| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03576266594418857, | |
| "eval_steps": 50, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00036123904994129867, | |
| "grad_norm": 11.258881568908691, | |
| "learning_rate": 5e-05, | |
| "loss": 4.9805, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00036123904994129867, | |
| "eval_loss": 1.3654987812042236, | |
| "eval_runtime": 65.6871, | |
| "eval_samples_per_second": 17.751, | |
| "eval_steps_per_second": 8.875, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007224780998825973, | |
| "grad_norm": 11.392064094543457, | |
| "learning_rate": 0.0001, | |
| "loss": 5.6516, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.001083717149823896, | |
| "grad_norm": 10.067161560058594, | |
| "learning_rate": 0.00015, | |
| "loss": 5.1962, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0014449561997651947, | |
| "grad_norm": 8.352842330932617, | |
| "learning_rate": 0.0002, | |
| "loss": 4.3996, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0018061952497064934, | |
| "grad_norm": 6.142587661743164, | |
| "learning_rate": 0.00025, | |
| "loss": 3.9857, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.002167434299647792, | |
| "grad_norm": 12.07518196105957, | |
| "learning_rate": 0.0003, | |
| "loss": 5.3352, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0025286733495890907, | |
| "grad_norm": 7.874082088470459, | |
| "learning_rate": 0.00035, | |
| "loss": 4.9328, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0028899123995303894, | |
| "grad_norm": 4.830751895904541, | |
| "learning_rate": 0.0004, | |
| "loss": 3.4086, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.003251151449471688, | |
| "grad_norm": 4.93334436416626, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 3.7905, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0036123904994129867, | |
| "grad_norm": 5.134774684906006, | |
| "learning_rate": 0.0005, | |
| "loss": 4.1506, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.003973629549354285, | |
| "grad_norm": 5.855077266693115, | |
| "learning_rate": 0.0004998442655654946, | |
| "loss": 3.7257, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.004334868599295584, | |
| "grad_norm": 5.537413120269775, | |
| "learning_rate": 0.0004993772562876909, | |
| "loss": 4.3003, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004696107649236883, | |
| "grad_norm": 5.094138145446777, | |
| "learning_rate": 0.0004985995540019955, | |
| "loss": 4.8794, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.005057346699178181, | |
| "grad_norm": 4.8982930183410645, | |
| "learning_rate": 0.0004975121276286136, | |
| "loss": 4.6231, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.00541858574911948, | |
| "grad_norm": 4.569957733154297, | |
| "learning_rate": 0.0004961163319653958, | |
| "loss": 4.0061, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005779824799060779, | |
| "grad_norm": 7.189935207366943, | |
| "learning_rate": 0.0004944139059999286, | |
| "loss": 5.0499, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.006141063849002077, | |
| "grad_norm": 4.611532688140869, | |
| "learning_rate": 0.000492406970742972, | |
| "loss": 3.2843, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.006502302898943376, | |
| "grad_norm": 4.90513277053833, | |
| "learning_rate": 0.0004900980265859448, | |
| "loss": 3.4266, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.006863541948884675, | |
| "grad_norm": 6.561873435974121, | |
| "learning_rate": 0.0004874899501857477, | |
| "loss": 4.2261, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.007224780998825973, | |
| "grad_norm": 11.763420104980469, | |
| "learning_rate": 0.00048458599088080736, | |
| "loss": 3.9256, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.007586020048767272, | |
| "grad_norm": 5.913724422454834, | |
| "learning_rate": 0.0004813897666428053, | |
| "loss": 3.4543, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00794725909870857, | |
| "grad_norm": 8.865498542785645, | |
| "learning_rate": 0.00047790525956913543, | |
| "loss": 5.0105, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.00830849814864987, | |
| "grad_norm": 6.202934265136719, | |
| "learning_rate": 0.0004741368109217071, | |
| "loss": 4.0654, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.008669737198591168, | |
| "grad_norm": 9.578519821166992, | |
| "learning_rate": 0.00047008911571827283, | |
| "loss": 5.3396, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.009030976248532467, | |
| "grad_norm": 7.616293430328369, | |
| "learning_rate": 0.00046576721688302105, | |
| "loss": 3.7428, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.009392215298473765, | |
| "grad_norm": 5.799994945526123, | |
| "learning_rate": 0.0004611764989637205, | |
| "loss": 4.0877, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.009753454348415064, | |
| "grad_norm": 5.58920955657959, | |
| "learning_rate": 0.0004563226814232444, | |
| "loss": 4.2916, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.010114693398356363, | |
| "grad_norm": 4.770939826965332, | |
| "learning_rate": 0.0004512118115138315, | |
| "loss": 4.2956, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.010475932448297661, | |
| "grad_norm": 4.9692063331604, | |
| "learning_rate": 0.0004458502567429631, | |
| "loss": 3.542, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.01083717149823896, | |
| "grad_norm": 4.753838539123535, | |
| "learning_rate": 0.00044024469694024196, | |
| "loss": 3.0264, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.011198410548180259, | |
| "grad_norm": 5.1347575187683105, | |
| "learning_rate": 0.00043440211593515554, | |
| "loss": 3.1049, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.011559649598121557, | |
| "grad_norm": 5.5637006759643555, | |
| "learning_rate": 0.0004283297928560951, | |
| "loss": 3.8163, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.011920888648062856, | |
| "grad_norm": 5.401363849639893, | |
| "learning_rate": 0.0004220352930614672, | |
| "loss": 3.8318, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.012282127698004155, | |
| "grad_norm": 5.22984504699707, | |
| "learning_rate": 0.00041552645871420013, | |
| "loss": 3.8126, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.012643366747945453, | |
| "grad_norm": 4.754082679748535, | |
| "learning_rate": 0.00040881139901138467, | |
| "loss": 4.0395, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.013004605797886752, | |
| "grad_norm": 4.820829391479492, | |
| "learning_rate": 0.00040189848008122475, | |
| "loss": 3.1665, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01336584484782805, | |
| "grad_norm": 3.9806621074676514, | |
| "learning_rate": 0.00039479631455988334, | |
| "loss": 2.9394, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.01372708389776935, | |
| "grad_norm": 4.494911193847656, | |
| "learning_rate": 0.0003875137508612103, | |
| "loss": 3.1021, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.014088322947710648, | |
| "grad_norm": 4.835848331451416, | |
| "learning_rate": 0.00038005986215272055, | |
| "loss": 3.6138, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.014449561997651947, | |
| "grad_norm": 5.597693920135498, | |
| "learning_rate": 0.0003724439350515571, | |
| "loss": 4.5553, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.014810801047593245, | |
| "grad_norm": 4.290703773498535, | |
| "learning_rate": 0.0003646754580545226, | |
| "loss": 3.271, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.015172040097534544, | |
| "grad_norm": 4.479038238525391, | |
| "learning_rate": 0.000356764109716594, | |
| "loss": 3.1925, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.015533279147475843, | |
| "grad_norm": 4.6170654296875, | |
| "learning_rate": 0.00034871974659264783, | |
| "loss": 2.8589, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.01589451819741714, | |
| "grad_norm": 5.0329060554504395, | |
| "learning_rate": 0.0003405523909574206, | |
| "loss": 3.1158, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.01625575724735844, | |
| "grad_norm": 5.719956398010254, | |
| "learning_rate": 0.0003322722183190025, | |
| "loss": 3.856, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.01661699629729974, | |
| "grad_norm": 6.794640064239502, | |
| "learning_rate": 0.0003238895447414211, | |
| "loss": 4.4511, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.016978235347241036, | |
| "grad_norm": 3.9683847427368164, | |
| "learning_rate": 0.0003154148139921102, | |
| "loss": 2.8105, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.017339474397182336, | |
| "grad_norm": 5.173192024230957, | |
| "learning_rate": 0.00030685858453027663, | |
| "loss": 2.7872, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.017700713447123633, | |
| "grad_norm": 3.741673469543457, | |
| "learning_rate": 0.0002982315163523742, | |
| "loss": 2.6865, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.018061952497064934, | |
| "grad_norm": 3.9549710750579834, | |
| "learning_rate": 0.000289544357711076, | |
| "loss": 3.0372, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.018061952497064934, | |
| "eval_loss": 0.8259496092796326, | |
| "eval_runtime": 66.5192, | |
| "eval_samples_per_second": 17.529, | |
| "eval_steps_per_second": 8.764, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01842319154700623, | |
| "grad_norm": 5.567710876464844, | |
| "learning_rate": 0.0002808079317242896, | |
| "loss": 3.7515, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.01878443059694753, | |
| "grad_norm": 7.243776321411133, | |
| "learning_rate": 0.0002720331228909005, | |
| "loss": 2.6164, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.019145669646888828, | |
| "grad_norm": 4.8548431396484375, | |
| "learning_rate": 0.00026323086353004075, | |
| "loss": 3.2759, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.019506908696830128, | |
| "grad_norm": 3.919628858566284, | |
| "learning_rate": 0.0002544121201607822, | |
| "loss": 2.9809, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.019868147746771425, | |
| "grad_norm": 5.447530269622803, | |
| "learning_rate": 0.00024558787983921783, | |
| "loss": 3.5247, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.020229386796712726, | |
| "grad_norm": 5.87801456451416, | |
| "learning_rate": 0.0002367691364699592, | |
| "loss": 2.7521, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.020590625846654022, | |
| "grad_norm": 4.117478847503662, | |
| "learning_rate": 0.00022796687710909964, | |
| "loss": 3.7246, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.020951864896595323, | |
| "grad_norm": 3.548586845397949, | |
| "learning_rate": 0.00021919206827571036, | |
| "loss": 3.4873, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.02131310394653662, | |
| "grad_norm": 4.6662702560424805, | |
| "learning_rate": 0.00021045564228892402, | |
| "loss": 3.6514, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.02167434299647792, | |
| "grad_norm": 3.70032000541687, | |
| "learning_rate": 0.00020176848364762578, | |
| "loss": 3.4641, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.022035582046419217, | |
| "grad_norm": 5.118409633636475, | |
| "learning_rate": 0.00019314141546972343, | |
| "loss": 3.7131, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.022396821096360518, | |
| "grad_norm": 5.562695026397705, | |
| "learning_rate": 0.00018458518600788986, | |
| "loss": 3.2622, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.022758060146301815, | |
| "grad_norm": 6.675144195556641, | |
| "learning_rate": 0.00017611045525857898, | |
| "loss": 3.6492, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.023119299196243115, | |
| "grad_norm": 4.012448310852051, | |
| "learning_rate": 0.0001677277816809975, | |
| "loss": 2.7895, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.023480538246184412, | |
| "grad_norm": 3.4179670810699463, | |
| "learning_rate": 0.00015944760904257942, | |
| "loss": 2.6486, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.023841777296125712, | |
| "grad_norm": 5.845145225524902, | |
| "learning_rate": 0.0001512802534073522, | |
| "loss": 4.1475, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.02420301634606701, | |
| "grad_norm": 3.7966604232788086, | |
| "learning_rate": 0.00014323589028340596, | |
| "loss": 3.6578, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.02456425539600831, | |
| "grad_norm": 4.147678852081299, | |
| "learning_rate": 0.00013532454194547733, | |
| "loss": 3.165, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.024925494445949607, | |
| "grad_norm": 4.22127628326416, | |
| "learning_rate": 0.00012755606494844294, | |
| "loss": 3.6101, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.025286733495890907, | |
| "grad_norm": 3.980015516281128, | |
| "learning_rate": 0.00011994013784727947, | |
| "loss": 2.9809, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.025647972545832204, | |
| "grad_norm": 3.8896799087524414, | |
| "learning_rate": 0.00011248624913878966, | |
| "loss": 2.7593, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.026009211595773504, | |
| "grad_norm": 3.7454440593719482, | |
| "learning_rate": 0.0001052036854401166, | |
| "loss": 2.7478, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0263704506457148, | |
| "grad_norm": 3.633699417114258, | |
| "learning_rate": 9.810151991877531e-05, | |
| "loss": 3.4857, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.0267316896956561, | |
| "grad_norm": 3.71138334274292, | |
| "learning_rate": 9.118860098861537e-05, | |
| "loss": 2.4003, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.0270929287455974, | |
| "grad_norm": 3.327180862426758, | |
| "learning_rate": 8.44735412857999e-05, | |
| "loss": 2.5128, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0274541677955387, | |
| "grad_norm": 3.146794080734253, | |
| "learning_rate": 7.79647069385328e-05, | |
| "loss": 2.2593, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.027815406845479996, | |
| "grad_norm": 3.6162102222442627, | |
| "learning_rate": 7.167020714390501e-05, | |
| "loss": 2.586, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.028176645895421296, | |
| "grad_norm": 3.828462600708008, | |
| "learning_rate": 6.559788406484446e-05, | |
| "loss": 2.3276, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.028537884945362593, | |
| "grad_norm": 3.7188730239868164, | |
| "learning_rate": 5.975530305975807e-05, | |
| "loss": 3.0949, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.028899123995303894, | |
| "grad_norm": 4.314111709594727, | |
| "learning_rate": 5.414974325703686e-05, | |
| "loss": 2.537, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.02926036304524519, | |
| "grad_norm": 4.148589134216309, | |
| "learning_rate": 4.8788188486168616e-05, | |
| "loss": 2.833, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.02962160209518649, | |
| "grad_norm": 3.602301836013794, | |
| "learning_rate": 4.367731857675569e-05, | |
| "loss": 2.4284, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.029982841145127788, | |
| "grad_norm": 5.501028537750244, | |
| "learning_rate": 3.882350103627952e-05, | |
| "loss": 2.9707, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.03034408019506909, | |
| "grad_norm": 4.275007247924805, | |
| "learning_rate": 3.423278311697897e-05, | |
| "loss": 2.373, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.030705319245010385, | |
| "grad_norm": 3.906873941421509, | |
| "learning_rate": 2.9910884281727225e-05, | |
| "loss": 3.7784, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.031066558294951686, | |
| "grad_norm": 3.9153549671173096, | |
| "learning_rate": 2.586318907829291e-05, | |
| "loss": 3.0894, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.03142779734489298, | |
| "grad_norm": 4.450911521911621, | |
| "learning_rate": 2.209474043086457e-05, | |
| "loss": 3.2057, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.03178903639483428, | |
| "grad_norm": 3.1464052200317383, | |
| "learning_rate": 1.861023335719475e-05, | |
| "loss": 2.3528, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.03215027544477558, | |
| "grad_norm": 3.700227737426758, | |
| "learning_rate": 1.5414009119192633e-05, | |
| "loss": 2.8889, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.03251151449471688, | |
| "grad_norm": 3.8259425163269043, | |
| "learning_rate": 1.25100498142523e-05, | |
| "loss": 2.3349, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03287275354465818, | |
| "grad_norm": 4.0107316970825195, | |
| "learning_rate": 9.901973414055187e-06, | |
| "loss": 2.3647, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.03323399259459948, | |
| "grad_norm": 3.5053553581237793, | |
| "learning_rate": 7.593029257027956e-06, | |
| "loss": 3.2543, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.03359523164454078, | |
| "grad_norm": 3.6908042430877686, | |
| "learning_rate": 5.5860940000714015e-06, | |
| "loss": 2.9373, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.03395647069448207, | |
| "grad_norm": 3.165818214416504, | |
| "learning_rate": 3.8836680346041594e-06, | |
| "loss": 2.6173, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.03431770974442337, | |
| "grad_norm": 4.046987056732178, | |
| "learning_rate": 2.487872371386424e-06, | |
| "loss": 3.5724, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.03467894879436467, | |
| "grad_norm": 4.770341873168945, | |
| "learning_rate": 1.4004459980045125e-06, | |
| "loss": 3.2451, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.03504018784430597, | |
| "grad_norm": 3.3098256587982178, | |
| "learning_rate": 6.22743712309054e-07, | |
| "loss": 2.3355, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.035401426894247266, | |
| "grad_norm": 3.6350321769714355, | |
| "learning_rate": 1.557344345054501e-07, | |
| "loss": 3.2009, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.03576266594418857, | |
| "grad_norm": 4.967100620269775, | |
| "learning_rate": 0.0, | |
| "loss": 3.182, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 99, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 300, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.308168021016576e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |