, | |
| "eval_loss": 3.395188331604004, | |
| "eval_runtime": 153.8316, | |
| "eval_samples_per_second": 376.503, | |
| "eval_steps_per_second": 5.883, | |
| "step": 297504 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "grad_norm": 0.320112407207489, | |
| "learning_rate": 0.00021769742261974816, | |
| "loss": 2.8136, | |
| "step": 298000 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "grad_norm": 0.3046302795410156, | |
| "learning_rate": 0.00021475520772037192, | |
| "loss": 2.7897, | |
| "step": 299000 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "grad_norm": 0.2765304446220398, | |
| "learning_rate": 0.00021181299282099563, | |
| "loss": 2.7962, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "grad_norm": 0.29775017499923706, | |
| "learning_rate": 0.0002088737201365188, | |
| "loss": 2.7979, | |
| "step": 301000 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "grad_norm": 0.2856539189815521, | |
| "learning_rate": 0.00020593150523714253, | |
| "loss": 2.8003, | |
| "step": 302000 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "grad_norm": 0.27068325877189636, | |
| "learning_rate": 0.00020299223255266567, | |
| "loss": 2.7995, | |
| "step": 303000 | |
| }, | |
| { | |
| "epoch": 16.35, | |
| "grad_norm": 0.28066492080688477, | |
| "learning_rate": 0.0002000500176532894, | |
| "loss": 2.8054, | |
| "step": 304000 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "grad_norm": 0.3302575349807739, | |
| "learning_rate": 0.00019710780275391317, | |
| "loss": 2.8034, | |
| "step": 305000 | |
| }, | |
| { | |
| "epoch": 16.46, | |
| "grad_norm": 0.2813708782196045, | |
| "learning_rate": 0.00019416558785453688, | |
| "loss": 2.809, | |
| "step": 306000 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "grad_norm": 0.32888713479042053, | |
| "learning_rate": 0.00019122631517006004, | |
| "loss": 2.8106, | |
| "step": 307000 | |
| }, | |
| { | |
| "epoch": 16.56, | |
| "grad_norm": 0.27354925870895386, | |
| "learning_rate": 0.00018828410027068378, | |
| "loss": 2.8067, | |
| "step": 308000 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "grad_norm": 0.2700726389884949, | |
| "learning_rate": 0.0001853448275862069, | |
| "loss": 2.811, | |
| "step": 309000 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "grad_norm": 0.27493104338645935, | |
| "learning_rate": 0.00018240555490173002, | |
| "loss": 2.8089, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 16.73, | |
| "grad_norm": 0.29335400462150574, | |
| "learning_rate": 0.00017946334000235379, | |
| "loss": 2.8103, | |
| "step": 311000 | |
| }, | |
| { | |
| "epoch": 16.78, | |
| "grad_norm": 0.27957841753959656, | |
| "learning_rate": 0.00017652112510297752, | |
| "loss": 2.8143, | |
| "step": 312000 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "grad_norm": 0.27593499422073364, | |
| "learning_rate": 0.0001735789102036013, | |
| "loss": 2.8125, | |
| "step": 313000 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "grad_norm": 0.29090675711631775, | |
| "learning_rate": 0.00017063669530422503, | |
| "loss": 2.8127, | |
| "step": 314000 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "grad_norm": 0.2894745171070099, | |
| "learning_rate": 0.00016769448040484877, | |
| "loss": 2.8173, | |
| "step": 315000 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "grad_norm": 0.3210045397281647, | |
| "learning_rate": 0.0001647552077203719, | |
| "loss": 2.8158, | |
| "step": 316000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.41055447638772274, | |
| "eval_loss": 3.401242733001709, | |
| "eval_runtime": 153.8825, | |
| "eval_samples_per_second": 376.378, | |
| "eval_steps_per_second": 5.881, | |
| "step": 316098 | |
| }, | |
| { | |
| "epoch": 17.05, | |
| "grad_norm": 0.32659468054771423, | |
| "learning_rate": 0.00016181299282099566, | |
| "loss": 2.7802, | |
| "step": 317000 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "grad_norm": 0.2894715964794159, | |
| "learning_rate": 0.00015887077792161938, | |
| "loss": 2.7795, | |
| "step": 318000 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "grad_norm": 0.29618579149246216, | |
| "learning_rate": 0.00015593150523714254, | |
| "loss": 2.7792, | |
| "step": 319000 | |
| }, | |
| { | |
| "epoch": 17.21, | |
| "grad_norm": 0.32222601771354675, | |
| "learning_rate": 0.00015298929033776628, | |
| "loss": 2.7825, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 17.26, | |
| "grad_norm": 0.2724851667881012, | |
| "learning_rate": 0.0001500500176532894, | |
| "loss": 2.7885, | |
| "step": 321000 | |
| }, | |
| { | |
| "epoch": 17.32, | |
| "grad_norm": 0.3210499584674835, | |
| "learning_rate": 0.00014710780275391315, | |
| "loss": 2.7837, | |
| "step": 322000 | |
| }, | |
| { | |
| "epoch": 17.37, | |
| "grad_norm": 0.27571025490760803, | |
| "learning_rate": 0.0001441655878545369, | |
| "loss": 2.7831, | |
| "step": 323000 | |
| }, | |
| { | |
| "epoch": 17.42, | |
| "grad_norm": 0.30725350975990295, | |
| "learning_rate": 0.00014122337295516065, | |
| "loss": 2.7861, | |
| "step": 324000 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "grad_norm": 0.3012644052505493, | |
| "learning_rate": 0.00013828410027068378, | |
| "loss": 2.7897, | |
| "step": 325000 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "grad_norm": 0.30297723412513733, | |
| "learning_rate": 0.00013534188537130752, | |
| "loss": 2.7938, | |
| "step": 326000 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "grad_norm": 0.30885782837867737, | |
| "learning_rate": 0.00013240261268683066, | |
| "loss": 2.7955, | |
| "step": 327000 | |
| }, | |
| { | |
| "epoch": 17.64, | |
| "grad_norm": 0.294734925031662, | |
| "learning_rate": 0.0001294603977874544, | |
| "loss": 2.7934, | |
| "step": 328000 | |
| }, | |
| { | |
| "epoch": 17.69, | |
| "grad_norm": 0.28373879194259644, | |
| "learning_rate": 0.00012651818288807816, | |
| "loss": 2.7934, | |
| "step": 329000 | |
| }, | |
| { | |
| "epoch": 17.75, | |
| "grad_norm": 0.2635454535484314, | |
| "learning_rate": 0.00012357891020360127, | |
| "loss": 2.794, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 17.8, | |
| "grad_norm": 0.29980969429016113, | |
| "learning_rate": 0.00012063669530422502, | |
| "loss": 2.7911, | |
| "step": 331000 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "grad_norm": 0.2912716567516327, | |
| "learning_rate": 0.00011769742261974815, | |
| "loss": 2.7942, | |
| "step": 332000 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "grad_norm": 0.2887323498725891, | |
| "learning_rate": 0.0001147552077203719, | |
| "loss": 2.7984, | |
| "step": 333000 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "grad_norm": 0.28569164872169495, | |
| "learning_rate": 0.00011181299282099564, | |
| "loss": 2.8018, | |
| "step": 334000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.4104207840872034, | |
| "eval_loss": 3.40279221534729, | |
| "eval_runtime": 155.722, | |
| "eval_samples_per_second": 371.932, | |
| "eval_steps_per_second": 5.812, | |
| "step": 334692 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "grad_norm": 0.3532341718673706, | |
| "learning_rate": 0.0001088707779216194, | |
| "loss": 2.7902, | |
| "step": 335000 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "grad_norm": 0.340742826461792, | |
| "learning_rate": 0.00010593150523714253, | |
| "loss": 2.7657, | |
| "step": 336000 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "grad_norm": 0.29802101850509644, | |
| "learning_rate": 0.00010298929033776628, | |
| "loss": 2.7679, | |
| "step": 337000 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "grad_norm": 0.2924386262893677, | |
| "learning_rate": 0.00010004707543839002, | |
| "loss": 2.7702, | |
| "step": 338000 | |
| }, | |
| { | |
| "epoch": 18.23, | |
| "grad_norm": 0.2923070788383484, | |
| "learning_rate": 9.710780275391315e-05, | |
| "loss": 2.7676, | |
| "step": 339000 | |
| }, | |
| { | |
| "epoch": 18.29, | |
| "grad_norm": 0.2915445566177368, | |
| "learning_rate": 9.41655878545369e-05, | |
| "loss": 2.7725, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "grad_norm": 0.2909591794013977, | |
| "learning_rate": 9.122631517006002e-05, | |
| "loss": 2.7718, | |
| "step": 341000 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "grad_norm": 0.3041108548641205, | |
| "learning_rate": 8.828704248558316e-05, | |
| "loss": 2.7736, | |
| "step": 342000 | |
| }, | |
| { | |
| "epoch": 18.45, | |
| "grad_norm": 0.297532320022583, | |
| "learning_rate": 8.534482758620691e-05, | |
| "loss": 2.7728, | |
| "step": 343000 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "grad_norm": 0.3025277554988861, | |
| "learning_rate": 8.240555490173002e-05, | |
| "loss": 2.776, | |
| "step": 344000 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "grad_norm": 0.2893424332141876, | |
| "learning_rate": 7.946334000235378e-05, | |
| "loss": 2.7743, | |
| "step": 345000 | |
| }, | |
| { | |
| "epoch": 18.61, | |
| "grad_norm": 0.2967279553413391, | |
| "learning_rate": 7.652112510297753e-05, | |
| "loss": 2.7797, | |
| "step": 346000 | |
| }, | |
| { | |
| "epoch": 18.66, | |
| "grad_norm": 0.3163110017776489, | |
| "learning_rate": 7.357891020360127e-05, | |
| "loss": 2.7776, | |
| "step": 347000 | |
| }, | |
| { | |
| "epoch": 18.72, | |
| "grad_norm": 0.27627646923065186, | |
| "learning_rate": 7.06396375191244e-05, | |
| "loss": 2.7742, | |
| "step": 348000 | |
| }, | |
| { | |
| "epoch": 18.77, | |
| "grad_norm": 0.29440227150917053, | |
| "learning_rate": 6.769742261974816e-05, | |
| "loss": 2.7737, | |
| "step": 349000 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "grad_norm": 0.28819799423217773, | |
| "learning_rate": 6.47552077203719e-05, | |
| "loss": 2.7803, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 18.88, | |
| "grad_norm": 0.36465972661972046, | |
| "learning_rate": 6.181299282099565e-05, | |
| "loss": 2.7794, | |
| "step": 351000 | |
| }, | |
| { | |
| "epoch": 18.93, | |
| "grad_norm": 0.2758486270904541, | |
| "learning_rate": 5.887372013651877e-05, | |
| "loss": 2.7776, | |
| "step": 352000 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "grad_norm": 0.35758283734321594, | |
| "learning_rate": 5.593150523714252e-05, | |
| "loss": 2.7777, | |
| "step": 353000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.4102629734269421, | |
| "eval_loss": 3.4115071296691895, | |
| "eval_runtime": 155.5477, | |
| "eval_samples_per_second": 372.349, | |
| "eval_steps_per_second": 5.818, | |
| "step": 353286 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "grad_norm": 0.30845823884010315, | |
| "learning_rate": 5.299223255266565e-05, | |
| "loss": 2.764, | |
| "step": 354000 | |
| }, | |
| { | |
| "epoch": 19.09, | |
| "grad_norm": 0.30031487345695496, | |
| "learning_rate": 5.00500176532894e-05, | |
| "loss": 2.76, | |
| "step": 355000 | |
| }, | |
| { | |
| "epoch": 19.15, | |
| "grad_norm": 0.3072240352630615, | |
| "learning_rate": 4.710780275391314e-05, | |
| "loss": 2.7564, | |
| "step": 356000 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "grad_norm": 0.2831727862358093, | |
| "learning_rate": 4.4168530069436276e-05, | |
| "loss": 2.756, | |
| "step": 357000 | |
| }, | |
| { | |
| "epoch": 19.25, | |
| "grad_norm": 0.29800769686698914, | |
| "learning_rate": 4.122631517006002e-05, | |
| "loss": 2.7556, | |
| "step": 358000 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "grad_norm": 0.31122684478759766, | |
| "learning_rate": 3.828410027068377e-05, | |
| "loss": 2.7633, | |
| "step": 359000 | |
| }, | |
| { | |
| "epoch": 19.36, | |
| "grad_norm": 0.3061014413833618, | |
| "learning_rate": 3.53448275862069e-05, | |
| "loss": 2.7596, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 19.41, | |
| "grad_norm": 0.29348084330558777, | |
| "learning_rate": 3.2402612686830645e-05, | |
| "loss": 2.7616, | |
| "step": 361000 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "grad_norm": 0.2803378403186798, | |
| "learning_rate": 2.9460397787454396e-05, | |
| "loss": 2.7585, | |
| "step": 362000 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "grad_norm": 0.31245797872543335, | |
| "learning_rate": 2.6521125102977523e-05, | |
| "loss": 2.7632, | |
| "step": 363000 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "grad_norm": 0.28981760144233704, | |
| "learning_rate": 2.357891020360127e-05, | |
| "loss": 2.7584, | |
| "step": 364000 | |
| }, | |
| { | |
| "epoch": 19.63, | |
| "grad_norm": 0.311199814081192, | |
| "learning_rate": 2.063669530422502e-05, | |
| "loss": 2.7607, | |
| "step": 365000 | |
| }, | |
| { | |
| "epoch": 19.68, | |
| "grad_norm": 0.28949621319770813, | |
| "learning_rate": 1.7694480404848772e-05, | |
| "loss": 2.7595, | |
| "step": 366000 | |
| }, | |
| { | |
| "epoch": 19.74, | |
| "grad_norm": 0.2877204716205597, | |
| "learning_rate": 1.4755207720371896e-05, | |
| "loss": 2.7597, | |
| "step": 367000 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "grad_norm": 0.28899291157722473, | |
| "learning_rate": 1.1812992820995646e-05, | |
| "loss": 2.7604, | |
| "step": 368000 | |
| }, | |
| { | |
| "epoch": 19.85, | |
| "grad_norm": 0.29409030079841614, | |
| "learning_rate": 8.873720136518773e-06, | |
| "loss": 2.761, | |
| "step": 369000 | |
| }, | |
| { | |
| "epoch": 19.9, | |
| "grad_norm": 0.2844576835632324, | |
| "learning_rate": 5.934447452041897e-06, | |
| "loss": 2.7607, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 19.95, | |
| "grad_norm": 0.30620551109313965, | |
| "learning_rate": 2.992232552665647e-06, | |
| "loss": 2.7582, | |
| "step": 371000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.41011443389103847, | |
| "eval_loss": 3.4167680740356445, | |
| "eval_runtime": 156.3007, | |
| "eval_samples_per_second": 370.555, | |
| "eval_steps_per_second": 5.79, | |
| "step": 371880 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 371880, | |
| "total_flos": 1.56683622053376e+18, | |
| "train_loss": 3.0287782370153007, | |
| "train_runtime": 81352.6361, | |
| "train_samples_per_second": 146.277, | |
| "train_steps_per_second": 4.571 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 371880, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 5000, | |
| "total_flos": 1.56683622053376e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |