| { | |
| "best_metric": 1.967491626739502, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Mistral-7B-v0.3_pct_ortho_r32/checkpoint-8", | |
| "epoch": 0.9981059842836993, | |
| "eval_steps": 8, | |
| "global_step": 387, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025790852307072333, | |
| "grad_norm": 7.501750946044922, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.2239, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010316340922828933, | |
| "grad_norm": 4.365421295166016, | |
| "learning_rate": 5e-05, | |
| "loss": 2.1072, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020632681845657867, | |
| "grad_norm": 3.2742128372192383, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9673, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020632681845657867, | |
| "eval_loss": 1.967491626739502, | |
| "eval_runtime": 94.9785, | |
| "eval_samples_per_second": 2.59, | |
| "eval_steps_per_second": 2.59, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0309490227684868, | |
| "grad_norm": 3.315385580062866, | |
| "learning_rate": 9.997251843068762e-05, | |
| "loss": 1.9892, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04126536369131573, | |
| "grad_norm": 3.2009124755859375, | |
| "learning_rate": 9.989010393221656e-05, | |
| "loss": 1.982, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04126536369131573, | |
| "eval_loss": 1.976182460784912, | |
| "eval_runtime": 122.0257, | |
| "eval_samples_per_second": 2.016, | |
| "eval_steps_per_second": 2.016, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05158170461414467, | |
| "grad_norm": 2.4668922424316406, | |
| "learning_rate": 9.97528470997769e-05, | |
| "loss": 1.931, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0618980455369736, | |
| "grad_norm": 2.5818123817443848, | |
| "learning_rate": 9.956089881469482e-05, | |
| "loss": 1.9563, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0618980455369736, | |
| "eval_loss": 1.9745111465454102, | |
| "eval_runtime": 82.5917, | |
| "eval_samples_per_second": 2.979, | |
| "eval_steps_per_second": 2.979, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07221438645980253, | |
| "grad_norm": 2.7342212200164795, | |
| "learning_rate": 9.931447007857432e-05, | |
| "loss": 2.003, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08253072738263147, | |
| "grad_norm": 2.6568496227264404, | |
| "learning_rate": 9.901383178135113e-05, | |
| "loss": 1.957, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08253072738263147, | |
| "eval_loss": 1.97785484790802, | |
| "eval_runtime": 110.4681, | |
| "eval_samples_per_second": 2.227, | |
| "eval_steps_per_second": 2.227, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09284706830546041, | |
| "grad_norm": 3.633096218109131, | |
| "learning_rate": 9.865931440351337e-05, | |
| "loss": 1.9643, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10316340922828934, | |
| "grad_norm": 2.6414403915405273, | |
| "learning_rate": 9.825130765281668e-05, | |
| "loss": 2.0248, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10316340922828934, | |
| "eval_loss": 1.9851678609848022, | |
| "eval_runtime": 85.4255, | |
| "eval_samples_per_second": 2.88, | |
| "eval_steps_per_second": 2.88, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11347975015111827, | |
| "grad_norm": 4.06710958480835, | |
| "learning_rate": 9.779026003589304e-05, | |
| "loss": 1.9751, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.1237960910739472, | |
| "grad_norm": 2.688786506652832, | |
| "learning_rate": 9.727667836522407e-05, | |
| "loss": 1.9753, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1237960910739472, | |
| "eval_loss": 1.9988040924072266, | |
| "eval_runtime": 83.5761, | |
| "eval_samples_per_second": 2.943, | |
| "eval_steps_per_second": 2.943, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13411243199677614, | |
| "grad_norm": 2.983027696609497, | |
| "learning_rate": 9.6711127202021e-05, | |
| "loss": 2.0274, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14442877291960507, | |
| "grad_norm": 2.963914394378662, | |
| "learning_rate": 9.609422823562345e-05, | |
| "loss": 1.9752, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14442877291960507, | |
| "eval_loss": 1.997394323348999, | |
| "eval_runtime": 109.8713, | |
| "eval_samples_per_second": 2.239, | |
| "eval_steps_per_second": 2.239, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.154745113842434, | |
| "grad_norm": 2.4576425552368164, | |
| "learning_rate": 9.542665960009959e-05, | |
| "loss": 1.9912, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16506145476526293, | |
| "grad_norm": 2.9133598804473877, | |
| "learning_rate": 9.470915512879852e-05, | |
| "loss": 2.0253, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.16506145476526293, | |
| "eval_loss": 1.994693398475647, | |
| "eval_runtime": 85.1875, | |
| "eval_samples_per_second": 2.888, | |
| "eval_steps_per_second": 2.888, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1753777956880919, | |
| "grad_norm": 2.7807934284210205, | |
| "learning_rate": 9.394250354767467e-05, | |
| "loss": 2.0054, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.18569413661092082, | |
| "grad_norm": 2.75816011428833, | |
| "learning_rate": 9.312754760827061e-05, | |
| "loss": 2.0073, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18569413661092082, | |
| "eval_loss": 1.9871970415115356, | |
| "eval_runtime": 116.486, | |
| "eval_samples_per_second": 2.112, | |
| "eval_steps_per_second": 2.112, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19601047753374976, | |
| "grad_norm": 2.501340389251709, | |
| "learning_rate": 9.226518316131176e-05, | |
| "loss": 2.013, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.2063268184565787, | |
| "grad_norm": 3.0418202877044678, | |
| "learning_rate": 9.1356358171931e-05, | |
| "loss": 1.9826, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2063268184565787, | |
| "eval_loss": 1.9953396320343018, | |
| "eval_runtime": 79.6585, | |
| "eval_samples_per_second": 3.088, | |
| "eval_steps_per_second": 3.088, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21664315937940762, | |
| "grad_norm": 2.696978807449341, | |
| "learning_rate": 9.040207167760586e-05, | |
| "loss": 1.9693, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22695950030223655, | |
| "grad_norm": 2.604274034500122, | |
| "learning_rate": 8.940337268995385e-05, | |
| "loss": 1.9907, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22695950030223655, | |
| "eval_loss": 2.0014798641204834, | |
| "eval_runtime": 119.4751, | |
| "eval_samples_per_second": 2.059, | |
| "eval_steps_per_second": 2.059, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.23727584122506548, | |
| "grad_norm": 2.6495442390441895, | |
| "learning_rate": 8.836135904159302e-05, | |
| "loss": 2.0053, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.2475921821478944, | |
| "grad_norm": 2.7287392616271973, | |
| "learning_rate": 8.727717617933544e-05, | |
| "loss": 1.9795, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2475921821478944, | |
| "eval_loss": 1.9950815439224243, | |
| "eval_runtime": 82.9311, | |
| "eval_samples_per_second": 2.966, | |
| "eval_steps_per_second": 2.966, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.25790852307072337, | |
| "grad_norm": 2.6722488403320312, | |
| "learning_rate": 8.615201590504017e-05, | |
| "loss": 2.033, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2682248639935523, | |
| "grad_norm": 3.8678576946258545, | |
| "learning_rate": 8.498711506550983e-05, | |
| "loss": 1.9882, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2682248639935523, | |
| "eval_loss": 2.0020015239715576, | |
| "eval_runtime": 79.5681, | |
| "eval_samples_per_second": 3.092, | |
| "eval_steps_per_second": 3.092, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.27854120491638124, | |
| "grad_norm": 3.1447510719299316, | |
| "learning_rate": 8.378375419287099e-05, | |
| "loss": 2.0347, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.28885754583921014, | |
| "grad_norm": 2.7415413856506348, | |
| "learning_rate": 8.25432560969328e-05, | |
| "loss": 1.9896, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.28885754583921014, | |
| "eval_loss": 1.9962644577026367, | |
| "eval_runtime": 119.6182, | |
| "eval_samples_per_second": 2.057, | |
| "eval_steps_per_second": 2.057, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2991738867620391, | |
| "grad_norm": 2.7823774814605713, | |
| "learning_rate": 8.126698441107146e-05, | |
| "loss": 1.9733, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.309490227684868, | |
| "grad_norm": 2.875882387161255, | |
| "learning_rate": 7.995634209323886e-05, | |
| "loss": 2.0177, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.309490227684868, | |
| "eval_loss": 2.0145890712738037, | |
| "eval_runtime": 82.837, | |
| "eval_samples_per_second": 2.97, | |
| "eval_steps_per_second": 2.97, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.31980656860769696, | |
| "grad_norm": 2.600053548812866, | |
| "learning_rate": 7.861276988374302e-05, | |
| "loss": 2.0148, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.33012290953052587, | |
| "grad_norm": 3.0531198978424072, | |
| "learning_rate": 7.723774472149601e-05, | |
| "loss": 2.0131, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.33012290953052587, | |
| "eval_loss": 2.0012505054473877, | |
| "eval_runtime": 112.1719, | |
| "eval_samples_per_second": 2.193, | |
| "eval_steps_per_second": 2.193, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3404392504533548, | |
| "grad_norm": 2.9197065830230713, | |
| "learning_rate": 7.583277812046993e-05, | |
| "loss": 2.0205, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.3507555913761838, | |
| "grad_norm": 2.766697406768799, | |
| "learning_rate": 7.439941450814591e-05, | |
| "loss": 2.0384, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3507555913761838, | |
| "eval_loss": 2.001709222793579, | |
| "eval_runtime": 84.1224, | |
| "eval_samples_per_second": 2.924, | |
| "eval_steps_per_second": 2.924, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3610719322990127, | |
| "grad_norm": 3.0844602584838867, | |
| "learning_rate": 7.293922952778239e-05, | |
| "loss": 2.0319, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.37138827322184165, | |
| "grad_norm": 3.19889760017395, | |
| "learning_rate": 7.145382830636924e-05, | |
| "loss": 2.0587, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.37138827322184165, | |
| "eval_loss": 2.001880168914795, | |
| "eval_runtime": 114.9668, | |
| "eval_samples_per_second": 2.14, | |
| "eval_steps_per_second": 2.14, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.38170461414467055, | |
| "grad_norm": 3.044001817703247, | |
| "learning_rate": 6.994484369017143e-05, | |
| "loss": 2.0052, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.3920209550674995, | |
| "grad_norm": 3.1708483695983887, | |
| "learning_rate": 6.841393444980177e-05, | |
| "loss": 1.9998, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.3920209550674995, | |
| "eval_loss": 1.996483325958252, | |
| "eval_runtime": 111.8248, | |
| "eval_samples_per_second": 2.2, | |
| "eval_steps_per_second": 2.2, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.4023372959903284, | |
| "grad_norm": 2.9223573207855225, | |
| "learning_rate": 6.686278345679625e-05, | |
| "loss": 1.9855, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.4126536369131574, | |
| "grad_norm": 2.8330798149108887, | |
| "learning_rate": 6.529309583369605e-05, | |
| "loss": 1.9729, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4126536369131574, | |
| "eval_loss": 1.9904886484146118, | |
| "eval_runtime": 85.3147, | |
| "eval_samples_per_second": 2.883, | |
| "eval_steps_per_second": 2.883, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4229699778359863, | |
| "grad_norm": 2.665085792541504, | |
| "learning_rate": 6.370659707966967e-05, | |
| "loss": 1.994, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.43328631875881524, | |
| "grad_norm": 3.091874122619629, | |
| "learning_rate": 6.2105031173736e-05, | |
| "loss": 2.0339, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.43328631875881524, | |
| "eval_loss": 2.0232627391815186, | |
| "eval_runtime": 113.1394, | |
| "eval_samples_per_second": 2.174, | |
| "eval_steps_per_second": 2.174, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.44360265968164414, | |
| "grad_norm": 3.6090798377990723, | |
| "learning_rate": 6.049015865767318e-05, | |
| "loss": 1.9818, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.4539190006044731, | |
| "grad_norm": 3.000669479370117, | |
| "learning_rate": 5.88637547007204e-05, | |
| "loss": 2.0029, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.4539190006044731, | |
| "eval_loss": 1.9972038269042969, | |
| "eval_runtime": 80.5833, | |
| "eval_samples_per_second": 3.053, | |
| "eval_steps_per_second": 3.053, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.46423534152730206, | |
| "grad_norm": 2.6875836849212646, | |
| "learning_rate": 5.722760714820057e-05, | |
| "loss": 2.0263, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.47455168245013096, | |
| "grad_norm": 3.1060383319854736, | |
| "learning_rate": 5.5583514556208514e-05, | |
| "loss": 1.997, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.47455168245013096, | |
| "eval_loss": 1.997556209564209, | |
| "eval_runtime": 116.9074, | |
| "eval_samples_per_second": 2.104, | |
| "eval_steps_per_second": 2.104, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4848680233729599, | |
| "grad_norm": 2.5845789909362793, | |
| "learning_rate": 5.393328421452514e-05, | |
| "loss": 1.9811, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.4951843642957888, | |
| "grad_norm": 2.890994071960449, | |
| "learning_rate": 5.2278730159931076e-05, | |
| "loss": 1.9808, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.4951843642957888, | |
| "eval_loss": 2.000748872756958, | |
| "eval_runtime": 83.3047, | |
| "eval_samples_per_second": 2.953, | |
| "eval_steps_per_second": 2.953, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.5055007052186178, | |
| "grad_norm": 3.2054591178894043, | |
| "learning_rate": 5.062167118210367e-05, | |
| "loss": 2.0514, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5158170461414467, | |
| "grad_norm": 2.911627769470215, | |
| "learning_rate": 4.896392882428901e-05, | |
| "loss": 2.0169, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5158170461414467, | |
| "eval_loss": 1.9872047901153564, | |
| "eval_runtime": 78.3344, | |
| "eval_samples_per_second": 3.14, | |
| "eval_steps_per_second": 3.14, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5261333870642756, | |
| "grad_norm": 3.094301223754883, | |
| "learning_rate": 4.730732538094749e-05, | |
| "loss": 1.9815, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5364497279871046, | |
| "grad_norm": 2.6194097995758057, | |
| "learning_rate": 4.565368189457313e-05, | |
| "loss": 1.9605, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5364497279871046, | |
| "eval_loss": 1.9974828958511353, | |
| "eval_runtime": 116.5117, | |
| "eval_samples_per_second": 2.111, | |
| "eval_steps_per_second": 2.111, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5467660689099335, | |
| "grad_norm": 3.0435950756073, | |
| "learning_rate": 4.400481615388948e-05, | |
| "loss": 1.9643, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5570824098327625, | |
| "grad_norm": 2.7957849502563477, | |
| "learning_rate": 4.236254069562213e-05, | |
| "loss": 2.0195, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5570824098327625, | |
| "eval_loss": 1.9963051080703735, | |
| "eval_runtime": 83.9136, | |
| "eval_samples_per_second": 2.932, | |
| "eval_steps_per_second": 2.932, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5673987507555914, | |
| "grad_norm": 2.5869903564453125, | |
| "learning_rate": 4.0728660812044536e-05, | |
| "loss": 2.0336, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5777150916784203, | |
| "grad_norm": 2.6142055988311768, | |
| "learning_rate": 3.910497256648742e-05, | |
| "loss": 1.9619, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5777150916784203, | |
| "eval_loss": 1.9877939224243164, | |
| "eval_runtime": 113.1041, | |
| "eval_samples_per_second": 2.175, | |
| "eval_steps_per_second": 2.175, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5880314326012492, | |
| "grad_norm": 2.57497501373291, | |
| "learning_rate": 3.749326081899329e-05, | |
| "loss": 1.9923, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5983477735240782, | |
| "grad_norm": 2.97292160987854, | |
| "learning_rate": 3.589529726428615e-05, | |
| "loss": 1.9361, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5983477735240782, | |
| "eval_loss": 2.004483699798584, | |
| "eval_runtime": 85.1956, | |
| "eval_samples_per_second": 2.887, | |
| "eval_steps_per_second": 2.887, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6086641144469072, | |
| "grad_norm": 2.6015868186950684, | |
| "learning_rate": 3.431283848421347e-05, | |
| "loss": 2.033, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.618980455369736, | |
| "grad_norm": 2.7763192653656006, | |
| "learning_rate": 3.274762401680124e-05, | |
| "loss": 1.9932, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.618980455369736, | |
| "eval_loss": 1.9814573526382446, | |
| "eval_runtime": 114.536, | |
| "eval_samples_per_second": 2.148, | |
| "eval_steps_per_second": 2.148, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.629296796292565, | |
| "grad_norm": 2.6632089614868164, | |
| "learning_rate": 3.120137444404442e-05, | |
| "loss": 1.9622, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6396131372153939, | |
| "grad_norm": 2.701653480529785, | |
| "learning_rate": 2.9675789500535328e-05, | |
| "loss": 1.9519, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6396131372153939, | |
| "eval_loss": 1.9895679950714111, | |
| "eval_runtime": 79.8573, | |
| "eval_samples_per_second": 3.08, | |
| "eval_steps_per_second": 3.08, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6499294781382229, | |
| "grad_norm": 2.8519248962402344, | |
| "learning_rate": 2.8172546205008683e-05, | |
| "loss": 2.0187, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.6602458190610517, | |
| "grad_norm": 2.4840056896209717, | |
| "learning_rate": 2.6693297016857188e-05, | |
| "loss": 1.9843, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6602458190610517, | |
| "eval_loss": 1.9900814294815063, | |
| "eval_runtime": 89.9348, | |
| "eval_samples_per_second": 2.735, | |
| "eval_steps_per_second": 2.735, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6705621599838807, | |
| "grad_norm": 2.6297292709350586, | |
| "learning_rate": 2.523966801964468e-05, | |
| "loss": 1.985, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6808785009067096, | |
| "grad_norm": 2.6009626388549805, | |
| "learning_rate": 2.3813257133612827e-05, | |
| "loss": 1.963, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6808785009067096, | |
| "eval_loss": 1.9819916486740112, | |
| "eval_runtime": 117.6742, | |
| "eval_samples_per_second": 2.091, | |
| "eval_steps_per_second": 2.091, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6911948418295386, | |
| "grad_norm": 2.4555246829986572, | |
| "learning_rate": 2.2415632359146856e-05, | |
| "loss": 1.9888, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.7015111827523676, | |
| "grad_norm": 2.845592975616455, | |
| "learning_rate": 2.104833005313131e-05, | |
| "loss": 1.9376, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7015111827523676, | |
| "eval_loss": 1.979328989982605, | |
| "eval_runtime": 83.6262, | |
| "eval_samples_per_second": 2.942, | |
| "eval_steps_per_second": 2.942, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7118275236751964, | |
| "grad_norm": 2.663501262664795, | |
| "learning_rate": 1.971285324008994e-05, | |
| "loss": 2.0383, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7221438645980254, | |
| "grad_norm": 2.5632455348968506, | |
| "learning_rate": 1.84106699599668e-05, | |
| "loss": 1.9876, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7221438645980254, | |
| "eval_loss": 1.9884700775146484, | |
| "eval_runtime": 126.7545, | |
| "eval_samples_per_second": 1.941, | |
| "eval_steps_per_second": 1.941, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7324602055208543, | |
| "grad_norm": 2.6025450229644775, | |
| "learning_rate": 1.7143211654364762e-05, | |
| "loss": 2.0093, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7427765464436833, | |
| "grad_norm": 3.391676187515259, | |
| "learning_rate": 1.5911871593014837e-05, | |
| "loss": 2.0157, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7427765464436833, | |
| "eval_loss": 1.9833872318267822, | |
| "eval_runtime": 84.5923, | |
| "eval_samples_per_second": 2.908, | |
| "eval_steps_per_second": 2.908, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7530928873665121, | |
| "grad_norm": 2.5039122104644775, | |
| "learning_rate": 1.4718003342206722e-05, | |
| "loss": 1.9728, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.7634092282893411, | |
| "grad_norm": 3.065073013305664, | |
| "learning_rate": 1.3562919276863844e-05, | |
| "loss": 2.011, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7634092282893411, | |
| "eval_loss": 1.9842520952224731, | |
| "eval_runtime": 81.8194, | |
| "eval_samples_per_second": 3.007, | |
| "eval_steps_per_second": 3.007, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7737255692121701, | |
| "grad_norm": 2.7710351943969727, | |
| "learning_rate": 1.2447889137898293e-05, | |
| "loss": 2.0308, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.784041910134999, | |
| "grad_norm": 3.067017078399658, | |
| "learning_rate": 1.1374138636432053e-05, | |
| "loss": 2.0179, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.784041910134999, | |
| "eval_loss": 1.9779284000396729, | |
| "eval_runtime": 128.2963, | |
| "eval_samples_per_second": 1.917, | |
| "eval_steps_per_second": 1.917, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.794358251057828, | |
| "grad_norm": 2.3697543144226074, | |
| "learning_rate": 1.0342848106418368e-05, | |
| "loss": 1.9998, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.8046745919806568, | |
| "grad_norm": 2.8660995960235596, | |
| "learning_rate": 9.35515120714447e-06, | |
| "loss": 1.9693, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8046745919806568, | |
| "eval_loss": 1.978676676750183, | |
| "eval_runtime": 85.106, | |
| "eval_samples_per_second": 2.891, | |
| "eval_steps_per_second": 2.891, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8149909329034858, | |
| "grad_norm": 2.3432037830352783, | |
| "learning_rate": 8.41213367704224e-06, | |
| "loss": 2.0149, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.8253072738263147, | |
| "grad_norm": 2.753044605255127, | |
| "learning_rate": 7.51483214017637e-06, | |
| "loss": 1.9632, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8253072738263147, | |
| "eval_loss": 1.982351541519165, | |
| "eval_runtime": 114.3934, | |
| "eval_samples_per_second": 2.15, | |
| "eval_steps_per_second": 2.15, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8356236147491437, | |
| "grad_norm": 2.3161306381225586, | |
| "learning_rate": 6.664232966721995e-06, | |
| "loss": 1.9832, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.8459399556719726, | |
| "grad_norm": 2.218827486038208, | |
| "learning_rate": 5.8612711886848196e-06, | |
| "loss": 1.9367, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8459399556719726, | |
| "eval_loss": 1.9775891304016113, | |
| "eval_runtime": 90.4335, | |
| "eval_samples_per_second": 2.72, | |
| "eval_steps_per_second": 2.72, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8562562965948015, | |
| "grad_norm": 2.4218974113464355, | |
| "learning_rate": 5.106829472055202e-06, | |
| "loss": 1.9481, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8665726375176305, | |
| "grad_norm": 2.6613683700561523, | |
| "learning_rate": 4.401737146526219e-06, | |
| "loss": 1.9824, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8665726375176305, | |
| "eval_loss": 1.9729845523834229, | |
| "eval_runtime": 119.6431, | |
| "eval_samples_per_second": 2.056, | |
| "eval_steps_per_second": 2.056, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8768889784404594, | |
| "grad_norm": 2.633181095123291, | |
| "learning_rate": 3.7467692938425057e-06, | |
| "loss": 1.9396, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.8872053193632883, | |
| "grad_norm": 2.728327989578247, | |
| "learning_rate": 3.142645895781715e-06, | |
| "loss": 1.9911, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8872053193632883, | |
| "eval_loss": 1.9718983173370361, | |
| "eval_runtime": 82.9362, | |
| "eval_samples_per_second": 2.966, | |
| "eval_steps_per_second": 2.966, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8975216602861172, | |
| "grad_norm": 2.4041829109191895, | |
| "learning_rate": 2.5900310427053044e-06, | |
| "loss": 1.9742, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9078380012089462, | |
| "grad_norm": 2.4489846229553223, | |
| "learning_rate": 2.089532203548794e-06, | |
| "loss": 2.0075, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9078380012089462, | |
| "eval_loss": 1.9730160236358643, | |
| "eval_runtime": 91.0108, | |
| "eval_samples_per_second": 2.703, | |
| "eval_steps_per_second": 2.703, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9181543421317752, | |
| "grad_norm": 2.5976321697235107, | |
| "learning_rate": 1.6416995580537664e-06, | |
| "loss": 1.9874, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.9284706830546041, | |
| "grad_norm": 3.244720458984375, | |
| "learning_rate": 1.247025391975698e-06, | |
| "loss": 1.9809, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9284706830546041, | |
| "eval_loss": 1.9729856252670288, | |
| "eval_runtime": 118.5331, | |
| "eval_samples_per_second": 2.075, | |
| "eval_steps_per_second": 2.075, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.938787023977433, | |
| "grad_norm": 2.2416179180145264, | |
| "learning_rate": 9.059435559326257e-07, | |
| "loss": 2.0036, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9491033649002619, | |
| "grad_norm": 2.4049975872039795, | |
| "learning_rate": 6.188289884893062e-07, | |
| "loss": 1.9971, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9491033649002619, | |
| "eval_loss": 1.9721975326538086, | |
| "eval_runtime": 83.8998, | |
| "eval_samples_per_second": 2.932, | |
| "eval_steps_per_second": 2.932, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9594197058230909, | |
| "grad_norm": 2.46026611328125, | |
| "learning_rate": 3.8599730400115107e-07, | |
| "loss": 1.9341, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.9697360467459198, | |
| "grad_norm": 2.334559917449951, | |
| "learning_rate": 2.0770444567118075e-07, | |
| "loss": 1.9913, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9697360467459198, | |
| "eval_loss": 1.9720317125320435, | |
| "eval_runtime": 127.7811, | |
| "eval_samples_per_second": 1.925, | |
| "eval_steps_per_second": 1.925, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9800523876687487, | |
| "grad_norm": 2.3480026721954346, | |
| "learning_rate": 8.414640420116305e-08, | |
| "loss": 1.9948, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9903687285915777, | |
| "grad_norm": 2.4215712547302246, | |
| "learning_rate": 1.5459002346324135e-08, | |
| "loss": 1.916, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9903687285915777, | |
| "eval_loss": 1.972062587738037, | |
| "eval_runtime": 84.2065, | |
| "eval_samples_per_second": 2.921, | |
| "eval_steps_per_second": 2.921, | |
| "step": 384 | |
| } | |
| ], | |
| "logging_steps": 4, | |
| "max_steps": 387, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.5374998322774016e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |