| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.148179292678833, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5675, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.114451289176941, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6005, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.8947972655296326, | |
| "learning_rate": 0.00012, | |
| "loss": 0.5333, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.6874412298202515, | |
| "learning_rate": 0.00016, | |
| "loss": 0.5126, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.6369613409042358, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5243, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.5632033944129944, | |
| "learning_rate": 0.00019994532573409262, | |
| "loss": 0.5335, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.4783449172973633, | |
| "learning_rate": 0.00019978136272187747, | |
| "loss": 0.5066, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.4505637586116791, | |
| "learning_rate": 0.00019950829025450114, | |
| "loss": 0.4819, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.4964255690574646, | |
| "learning_rate": 0.00019912640693269752, | |
| "loss": 0.4949, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.48732897639274597, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 0.5023, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.46126267313957214, | |
| "learning_rate": 0.00019803799658748094, | |
| "loss": 0.4869, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.4295591711997986, | |
| "learning_rate": 0.0001973326597248006, | |
| "loss": 0.4706, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.4603804647922516, | |
| "learning_rate": 0.00019652089102773488, | |
| "loss": 0.4861, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.43054094910621643, | |
| "learning_rate": 0.00019560357815343577, | |
| "loss": 0.5147, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.41392526030540466, | |
| "learning_rate": 0.00019458172417006347, | |
| "loss": 0.4657, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.3919243812561035, | |
| "learning_rate": 0.0001934564464599461, | |
| "loss": 0.4649, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.38946983218193054, | |
| "learning_rate": 0.00019222897549773848, | |
| "loss": 0.4626, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.3812195956707001, | |
| "learning_rate": 0.00019090065350491626, | |
| "loss": 0.4537, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.37084001302719116, | |
| "learning_rate": 0.00018947293298207635, | |
| "loss": 0.4489, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.40874817967414856, | |
| "learning_rate": 0.0001879473751206489, | |
| "loss": 0.4583, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.38442370295524597, | |
| "learning_rate": 0.00018632564809575742, | |
| "loss": 0.4279, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.3699128329753876, | |
| "learning_rate": 0.00018460952524209355, | |
| "loss": 0.4404, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.40334978699684143, | |
| "learning_rate": 0.00018280088311480201, | |
| "loss": 0.4448, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.3896102011203766, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.4421, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.3784579336643219, | |
| "learning_rate": 0.00017891405093963938, | |
| "loss": 0.4538, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.3841725289821625, | |
| "learning_rate": 0.00017684011108568592, | |
| "loss": 0.4251, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.3888923227787018, | |
| "learning_rate": 0.0001746821476984154, | |
| "loss": 0.3996, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.40714430809020996, | |
| "learning_rate": 0.00017244252047910892, | |
| "loss": 0.4298, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.395693838596344, | |
| "learning_rate": 0.00017012367842724887, | |
| "loss": 0.4044, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.3896782398223877, | |
| "learning_rate": 0.00016772815716257412, | |
| "loss": 0.4093, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.3841933310031891, | |
| "learning_rate": 0.00016525857615241687, | |
| "loss": 0.4139, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.3779061436653137, | |
| "learning_rate": 0.0001627176358473537, | |
| "loss": 0.3962, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.4033326208591461, | |
| "learning_rate": 0.00016010811472830252, | |
| "loss": 0.3883, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.3781256079673767, | |
| "learning_rate": 0.00015743286626829437, | |
| "loss": 0.3858, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.37716230750083923, | |
| "learning_rate": 0.00015469481581224272, | |
| "loss": 0.3938, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.395526260137558, | |
| "learning_rate": 0.00015189695737812152, | |
| "loss": 0.4122, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.368266224861145, | |
| "learning_rate": 0.00014904235038305083, | |
| "loss": 0.403, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.37806516885757446, | |
| "learning_rate": 0.0001461341162978688, | |
| "loss": 0.374, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.37161675095558167, | |
| "learning_rate": 0.00014317543523384928, | |
| "loss": 0.3792, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.3814436197280884, | |
| "learning_rate": 0.00014016954246529696, | |
| "loss": 0.3824, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.3864171504974365, | |
| "learning_rate": 0.00013711972489182208, | |
| "loss": 0.3693, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.35850077867507935, | |
| "learning_rate": 0.00013402931744416433, | |
| "loss": 0.3623, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.36775171756744385, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.3643, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.37669485807418823, | |
| "learning_rate": 0.00012774029087618446, | |
| "loss": 0.3756, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.35857564210891724, | |
| "learning_rate": 0.00012454854871407994, | |
| "loss": 0.3947, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.3690029978752136, | |
| "learning_rate": 0.0001213299630743747, | |
| "loss": 0.3548, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.3639805018901825, | |
| "learning_rate": 0.000118088053433211, | |
| "loss": 0.3627, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.37132132053375244, | |
| "learning_rate": 0.0001148263647711842, | |
| "loss": 0.3668, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.37001127004623413, | |
| "learning_rate": 0.00011154846369695863, | |
| "loss": 0.3687, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.3796997368335724, | |
| "learning_rate": 0.00010825793454723325, | |
| "loss": 0.3819, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 0.29557618498802185, | |
| "learning_rate": 0.00010495837546732224, | |
| "loss": 0.2758, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.31048524379730225, | |
| "learning_rate": 0.00010165339447663587, | |
| "loss": 0.2548, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 0.3087123930454254, | |
| "learning_rate": 9.834660552336415e-05, | |
| "loss": 0.2647, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.3230942487716675, | |
| "learning_rate": 9.504162453267777e-05, | |
| "loss": 0.2571, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.3078308701515198, | |
| "learning_rate": 9.174206545276677e-05, | |
| "loss": 0.247, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.3151338994503021, | |
| "learning_rate": 8.845153630304139e-05, | |
| "loss": 0.2649, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.1400000000000001, | |
| "grad_norm": 0.32113245129585266, | |
| "learning_rate": 8.517363522881579e-05, | |
| "loss": 0.2441, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.3001263737678528, | |
| "learning_rate": 8.191194656678904e-05, | |
| "loss": 0.2545, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 0.31521672010421753, | |
| "learning_rate": 7.867003692562534e-05, | |
| "loss": 0.25, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.3146115839481354, | |
| "learning_rate": 7.54514512859201e-05, | |
| "loss": 0.2758, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 0.30817195773124695, | |
| "learning_rate": 7.225970912381556e-05, | |
| "loss": 0.2639, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.3167981505393982, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.2523, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 0.33051297068595886, | |
| "learning_rate": 6.59706825558357e-05, | |
| "loss": 0.2478, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.32061681151390076, | |
| "learning_rate": 6.28802751081779e-05, | |
| "loss": 0.2292, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.32867640256881714, | |
| "learning_rate": 5.983045753470308e-05, | |
| "loss": 0.2506, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.32115456461906433, | |
| "learning_rate": 5.6824564766150726e-05, | |
| "loss": 0.2514, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 0.3387518525123596, | |
| "learning_rate": 5.386588370213124e-05, | |
| "loss": 0.244, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.32781389355659485, | |
| "learning_rate": 5.095764961694922e-05, | |
| "loss": 0.2424, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 0.3345761299133301, | |
| "learning_rate": 4.810304262187852e-05, | |
| "loss": 0.2513, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.3255421817302704, | |
| "learning_rate": 4.530518418775733e-05, | |
| "loss": 0.2456, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 0.3354507088661194, | |
| "learning_rate": 4.256713373170564e-05, | |
| "loss": 0.2457, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.3373594284057617, | |
| "learning_rate": 3.9891885271697496e-05, | |
| "loss": 0.2475, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 0.32666316628456116, | |
| "learning_rate": 3.7282364152646297e-05, | |
| "loss": 0.2301, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.3279685974121094, | |
| "learning_rate": 3.4741423847583134e-05, | |
| "loss": 0.2472, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.3301085829734802, | |
| "learning_rate": 3.227184283742591e-05, | |
| "loss": 0.2294, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.33756864070892334, | |
| "learning_rate": 2.9876321572751144e-05, | |
| "loss": 0.2319, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 0.3326968848705292, | |
| "learning_rate": 2.7557479520891104e-05, | |
| "loss": 0.2435, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.3414022624492645, | |
| "learning_rate": 2.5317852301584643e-05, | |
| "loss": 0.2365, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 0.3446585237979889, | |
| "learning_rate": 2.315988891431412e-05, | |
| "loss": 0.2353, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.33243560791015625, | |
| "learning_rate": 2.1085949060360654e-05, | |
| "loss": 0.2604, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 0.3418924808502197, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.2546, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.33969858288764954, | |
| "learning_rate": 1.7199116885197995e-05, | |
| "loss": 0.2546, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.6600000000000001, | |
| "grad_norm": 0.3274635076522827, | |
| "learning_rate": 1.5390474757906446e-05, | |
| "loss": 0.2336, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.3252343237400055, | |
| "learning_rate": 1.3674351904242611e-05, | |
| "loss": 0.2307, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.3318844735622406, | |
| "learning_rate": 1.2052624879351104e-05, | |
| "loss": 0.2273, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.3397023677825928, | |
| "learning_rate": 1.0527067017923654e-05, | |
| "loss": 0.237, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 0.32320481538772583, | |
| "learning_rate": 9.09934649508375e-06, | |
| "loss": 0.24, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.34301823377609253, | |
| "learning_rate": 7.771024502261526e-06, | |
| "loss": 0.2447, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 0.33157962560653687, | |
| "learning_rate": 6.543553540053926e-06, | |
| "loss": 0.2512, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.33687952160835266, | |
| "learning_rate": 5.418275829936537e-06, | |
| "loss": 0.2538, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.8199999999999998, | |
| "grad_norm": 0.3404587209224701, | |
| "learning_rate": 4.3964218465642355e-06, | |
| "loss": 0.258, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.3458779752254486, | |
| "learning_rate": 3.4791089722651436e-06, | |
| "loss": 0.2392, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.8599999999999999, | |
| "grad_norm": 0.33880794048309326, | |
| "learning_rate": 2.667340275199426e-06, | |
| "loss": 0.2542, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.3243839144706726, | |
| "learning_rate": 1.9620034125190644e-06, | |
| "loss": 0.2371, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.3197105824947357, | |
| "learning_rate": 1.3638696597277679e-06, | |
| "loss": 0.2237, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.32444778084754944, | |
| "learning_rate": 8.735930673024806e-07, | |
| "loss": 0.2331, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 0.3440603017807007, | |
| "learning_rate": 4.917097454988584e-07, | |
| "loss": 0.257, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.3220385015010834, | |
| "learning_rate": 2.1863727812254653e-07, | |
| "loss": 0.2357, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 0.3219694495201111, | |
| "learning_rate": 5.467426590739511e-08, | |
| "loss": 0.2483, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.3302796185016632, | |
| "learning_rate": 0.0, | |
| "loss": 0.2355, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.5112362971367014e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |