| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.976, | |
| "eval_steps": 500, | |
| "global_step": 93, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.032, | |
| "grad_norm": 5.849673926536721, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 0.8186, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": 5.8970181550625895, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.8458, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 5.697363226671905, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.8305, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 5.841753636905696, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.868, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 5.671243517334994, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.8036, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 4.599014608569536, | |
| "learning_rate": 3e-06, | |
| "loss": 0.8808, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.224, | |
| "grad_norm": 4.220193980086825, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.8373, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 2.379366858517578, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.7709, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 2.195706729610227, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.7403, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 3.8473606579954094, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7736, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.352, | |
| "grad_norm": 4.215166044150153, | |
| "learning_rate": 4.998209387040829e-06, | |
| "loss": 0.8299, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 4.295222651552718, | |
| "learning_rate": 4.992840113199131e-06, | |
| "loss": 0.7895, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.416, | |
| "grad_norm": 3.8988289446476796, | |
| "learning_rate": 4.983899869907963e-06, | |
| "loss": 0.723, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.448, | |
| "grad_norm": 3.839958855257017, | |
| "learning_rate": 4.971401463979722e-06, | |
| "loss": 0.7461, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.069526753856086, | |
| "learning_rate": 4.955362799260507e-06, | |
| "loss": 0.6986, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 2.7386584473852658, | |
| "learning_rate": 4.935806850983034e-06, | |
| "loss": 0.7205, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.544, | |
| "grad_norm": 2.02702044199497, | |
| "learning_rate": 4.912761632854834e-06, | |
| "loss": 0.7043, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 1.5415003473808875, | |
| "learning_rate": 4.8862601569288885e-06, | |
| "loss": 0.6786, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.608, | |
| "grad_norm": 1.6945417207348286, | |
| "learning_rate": 4.8563403863141825e-06, | |
| "loss": 0.7001, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 1.3996088584254769, | |
| "learning_rate": 4.823045180793914e-06, | |
| "loss": 0.6194, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 1.7673568712274899, | |
| "learning_rate": 4.786422235429269e-06, | |
| "loss": 0.6456, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.704, | |
| "grad_norm": 1.623347448295577, | |
| "learning_rate": 4.746524012236706e-06, | |
| "loss": 0.6652, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.736, | |
| "grad_norm": 1.2829766714727795, | |
| "learning_rate": 4.703407665036622e-06, | |
| "loss": 0.6032, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 1.3613789016207494, | |
| "learning_rate": 4.657134957581057e-06, | |
| "loss": 0.7089, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.9373480619558588, | |
| "learning_rate": 4.607772175077712e-06, | |
| "loss": 0.6395, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.832, | |
| "grad_norm": 1.12602050571956, | |
| "learning_rate": 4.555390029237026e-06, | |
| "loss": 0.7021, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 1.0086309702493645, | |
| "learning_rate": 4.5000635569783365e-06, | |
| "loss": 0.6345, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 1.234160112595515, | |
| "learning_rate": 4.4418720129402145e-06, | |
| "loss": 0.6273, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.928, | |
| "grad_norm": 0.9735941294239923, | |
| "learning_rate": 4.3808987559489536e-06, | |
| "loss": 0.6497, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.8806041823335348, | |
| "learning_rate": 4.317231129607859e-06, | |
| "loss": 0.6714, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.992, | |
| "grad_norm": 0.9271923458351733, | |
| "learning_rate": 4.2509603371783776e-06, | |
| "loss": 0.6328, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.024, | |
| "grad_norm": 1.5098478417524255, | |
| "learning_rate": 4.1821813109322975e-06, | |
| "loss": 0.9666, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 0.8108394298004445, | |
| "learning_rate": 4.110992576162193e-06, | |
| "loss": 0.5684, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.088, | |
| "grad_norm": 0.8548590696244006, | |
| "learning_rate": 4.037496110044885e-06, | |
| "loss": 0.5749, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.8745126302896333, | |
| "learning_rate": 3.961797195560118e-06, | |
| "loss": 0.614, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 0.7435496429501566, | |
| "learning_rate": 3.884004270673711e-06, | |
| "loss": 0.5273, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.184, | |
| "grad_norm": 0.7685963393034175, | |
| "learning_rate": 3.8042287730012117e-06, | |
| "loss": 0.6375, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.216, | |
| "grad_norm": 0.8434136566563109, | |
| "learning_rate": 3.7225849801745835e-06, | |
| "loss": 0.6295, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 0.7860421945854131, | |
| "learning_rate": 3.6391898461406045e-06, | |
| "loss": 0.5147, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.6986247608098803, | |
| "learning_rate": 3.55416283362546e-06, | |
| "loss": 0.5539, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.312, | |
| "grad_norm": 0.6004937658413751, | |
| "learning_rate": 3.4676257430055438e-06, | |
| "loss": 0.4879, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.6936826995472081, | |
| "learning_rate": 3.3797025378295826e-06, | |
| "loss": 0.627, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.376, | |
| "grad_norm": 0.5956362371010271, | |
| "learning_rate": 3.29051916724206e-06, | |
| "loss": 0.4996, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.408, | |
| "grad_norm": 0.6859466931561792, | |
| "learning_rate": 3.2002033855622683e-06, | |
| "loss": 0.6106, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.843708683499078, | |
| "learning_rate": 3.1088845692774798e-06, | |
| "loss": 0.644, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.472, | |
| "grad_norm": 0.6343088750841447, | |
| "learning_rate": 3.0166935317123824e-06, | |
| "loss": 0.5099, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.504, | |
| "grad_norm": 0.7271302183612709, | |
| "learning_rate": 2.9237623356402423e-06, | |
| "loss": 0.5716, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.7366637555716543, | |
| "learning_rate": 2.8302241041042564e-06, | |
| "loss": 0.5895, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.568, | |
| "grad_norm": 0.8301563843419107, | |
| "learning_rate": 2.7362128297200784e-06, | |
| "loss": 0.6191, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.6840596491206852, | |
| "learning_rate": 2.6418631827326857e-06, | |
| "loss": 0.5595, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.7095662290542607, | |
| "learning_rate": 2.547310318102548e-06, | |
| "loss": 0.5506, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.6640000000000001, | |
| "grad_norm": 0.7447871051048682, | |
| "learning_rate": 2.4526896818974534e-06, | |
| "loss": 0.6013, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.696, | |
| "grad_norm": 0.78939028510505, | |
| "learning_rate": 2.358136817267315e-06, | |
| "loss": 0.5804, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.7221168629786532, | |
| "learning_rate": 2.263787170279922e-06, | |
| "loss": 0.5112, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.6843052552942253, | |
| "learning_rate": 2.169775895895745e-06, | |
| "loss": 0.5302, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.792, | |
| "grad_norm": 0.6466865515836445, | |
| "learning_rate": 2.0762376643597586e-06, | |
| "loss": 0.5912, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.6241622145357679, | |
| "learning_rate": 1.9833064682876175e-06, | |
| "loss": 0.524, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.8559999999999999, | |
| "grad_norm": 0.6681670660427579, | |
| "learning_rate": 1.8911154307225204e-06, | |
| "loss": 0.567, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.888, | |
| "grad_norm": 0.6084101203959995, | |
| "learning_rate": 1.7997966144377328e-06, | |
| "loss": 0.4889, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6661890505529855, | |
| "learning_rate": 1.7094808327579401e-06, | |
| "loss": 0.5566, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.952, | |
| "grad_norm": 0.7309734200574032, | |
| "learning_rate": 1.6202974621704176e-06, | |
| "loss": 0.6002, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.984, | |
| "grad_norm": 0.6816696022945047, | |
| "learning_rate": 1.5323742569944573e-06, | |
| "loss": 0.5745, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.3649323993886235, | |
| "learning_rate": 1.44583716637454e-06, | |
| "loss": 0.8374, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.048, | |
| "grad_norm": 0.6467643294360865, | |
| "learning_rate": 1.3608101538593965e-06, | |
| "loss": 0.5162, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.685164030556072, | |
| "learning_rate": 1.277415019825417e-06, | |
| "loss": 0.5509, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.6039417194223102, | |
| "learning_rate": 1.195771226998789e-06, | |
| "loss": 0.505, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.144, | |
| "grad_norm": 0.6457046464618058, | |
| "learning_rate": 1.1159957293262888e-06, | |
| "loss": 0.5505, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.176, | |
| "grad_norm": 0.51006085110857, | |
| "learning_rate": 1.0382028044398823e-06, | |
| "loss": 0.4181, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.5817319940273514, | |
| "learning_rate": 9.625038899551162e-07, | |
| "loss": 0.5063, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.616719797907229, | |
| "learning_rate": 8.890074238378074e-07, | |
| "loss": 0.5717, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.2720000000000002, | |
| "grad_norm": 0.5698471944044463, | |
| "learning_rate": 8.178186890677029e-07, | |
| "loss": 0.5182, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.6181330817750901, | |
| "learning_rate": 7.490396628216237e-07, | |
| "loss": 0.5755, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.336, | |
| "grad_norm": 0.5507693283896078, | |
| "learning_rate": 6.827688703921407e-07, | |
| "loss": 0.4823, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.368, | |
| "grad_norm": 0.5955200432739955, | |
| "learning_rate": 6.191012440510469e-07, | |
| "loss": 0.53, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.5637517004790812, | |
| "learning_rate": 5.581279870597866e-07, | |
| "loss": 0.4502, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.432, | |
| "grad_norm": 0.7466890560657303, | |
| "learning_rate": 4.999364430216639e-07, | |
| "loss": 0.6202, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.464, | |
| "grad_norm": 0.5634548289522667, | |
| "learning_rate": 4.4460997076297504e-07, | |
| "loss": 0.4654, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.5460988308014747, | |
| "learning_rate": 3.922278249222894e-07, | |
| "loss": 0.4881, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.528, | |
| "grad_norm": 0.5526022579328083, | |
| "learning_rate": 3.4286504241894283e-07, | |
| "loss": 0.4975, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.622419564871221, | |
| "learning_rate": 2.965923349633779e-07, | |
| "loss": 0.605, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.6459977525358341, | |
| "learning_rate": 2.53475987763295e-07, | |
| "loss": 0.5747, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.624, | |
| "grad_norm": 0.577093639398248, | |
| "learning_rate": 2.135777645707318e-07, | |
| "loss": 0.4929, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.656, | |
| "grad_norm": 0.5651216938831269, | |
| "learning_rate": 1.7695481920608716e-07, | |
| "loss": 0.5086, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6067146555637166, | |
| "learning_rate": 1.4365961368581844e-07, | |
| "loss": 0.565, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.6317650182194224, | |
| "learning_rate": 1.137398430711123e-07, | |
| "loss": 0.603, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.752, | |
| "grad_norm": 0.44966045140241284, | |
| "learning_rate": 8.723836714516681e-08, | |
| "loss": 0.3893, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5953264863370202, | |
| "learning_rate": 6.419314901696671e-08, | |
| "loss": 0.5293, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.816, | |
| "grad_norm": 0.5479215502914991, | |
| "learning_rate": 4.4637200739493514e-08, | |
| "loss": 0.5167, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.848, | |
| "grad_norm": 0.48730983980669906, | |
| "learning_rate": 2.8598536020278678e-08, | |
| "loss": 0.4821, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.6545838078994339, | |
| "learning_rate": 1.6100130092037704e-08, | |
| "loss": 0.645, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.912, | |
| "grad_norm": 0.5375454814372095, | |
| "learning_rate": 7.159886800869875e-09, | |
| "loss": 0.4611, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.944, | |
| "grad_norm": 0.5959972228051752, | |
| "learning_rate": 1.7906129591713228e-09, | |
| "loss": 0.5646, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.5297891060878839, | |
| "learning_rate": 0.0, | |
| "loss": 0.4423, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "step": 93, | |
| "total_flos": 5.107931021941146e+16, | |
| "train_loss": 0.6129485577024439, | |
| "train_runtime": 3284.9265, | |
| "train_samples_per_second": 0.91, | |
| "train_steps_per_second": 0.028 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 93, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 5.107931021941146e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |